New Cisco Catalyst QoS book [7:75229]
Hi, I was wondering whether it's worth reading the new Cisco Catalyst QoS IOS book whilst revising for the CCIE theory or even perhaps the practical exams. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=75229&t=75229 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: QoS Exam 642-641 [7:74081]
Charlie, Thanks. Actually, with knowledgenet, one may not need to buy practice exams a knowledgenet course, if it targets a given certification exam, is sufficient. How I wish that I could have unlimited means to buy these online courses:-) Anyway, Thanks a lot. Mwalie Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=74149&t=74081 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: QoS Exam 642-641 [7:74081]
Charlie, Congrats!! Good to let us know; I should do it some day. (Which materials did you use for this one?) Again, Congrats!! Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=74140&t=74081 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: QoS Exam 642-641 [7:74081]
I used the knowledgenet QoS training course and Boson #1 QoS practice test to study for the test. (I probably could have gotten away with just using the knowledgenet QoS training course though.) Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=74142&t=74081 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: QoS Exam 642-641 [7:74081]
Yea! I passed. It was pretty easy though. (No tricks or hazy questions in this test.) I guess I'm still bitter after having to take the Safe Exam 2x to pass. Now onto the CCNP recert which I hear is quite fun. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=74132&t=74081 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
QoS Exam 642-641 [7:74081]
Taking this bad boy tomorrow... and advice? All of the new exams seem to be quite a bit more painful than the old ones. Or at least more difficult in my opinion... Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=74081&t=74081 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
BGP and QOS Beta exams [7:73599]
Folks I have seen a few mentions of the BGP and QOS beta exams recently (also mentioning the results). My question is, am I the only person still waiting for results for these exams? My Vue exam history shows Tue February 18, 2003 02:30 PM 641-661: BGP Corefacts, Cambridge, GBR taken Thu March 27, 2003 02:00 PM 643-641: Quality of Service Corefacts, Cambridge, GBR taken and certmanager doesnt mention either. Is it time to start chasing vue/cisco? As an aside, I took the CCNP support beta last year and never actually received results at all, although it did show up in certmanager as a pass about 3 months after I had given up waiting and passed the non beta version. Thanks Peter Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=73599&t=73599 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: BGP and QOS Beta exams [7:73599]
--On 07 August 2003 02:50 + Mwalie W wrote: > > Yes, you will have to begin chasing VUE and Cisco. > Thanks, that is what I thought > > Good Luck! You must be a very patient person:-) And this is also the > reason I do not like Beta exams now. > Actually, I am very impatient. Which is why I try to make a point of only doing Beta's when I dont need the exam, and then trying to ensure I "forget" about the exam. I really had put the exams out of my mind until I saw a couple of groupstudy messages in which people mentioned the results. I dont think there is a problem with my address as I have received results for CCIE Beta qualification exams I took before and after the QOS and BGP exams. Peter Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=73759&t=73599 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
RE: BGP and QOS Beta exams [7:73599]
Peter, Yes, you will have to begin chasing VUE and Cisco. For example, I did BGP Beta on May 30th 2003 and I got a letter about my passing from Prometric around 20th June 2003. After a few days, it also appeared in my Tracking System. The same with BSCI Beta 643-801. It could have something to do with your postal address. So, check with VUE, then they should guide concerning how to approach Cisco ... because my experience shows that Cisco asks for a copy of the result of a given exam, and this comes from Prometric/VUE. Good Luck! You must be a very patient person:-) And this is also the reason I do not like Beta exams now. Mwalie CCDP Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=73637&t=73599 -- **Please support GroupStudy by purchasing from the GroupStudy Store: http://shop.groupstudy.com FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Frottle (WAS: Bandwidth, QoS, and Contention netw [7:73478]
This thread is a couple weeks old, but I felt this is relevant. The discussion dealt with microsegmentation for wireless networks, and lo and behold this popped up on slashdot: http://frottle.sourceforge.net/. Apparently, someone else decided that wireless traffic needed better management. From the site: Frottle (Freenet throttle) is an open source GNU GPL project to control traffic on wireless networks. Such control eliminates the common hidden-node effect even on large scale wireless networks. Frottle is currently only available for Linux wireless gateways using iptables firewalls, with plans to develop a windows client in the future. Frottle works by scheduling the traffic of each client, using a master node to co-ordinate actions. This eliminates collisions, and prevents clients with stronger signals from receiving bandwidth bias. Frottle has been developed and tested on the large community wireless network of WaFreeNet. We have found running frottle has given us a significant improvment in the network usability. Testing results will be documented here as time permits. RAther than find a way to segment the RF spectrum, it looks like they're running standard 802.11b wireless in conjunction with a deterministic token passing method. In addition to granting a host permission to send, the token tells a host how much data it can send. Good times. Enjoy, Hal > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Monday, July 21, 2003 6:06 PM > To: [EMAIL PROTECTED] > Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] > > > Quick addendum: > > As Howard mentioned, the problem with a contention medium > isn't only how > much bandwidth is available but also how quickly a station > can access it. If > the stataion listens before it sends and can't send for long > periods of time > because the medium is not free, this is bad news for voice and video. > > Shared Ethernet has gone the way of LocalTalk, (almost?) but > shared wireless > is gaining popularity, as Chuck has been saying. I haven't > seen any studies > yet that address medium access delay on wireless networks > based on load and > access characteristics, but maybe they will get published at > some point, > (although it's pretty unpredicatable with bursy data sources, > of course). > > Suffice it to say, "microsegmentation" for wireless networks will be > necessary to some extent, just like it was with shared > Ethernet, depending > on delay and delay variance requirements of the applications. > > Priscilla > > "Chuck Whose Road is Ever Shorte wrote: > > > > ""Zsombor Papp"" wrote in message > > news:[EMAIL PROTECTED] > > > "Chuck Whose Road is Ever Shorte wrote: > > > > after viewing the presentation, you tell me - is > > > > this not saying > > > > that 5 megabits is more than adequate for voice, video, etc? > > > > > > I don't think that was the point of the presention. > > Regardless, I can tell > > > you that 5Mbps is enough for voice. :) For video, it depends > > on the > > quality. > > > > 30 frame per second video over ISDN requires 3 ISDN lines ( 6 > > B-channels ) > > for a total of 384K. I've spoken about video over WAN links on > > several > > occasions with the video guy in my group. He tells me he likes > > to reserve > > 500K over T1's typically. > > > > Howard, if your reading, I just looked over my unsent drafts of > > the question > > regarding bandwidth, and the point I failed to raise in this > > posted thread > > is that of global synchronization. One of the major benefits of > > such QoS > > mechanisms as RED and WRED is that the phenomenon of global > > synch can be > > controlled, meaning a more efficient use of bandwidth ( no > > periods of > > congestion followed by periods of low activity because of the > > TCP backoff > > mechanisms ) > > > > Now supposing, even in a contention medium, I could fine tune > > my queueing > > such that I no longer suffered from global synch. I set my > > voice queues and > > my delay sensative queues such that sufficient badwidth was > > available, and I > > used RED or WRED for the general queue. I'm wondering if there > > are studies > > done, papers in the IETF working groups demonstrating that > > given proper > > queueing mechanisms that less bandwidth is required or > > necessary? > > > > Cisco does offer downstream only QoS on their wireless product > > line. Not > > sure I understa
Re: Bandwidth, QoS, and Contention networks [7:72645]
""Logan, Harold"" wrote in message news:[EMAIL PROTECTED] > be given no credibility whatsoever. I'm neither a wireless guru nor an RF > engineer. > > The problem of ethernet and contention-based media has been solved for the > most part by full duplex ethernet and layer 2 QoS. As far as the hosts on > the network are concerned, the media is always available. "Congestion" is > not encountered until data hits a switch, at which point a layer 2 (or layer > 3, depending on how cool the switch is) QoS policy can be enforced, > packets/frames can be prioritized, and life is good. > > I wonder then if it would be possible to microsegment a wireless network, as > Priscilla suggests, using different ranges of the RF spectrum? many things are possible, but with radio, as opposed to wire, you have a different set of problems. I'm not going to claim I have all this down yet either. Wireless experts I know all seem to have several years experience in military radio and satellite communications and operate in a different world than us wireheads. ;-> consider how many radio stations there are in a given area. why? why can't you have a radio station every couple of clicks? because a radio signal uses a certain amount of spectrum to transfer intelligence. 98.7 on your radio dial isn't exactly and only 98.7 megahertz. the signal includes frequencies to either side. if two stations' signals overlap, you have problems. radio spectrum is limited. for good reason. it must be rationed so that more stations can broadcast good signal. 802.11 is nothing more than radio, pure and simple. the unregulated ISM band is subject to the same laws of physics as are broadcast radio bands. you can't have two signals operating at the same frequency in the same area, or the signals interfere with eachother. DSSS permits some flexibility, but not much. There are 11 channels in the US available under 802.11b however, you have to have proper channel separation or the frequency overlap will cause problems, same as if you have two radio stations whose signals are in the same frequency range. >Most wireless > equipment I've worked on operates spread spectrum so that interference on > one channel doesn't cause a total data loss. If anyone on the list has used > the army SINCGARS radios, they (as well as most other military > communications equipment, I'm sure) incorporate frequency-hopping as well as > scrambling to make jamming and interception more difficult. Rather than hop > frequencies, would it be possible to have AP's that listen on all available > 802.11 bands, but "segment" the hosts into collision domains by having them > only transmit on certain channels? I would guess that for one thing, the military bands are not as crowded as the ISM bands. secondly, my quick look on google seems to indicate that even among the military bands, certain frequencies are reserved for certain activities. secondly, considering that the best in radio receivers these days still can't sort it out when two radio broadcasts on close frequencies collide, it would be a good guess that this would be problematic. the laws of physics still aplly. as RFC 1925 states, "No matter how hard you push and no matter what the priority, you can't increase the speed of light." Radio is not wire. Wire contains signal, meaning keeps it contained within the medium. it has been know to happen that data cables laid close to eachother have bled into eachother, causing interference on both cables. I believe that twisted pair construction has pretty much eliminated this problem. wireless radio cannot contain much of anything. the signal spreads out. nature of the beast. > > Does that sound doable or am I talking out of my arse? =) I've been wrong about "trekkie tech" many many times, so who knows? on the other hand, be careful what you wish for. I for one would not enjoy living in a world where we all had to be connected every moment of every day. > > Hal > > > -Original Message- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > > Sent: Monday, July 21, 2003 6:06 PM > > To: [EMAIL PROTECTED] > > Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] > > > > > > Quick addendum: > > > > As Howard mentioned, the problem with a contention medium > > isn't only how > > much bandwidth is available but also how quickly a station > > can access it. If > > the stataion listens before it sends and can't send for long > > periods of time > > because the medium is not free, this is bad news for voice and video. > > > > Shared Ethernet has gone the way of LocalTalk, (almost?) but > > shared wireless > > is gaining popul
RE: Bandwidth, QoS, and Contention networks [7:72645]
be given no credibility whatsoever. I'm neither a wireless guru nor an RF engineer. The problem of ethernet and contention-based media has been solved for the most part by full duplex ethernet and layer 2 QoS. As far as the hosts on the network are concerned, the media is always available. "Congestion" is not encountered until data hits a switch, at which point a layer 2 (or layer 3, depending on how cool the switch is) QoS policy can be enforced, packets/frames can be prioritized, and life is good. I wonder then if it would be possible to microsegment a wireless network, as Priscilla suggests, using different ranges of the RF spectrum? Most wireless equipment I've worked on operates spread spectrum so that interference on one channel doesn't cause a total data loss. If anyone on the list has used the army SINCGARS radios, they (as well as most other military communications equipment, I'm sure) incorporate frequency-hopping as well as scrambling to make jamming and interception more difficult. Rather than hop frequencies, would it be possible to have AP's that listen on all available 802.11 bands, but "segment" the hosts into collision domains by having them only transmit on certain channels? Does that sound doable or am I talking out of my arse? =) Hal > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Sent: Monday, July 21, 2003 6:06 PM > To: [EMAIL PROTECTED] > Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] > > > Quick addendum: > > As Howard mentioned, the problem with a contention medium > isn't only how > much bandwidth is available but also how quickly a station > can access it. If > the stataion listens before it sends and can't send for long > periods of time > because the medium is not free, this is bad news for voice and video. > > Shared Ethernet has gone the way of LocalTalk, (almost?) but > shared wireless > is gaining popularity, as Chuck has been saying. I haven't > seen any studies > yet that address medium access delay on wireless networks > based on load and > access characteristics, but maybe they will get published at > some point, > (although it's pretty unpredicatable with bursy data sources, > of course). > > Suffice it to say, "microsegmentation" for wireless networks will be > necessary to some extent, just like it was with shared > Ethernet, depending > on delay and delay variance requirements of the applications. > > Priscilla > > "Chuck Whose Road is Ever Shorte wrote: > > > > ""Zsombor Papp"" wrote in message > > news:[EMAIL PROTECTED] > > > "Chuck Whose Road is Ever Shorte wrote: > > > > after viewing the presentation, you tell me - is > > > > this not saying > > > > that 5 megabits is more than adequate for voice, video, etc? > > > > > > I don't think that was the point of the presention. > > Regardless, I can tell > > > you that 5Mbps is enough for voice. :) For video, it depends > > on the > > quality. > > > > 30 frame per second video over ISDN requires 3 ISDN lines ( 6 > > B-channels ) > > for a total of 384K. I've spoken about video over WAN links on > > several > > occasions with the video guy in my group. He tells me he likes > > to reserve > > 500K over T1's typically. > > > > Howard, if your reading, I just looked over my unsent drafts of > > the question > > regarding bandwidth, and the point I failed to raise in this > > posted thread > > is that of global synchronization. One of the major benefits of > > such QoS > > mechanisms as RED and WRED is that the phenomenon of global > > synch can be > > controlled, meaning a more efficient use of bandwidth ( no > > periods of > > congestion followed by periods of low activity because of the > > TCP backoff > > mechanisms ) > > > > Now supposing, even in a contention medium, I could fine tune > > my queueing > > such that I no longer suffered from global synch. I set my > > voice queues and > > my delay sensative queues such that sufficient badwidth was > > available, and I > > used RED or WRED for the general queue. I'm wondering if there > > are studies > > done, papers in the IETF working groups demonstrating that > > given proper > > queueing mechanisms that less bandwidth is required or > > necessary? > > > > Cisco does offer downstream only QoS on their wireless product > > line. Not > > sure I understand the mechanism completely, but I have to > > believe it is >
Re: Bandwidth, QoS, and Contention networks [7:72645]
Quick addendum: As Howard mentioned, the problem with a contention medium isn't only how much bandwidth is available but also how quickly a station can access it. If the stataion listens before it sends and can't send for long periods of time because the medium is not free, this is bad news for voice and video. Shared Ethernet has gone the way of LocalTalk, (almost?) but shared wireless is gaining popularity, as Chuck has been saying. I haven't seen any studies yet that address medium access delay on wireless networks based on load and access characteristics, but maybe they will get published at some point, (although it's pretty unpredicatable with bursy data sources, of course). Suffice it to say, "microsegmentation" for wireless networks will be necessary to some extent, just like it was with shared Ethernet, depending on delay and delay variance requirements of the applications. Priscilla "Chuck Whose Road is Ever Shorte wrote: > > ""Zsombor Papp"" wrote in message > news:[EMAIL PROTECTED] > > "Chuck Whose Road is Ever Shorte wrote: > > > after viewing the presentation, you tell me - is > > > this not saying > > > that 5 megabits is more than adequate for voice, video, etc? > > > > I don't think that was the point of the presention. > Regardless, I can tell > > you that 5Mbps is enough for voice. :) For video, it depends > on the > quality. > > 30 frame per second video over ISDN requires 3 ISDN lines ( 6 > B-channels ) > for a total of 384K. I've spoken about video over WAN links on > several > occasions with the video guy in my group. He tells me he likes > to reserve > 500K over T1's typically. > > Howard, if your reading, I just looked over my unsent drafts of > the question > regarding bandwidth, and the point I failed to raise in this > posted thread > is that of global synchronization. One of the major benefits of > such QoS > mechanisms as RED and WRED is that the phenomenon of global > synch can be > controlled, meaning a more efficient use of bandwidth ( no > periods of > congestion followed by periods of low activity because of the > TCP backoff > mechanisms ) > > Now supposing, even in a contention medium, I could fine tune > my queueing > such that I no longer suffered from global synch. I set my > voice queues and > my delay sensative queues such that sufficient badwidth was > available, and I > used RED or WRED for the general queue. I'm wondering if there > are studies > done, papers in the IETF working groups demonstrating that > given proper > queueing mechanisms that less bandwidth is required or > necessary? > > Cisco does offer downstream only QoS on their wireless product > line. Not > sure I understand the mechanism completely, but I have to > believe it is > based on enough solid study such that given a reasonable > design, there would > be less concern for voice in the mix. Recall that Cisco will > soon be > releasing their own wireless IP phone, and the whole point of > it is to > provide untethered mobility throughout an enterprise. > > > > > > > > I'm just asking - if people are more productive, > > > despite the obvious lack of bandwidth, and > > > despite the step back to a contention medium, > > > is there something to be said about the > > > perceived need for 100 megabits to the desktop? > > > > I agree with Fred, such a perception is probably misguided in > most cases. > > Most people are very happy even with their 1.5Mbps DSL line. > But all this > > depends on what you want to do. Full-screen DVD quality video > won't work > > over DSL. Even online gaming could use more than 1.5Mbps. > FWIW, I've heard > > that in Korea, there is a serious market for dedicated > 100Mbps connections > > to the *home* due to wide-spread online gaming (I don't know > if this is > > true, I find it a bit hard to believe). > > > can't comment much regarding the needs of on-line gaming, but > the guy > teaching the wireless class two weeks ago said he was an avvid > on line gamer > and that his DSL was plenty fine for what he did. He also said > he had a > couple friends around his neighborhood who did on-line gaming > via a sireless > AP that he set up for their use. Don't know the particular > game, so I can > say as to whether it is the same one you play. > > > > > Also consider that pure 10Mbps Ethernet interfaces are > getting pretty > rare; > > most of the Ethernet interfaces are 10/100. So in a campus > network, in > most > > cases
Re: Bandwidth, QoS, and Contention networks [7:72645]
I hear you Chuck. I think microsegmentation (that's what we used to call it, as you know I'm sure) was definitely oversold by the marketing types. Microsegmentation to a point certainly makes sense. I troubleshooted quite a few shared Ethernet networks that were approaching meltdown because of too many users sharing the bandwidth. But for most applications, 100 full duplex to the desktop isn't actually necessary. Links that carry multiple users' data need more bandwidth, but a link that just carries a single user's data for a user doing ordinary typical applications, will use a minimal amount of that 100 Mbps. I challenge my students to try to use just 10 Mbps. They are shocked at how hard it is. We tend to lose sight of the fact that switches do have some disadvantages too. They are more complex and more likely to have problems than hubs. They make protocol analysis difficult. But we can't go backwards. It's getting harder and harder to even buy a hub! But, as you say 11 Mbps shared RF, i.e. 802.11 is a step backwards, and it works! I'm using it right now! Is it making me more productive? Absolutely not. I'm trying to type on this awful little keyboard, looking at an eensy-weensy screen, doing Group Study when I should be eating lunch and decompressing and chatting (face-to-face) with colleauges. :-) Last year I attended a security conference. I think it was the first time they had wireless available. Much less work got done. All the guys sat hunkered over their notebook computers, working and trying to attack each other. The informal discussions that lead to brilliant ideas and collaboration were much harder to start. It was up to the women to start them. :-) Unforutately, we made up about 5% of the attendees. OK, now it really is lunch time. I hope some colleagues will be in the cafeteria and we can talk in person. Priscilla "Chuck Whose Road is Ever Shorte wrote: > > ""Zsombor Papp"" wrote in message > news:[EMAIL PROTECTED] > > I think comparing shared wireless to dedicated wired > connections is a bit > of > > an apple vs orange contest. You can get shared wireless where > you can't > get > > anything else (e.g. walking from one meeting room to the > other, or > attending > > a meeting with 10 other people in a room where there are only > 4 wired > > ports), so obviously it is better than all the other choices. > If you can > > choose between a 100Mbps switch port and a 11Mbps shared > wireless link > > without sacrificing anything (e.g. in case of servers or > desktop > machines), > > then the 100Mbps switch port is obviously better. > > well sure. one of the other reasons I got to pondering the > original question > ( are we overselling the value of bandwidth? ) is the following: > > http://newsroom.cisco.com/dlls/video_audio_archive/?video > > check out the long reach ethernet presentation, maybe 3/4 down > the page. > > LRE can provide up to 5 megabits full duplex over cat 3 phone > wire - data > and voice. after viewing the presentation, you tell me - is > this not saying > that 5 megabits is more than adequate for voice, video, etc? > > as for wireless - I fully understand that the requirements of > the > application drive the need for bandwidth. I'm just asking - if > people are > more productive, despite the obvious lack of bandwidth, and > despite the step > back to a contention medium, is there something to be said > about the > perceived need for 100 megabits to the desktop? > > > > > > > > > Thanks, > > > > Zsombor > > > > "Chuck Whose Road is Ever Shorte wrote: > > > > > > ""Howard C. Berkowitz"" wrote in message > > > news:[EMAIL PROTECTED] > > > > > > > What's the medium cost between the two cities? Can you > use > > > demand > > > > circuits as a backup? Can you live with one more PVC and > > > trust the > > > > physical connection? Is QoS-unpredictable cable or DSL > > > available? > > > > > > > > > > Funny you should ask this, Howard. I've been struggling for > > > several weeks > > > how to pose the question. Have we, the engineering / > technical > > > sales > > > community oversold the idea of dedicated bandwidth and QoS? > > > > > > Take, for example, wireless. > > > > > > Wireless is essentially a step backwards. For years we have > > > been convincing > > > customers to get rid of their hubs and move into a switched > > > domain, with > > > dedicated bandwidth for every user. This is
RE: Bandwidth, QoS, and Contention networks [7:72645]
We have customers using Vocera's wireless VoIP phones (they look like Star Trek communicators) with no problems. We did have to use wireless VLANs, but I was not involved in the configuration of the AP's. They are having no problems that I'm aware of. Fred Reimer - CCNA Eclipsys Corporation, 200 Ashford Center North, Atlanta, GA 30338 Phone: 404-847-5177 Cell: 770-490-3071 Pager: 888-260-2050 NOTICE; This email contains confidential or proprietary information which may be legally privileged. It is intended only for the named recipient(s). If an addressing or transmission error has misdirected the email, please notify the author by replying to this message. If you are not the named recipient, you are not authorized to use, disclose, distribute, copy, print or rely on this email, and should immediately delete it from your computer. -Original Message- From: "Chuck Whose Road is Ever Shorte [mailto:[EMAIL PROTECTED] Sent: Monday, July 21, 2003 11:08 AM To: [EMAIL PROTECTED] Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] ""Zsombor Papp"" wrote in message news:[EMAIL PROTECTED] > "Chuck Whose Road is Ever Shorte wrote: > > after viewing the presentation, you tell me - is > > this not saying > > that 5 megabits is more than adequate for voice, video, etc? > > I don't think that was the point of the presention. Regardless, I can tell > you that 5Mbps is enough for voice. :) For video, it depends on the quality. 30 frame per second video over ISDN requires 3 ISDN lines ( 6 B-channels ) for a total of 384K. I've spoken about video over WAN links on several occasions with the video guy in my group. He tells me he likes to reserve 500K over T1's typically. Howard, if your reading, I just looked over my unsent drafts of the question regarding bandwidth, and the point I failed to raise in this posted thread is that of global synchronization. One of the major benefits of such QoS mechanisms as RED and WRED is that the phenomenon of global synch can be controlled, meaning a more efficient use of bandwidth ( no periods of congestion followed by periods of low activity because of the TCP backoff mechanisms ) Now supposing, even in a contention medium, I could fine tune my queueing such that I no longer suffered from global synch. I set my voice queues and my delay sensative queues such that sufficient badwidth was available, and I used RED or WRED for the general queue. I'm wondering if there are studies done, papers in the IETF working groups demonstrating that given proper queueing mechanisms that less bandwidth is required or necessary? Cisco does offer downstream only QoS on their wireless product line. Not sure I understand the mechanism completely, but I have to believe it is based on enough solid study such that given a reasonable design, there would be less concern for voice in the mix. Recall that Cisco will soon be releasing their own wireless IP phone, and the whole point of it is to provide untethered mobility throughout an enterprise. > > > I'm just asking - if people are more productive, > > despite the obvious lack of bandwidth, and > > despite the step back to a contention medium, > > is there something to be said about the > > perceived need for 100 megabits to the desktop? > > I agree with Fred, such a perception is probably misguided in most cases. > Most people are very happy even with their 1.5Mbps DSL line. But all this > depends on what you want to do. Full-screen DVD quality video won't work > over DSL. Even online gaming could use more than 1.5Mbps. FWIW, I've heard > that in Korea, there is a serious market for dedicated 100Mbps connections > to the *home* due to wide-spread online gaming (I don't know if this is > true, I find it a bit hard to believe). can't comment much regarding the needs of on-line gaming, but the guy teaching the wireless class two weeks ago said he was an avvid on line gamer and that his DSL was plenty fine for what he did. He also said he had a couple friends around his neighborhood who did on-line gaming via a sireless AP that he set up for their use. Don't know the particular game, so I can say as to whether it is the same one you play. > > Also consider that pure 10Mbps Ethernet interfaces are getting pretty rare; > most of the Ethernet interfaces are 10/100. So in a campus network, in most > cases, there is no real reason to not have 100Mbps to the desktop. indeed. and with three teenage boys around the house, I am happy to sell lots of these things in order to keep the refridgerator full. ;-> yes, Mr Customer, you never can tell when your users will need this bandwidth, what with internet radio, lots of databases out there on the web, and all the e-mail attachments people need to read to get their work done. ;-> >
Re: Bandwidth, QoS, and Contention networks [7:72645]
""Zsombor Papp"" wrote in message news:[EMAIL PROTECTED] > "Chuck Whose Road is Ever Shorte wrote: > > after viewing the presentation, you tell me - is > > this not saying > > that 5 megabits is more than adequate for voice, video, etc? > > I don't think that was the point of the presention. Regardless, I can tell > you that 5Mbps is enough for voice. :) For video, it depends on the quality. 30 frame per second video over ISDN requires 3 ISDN lines ( 6 B-channels ) for a total of 384K. I've spoken about video over WAN links on several occasions with the video guy in my group. He tells me he likes to reserve 500K over T1's typically. Howard, if your reading, I just looked over my unsent drafts of the question regarding bandwidth, and the point I failed to raise in this posted thread is that of global synchronization. One of the major benefits of such QoS mechanisms as RED and WRED is that the phenomenon of global synch can be controlled, meaning a more efficient use of bandwidth ( no periods of congestion followed by periods of low activity because of the TCP backoff mechanisms ) Now supposing, even in a contention medium, I could fine tune my queueing such that I no longer suffered from global synch. I set my voice queues and my delay sensative queues such that sufficient badwidth was available, and I used RED or WRED for the general queue. I'm wondering if there are studies done, papers in the IETF working groups demonstrating that given proper queueing mechanisms that less bandwidth is required or necessary? Cisco does offer downstream only QoS on their wireless product line. Not sure I understand the mechanism completely, but I have to believe it is based on enough solid study such that given a reasonable design, there would be less concern for voice in the mix. Recall that Cisco will soon be releasing their own wireless IP phone, and the whole point of it is to provide untethered mobility throughout an enterprise. > > > I'm just asking - if people are more productive, > > despite the obvious lack of bandwidth, and > > despite the step back to a contention medium, > > is there something to be said about the > > perceived need for 100 megabits to the desktop? > > I agree with Fred, such a perception is probably misguided in most cases. > Most people are very happy even with their 1.5Mbps DSL line. But all this > depends on what you want to do. Full-screen DVD quality video won't work > over DSL. Even online gaming could use more than 1.5Mbps. FWIW, I've heard > that in Korea, there is a serious market for dedicated 100Mbps connections > to the *home* due to wide-spread online gaming (I don't know if this is > true, I find it a bit hard to believe). can't comment much regarding the needs of on-line gaming, but the guy teaching the wireless class two weeks ago said he was an avvid on line gamer and that his DSL was plenty fine for what he did. He also said he had a couple friends around his neighborhood who did on-line gaming via a sireless AP that he set up for their use. Don't know the particular game, so I can say as to whether it is the same one you play. > > Also consider that pure 10Mbps Ethernet interfaces are getting pretty rare; > most of the Ethernet interfaces are 10/100. So in a campus network, in most > cases, there is no real reason to not have 100Mbps to the desktop. indeed. and with three teenage boys around the house, I am happy to sell lots of these things in order to keep the refridgerator full. ;-> yes, Mr Customer, you never can tell when your users will need this bandwidth, what with internet radio, lots of databases out there on the web, and all the e-mail attachments people need to read to get their work done. ;-> > > Thanks, > > Zsombor Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72678&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Bandwidth, QoS, and Contention networks [7:72645]
""Reimer, Fred"" wrote in message news:[EMAIL PROTECTED] > Chuck - > " well sure. one of the other reasons I got to pondering the original > question > ( are we overselling the value of bandwidth? ) is the following: > > http://newsroom.cisco.com/dlls/video_audio_archive/?video > > check out the long reach ethernet presentation, maybe 3/4 down the page. > > LRE can provide up to 5 megabits full duplex over cat 3 phone wire - data > and voice. after viewing the presentation, you tell me - is this not saying > that 5 megabits is more than adequate for voice, video, etc? > > as for wireless - I fully understand that the requirements of the > application drive the need for bandwidth. I'm just asking - if people are > more productive, despite the obvious lack of bandwidth, and despite the step > back to a contention medium, is there something to be said about the > perceived need for 100 megabits to the desktop?" > > Me - > > I think you have to take into account the application requirements. I thought I said that. [ looks at my earlier statement] why yes, I did say that :-> >Would you want to stick an enterprise server on wireless? Heck no. Oh i don't know. The answer is "it depends" :-> but suppose I stick the server on my 3550-24PWR and run 23 access points off that switch, instead of putting in a 4507 with a port ( and wiring ) for each user? >Would you want to stick an Oracle database on wireless? I'd say not. do I have a thin client? am I downloading large reports? uploading large volumes of data? >Take a look at what wireless is actually being used for, and I think you will find that, in > most cases, the shared media is not a huge issue. It's not like we are > transferring hundreds of megabytes of data over wireless on a regular basis, > if at all. my point exactly - > > Is 100MBps really needed to the desktop, let alone 1Gb? I'd say 99.44% of > the time the answer is no. 10Mbps switched to the desktop should be more > than enough for most users. There will always be users and applications > that require greater bandwidth, but as I'm sure most everyone knows the > Gigabit downlinks to the core in most corporations are not being loaded to a > significant degree. Yes, there will be exceptions, but the general rule > I've seen is that the "average" Gb utilization is well below 10% on a > downlink. > > Still, I would not even think of proposing or installing a network today > that did not have 100Mbps capabilities at the edge, and Gb connections to > the core. neither would I. The kids need shoes and I have house payments to make ;-> > > Fred Reimer - CCNA > > > Eclipsys Corporation, 200 Ashford Center North, Atlanta, GA 30338 > Phone: 404-847-5177 Cell: 770-490-3071 Pager: 888-260-2050 > > > NOTICE; This email contains confidential or proprietary information which > may be legally privileged. It is intended only for the named recipient(s). > If an addressing or transmission error has misdirected the email, please > notify the author by replying to this message. If you are not the named > recipient, you are not authorized to use, disclose, distribute, copy, print > or rely on this email, and should immediately delete it from your computer. > > > -Original Message- > From: "Chuck Whose Road is Ever Shorter" [mailto:[EMAIL PROTECTED] > Sent: Sunday, July 20, 2003 2:50 PM > To: [EMAIL PROTECTED] > Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] > > ""Zsombor Papp"" wrote in message > news:[EMAIL PROTECTED] > > I think comparing shared wireless to dedicated wired connections is a bit > of > > an apple vs orange contest. You can get shared wireless where you can't > get > > anything else (e.g. walking from one meeting room to the other, or > attending > > a meeting with 10 other people in a room where there are only 4 wired > > ports), so obviously it is better than all the other choices. If you can > > choose between a 100Mbps switch port and a 11Mbps shared wireless link > > without sacrificing anything (e.g. in case of servers or desktop > machines), > > then the 100Mbps switch port is obviously better. > > well sure. one of the other reasons I got to pondering the original question > ( are we overselling the value of bandwidth? ) is the following: > > http://newsroom.cisco.com/dlls/video_audio_archive/?video > > check out the long reach ethernet presentation, maybe 3/4 down the page. > > LRE can provide up to 5 megabits full duplex over cat 3 phone wire - data > and voice. after viewing the presentation, you tell me - is this not
Re: Bandwidth, QoS, and Contention networks [7:72645]
"Chuck Whose Road is Ever Shorte wrote: > after viewing the presentation, you tell me - is > this not saying > that 5 megabits is more than adequate for voice, video, etc? I don't think that was the point of the presention. Regardless, I can tell you that 5Mbps is enough for voice. :) For video, it depends on the quality. > I'm just asking - if people are more productive, > despite the obvious lack of bandwidth, and > despite the step back to a contention medium, > is there something to be said about the > perceived need for 100 megabits to the desktop? I agree with Fred, such a perception is probably misguided in most cases. Most people are very happy even with their 1.5Mbps DSL line. But all this depends on what you want to do. Full-screen DVD quality video won't work over DSL. Even online gaming could use more than 1.5Mbps. FWIW, I've heard that in Korea, there is a serious market for dedicated 100Mbps connections to the *home* due to wide-spread online gaming (I don't know if this is true, I find it a bit hard to believe). Also consider that pure 10Mbps Ethernet interfaces are getting pretty rare; most of the Ethernet interfaces are 10/100. So in a campus network, in most cases, there is no real reason to not have 100Mbps to the desktop. Thanks, Zsombor Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72673&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Bandwidth, QoS, and Contention networks [7:72645]
Chuck - " well sure. one of the other reasons I got to pondering the original question ( are we overselling the value of bandwidth? ) is the following: http://newsroom.cisco.com/dlls/video_audio_archive/?video check out the long reach ethernet presentation, maybe 3/4 down the page. LRE can provide up to 5 megabits full duplex over cat 3 phone wire - data and voice. after viewing the presentation, you tell me - is this not saying that 5 megabits is more than adequate for voice, video, etc? as for wireless - I fully understand that the requirements of the application drive the need for bandwidth. I'm just asking - if people are more productive, despite the obvious lack of bandwidth, and despite the step back to a contention medium, is there something to be said about the perceived need for 100 megabits to the desktop?" Me - I think you have to take into account the application requirements. Would you want to stick an enterprise server on wireless? Heck no. Would you want to stick an Oracle database on wireless? I'd say not. Take a look at what wireless is actually being used for, and I think you will find that, in most cases, the shared media is not a huge issue. It's not like we are transferring hundreds of megabytes of data over wireless on a regular basis, if at all. Is 100MBps really needed to the desktop, let alone 1Gb? I'd say 99.44% of the time the answer is no. 10Mbps switched to the desktop should be more than enough for most users. There will always be users and applications that require greater bandwidth, but as I'm sure most everyone knows the Gigabit downlinks to the core in most corporations are not being loaded to a significant degree. Yes, there will be exceptions, but the general rule I've seen is that the "average" Gb utilization is well below 10% on a downlink. Still, I would not even think of proposing or installing a network today that did not have 100Mbps capabilities at the edge, and Gb connections to the core. Fred Reimer - CCNA Eclipsys Corporation, 200 Ashford Center North, Atlanta, GA 30338 Phone: 404-847-5177 Cell: 770-490-3071 Pager: 888-260-2050 NOTICE; This email contains confidential or proprietary information which may be legally privileged. It is intended only for the named recipient(s). If an addressing or transmission error has misdirected the email, please notify the author by replying to this message. If you are not the named recipient, you are not authorized to use, disclose, distribute, copy, print or rely on this email, and should immediately delete it from your computer. -Original Message- From: "Chuck Whose Road is Ever Shorter" [mailto:[EMAIL PROTECTED] Sent: Sunday, July 20, 2003 2:50 PM To: [EMAIL PROTECTED] Subject: Re: Bandwidth, QoS, and Contention networks [7:72645] ""Zsombor Papp"" wrote in message news:[EMAIL PROTECTED] > I think comparing shared wireless to dedicated wired connections is a bit of > an apple vs orange contest. You can get shared wireless where you can't get > anything else (e.g. walking from one meeting room to the other, or attending > a meeting with 10 other people in a room where there are only 4 wired > ports), so obviously it is better than all the other choices. If you can > choose between a 100Mbps switch port and a 11Mbps shared wireless link > without sacrificing anything (e.g. in case of servers or desktop machines), > then the 100Mbps switch port is obviously better. well sure. one of the other reasons I got to pondering the original question ( are we overselling the value of bandwidth? ) is the following: http://newsroom.cisco.com/dlls/video_audio_archive/?video check out the long reach ethernet presentation, maybe 3/4 down the page. LRE can provide up to 5 megabits full duplex over cat 3 phone wire - data and voice. after viewing the presentation, you tell me - is this not saying that 5 megabits is more than adequate for voice, video, etc? as for wireless - I fully understand that the requirements of the application drive the need for bandwidth. I'm just asking - if people are more productive, despite the obvious lack of bandwidth, and despite the step back to a contention medium, is there something to be said about the perceived need for 100 megabits to the desktop? > > Thanks, > > Zsombor > > "Chuck Whose Road is Ever Shorte wrote: > > > > ""Howard C. Berkowitz"" wrote in message > > news:[EMAIL PROTECTED] > > > > > What's the medium cost between the two cities? Can you use > > demand > > > circuits as a backup? Can you live with one more PVC and > > trust the > > > physical connection? Is QoS-unpredictable cable or DSL > > available? > > > > > > > Funny you should ask this, Howard. I've been struggling for > &
RE: Bandwidth, QoS, and Contention networks [7:72645]
At 5:29 PM + 7/20/03, Zsombor Papp wrote: >I think comparing shared wireless to dedicated wired connections is a bit of >an apple vs orange contest. AppleTalk, I thought, is off the exam... >You can get shared wireless where you can't get >anything else (e.g. walking from one meeting room to the other, or attending >a meeting with 10 other people in a room where there are only 4 wired >ports), so obviously it is better than all the other choices. If you can >choose between a 100Mbps switch port and a 11Mbps shared wireless link >without sacrificing anything (e.g. in case of servers or desktop machines), >then the 100Mbps switch port is obviously better. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72671&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Bandwidth, QoS, and Contention networks [7:72645]
""Zsombor Papp"" wrote in message news:[EMAIL PROTECTED] > I think comparing shared wireless to dedicated wired connections is a bit of > an apple vs orange contest. You can get shared wireless where you can't get > anything else (e.g. walking from one meeting room to the other, or attending > a meeting with 10 other people in a room where there are only 4 wired > ports), so obviously it is better than all the other choices. If you can > choose between a 100Mbps switch port and a 11Mbps shared wireless link > without sacrificing anything (e.g. in case of servers or desktop machines), > then the 100Mbps switch port is obviously better. well sure. one of the other reasons I got to pondering the original question ( are we overselling the value of bandwidth? ) is the following: http://newsroom.cisco.com/dlls/video_audio_archive/?video check out the long reach ethernet presentation, maybe 3/4 down the page. LRE can provide up to 5 megabits full duplex over cat 3 phone wire - data and voice. after viewing the presentation, you tell me - is this not saying that 5 megabits is more than adequate for voice, video, etc? as for wireless - I fully understand that the requirements of the application drive the need for bandwidth. I'm just asking - if people are more productive, despite the obvious lack of bandwidth, and despite the step back to a contention medium, is there something to be said about the perceived need for 100 megabits to the desktop? > > Thanks, > > Zsombor > > "Chuck Whose Road is Ever Shorte wrote: > > > > ""Howard C. Berkowitz"" wrote in message > > news:[EMAIL PROTECTED] > > > > > What's the medium cost between the two cities? Can you use > > demand > > > circuits as a backup? Can you live with one more PVC and > > trust the > > > physical connection? Is QoS-unpredictable cable or DSL > > available? > > > > > > > Funny you should ask this, Howard. I've been struggling for > > several weeks > > how to pose the question. Have we, the engineering / technical > > sales > > community oversold the idea of dedicated bandwidth and QoS? > > > > Take, for example, wireless. > > > > Wireless is essentially a step backwards. For years we have > > been convincing > > customers to get rid of their hubs and move into a switched > > domain, with > > dedicated bandwidth for every user. This is often done in the > > name of > > productivity. Fewer interruptions of data streams, meaning work > > completed > > faster.Now all the wireless vendors ( Cisco included ) are > > producing studies > > showing how wireless is increasing productivity to the tune of > > an hour a > > day. On a shared contention medium. Cisco will shortly release > > their > > wireless telephone as part of their AVVID suite of products, > > competing with > > the SpectraLink product that has been available for a couple of > > years. > > > > All this gives one reason to re-evaluate what we have been told > > for the last > > couple of years. a contention medium provides the means for > > greater > > productivity? > > > > You mention QoS in your response above. QoS is something being > > pushed as > > necessary for voice, video, and other delay sensitive traffic. > > Cisco > > wireless AP's offer one way quasi QoS. Wireless, however, > > remains a > > contention medium, and will remain so until the FCC changes the > > rules. I'm > > not sure they will be able to release sufficient radio spectrum > > to permit > > all the bandwidth and services that wired can. But wireless is > > so damn > > convenient! > > > > I'm not suggesting that dedicated bandwidth to the desktop is a > > bad thing or > > that there is not need for QoS. However, I'm wondering how all > > of us might > > reconcile two seemingly opposed points of view regarding > > bandwidth and QoS - > > recognizing that wireless, whatever it's limitations, is here > > to stay, and > > will become and remain essential to any and all networks, > > enterprise or > > small business, going forward. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72667&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Bandwidth, QoS, and Contention networks [7:72645]
""Howard C. Berkowitz"" wrote in message news:[EMAIL PROTECTED] > >Howard makes good points. Let me offer another view, perhaps more > >mainline-business" oriented, vs. the very special requirements in > >medicine (which I do appreciate -- I'm alive, thank you, > >radiologist and surgeon). > > > >The increase in productivity due to wireless is believed to > >come--and I can't be more specific than "believed" since I don't > >know the quality of the supposed studies--from capturing value in > >otherwise wasted time, and/or from making qualitative > >improvements in the work environment, leading to more output from > >the extant inputs. Whether it is a good idea in the long run to > >capture the work that could be done when not at the desk (via > >802.11 systems) has not been assessed; we are gathering much > >empirical evidence, though ;-). Personally, the time away from > >the desk is most useful to me--I decompress, and I think. > > Perhaps you have some perspective from your Air Force days on whether > having constant communications available is, in fact, a good thing. > My impression is that air traffic controllers, tactical controllers, > etc., have enforced rest breaks, unless emergency conditions require > otherwise. > > Wireless -- or perhaps more correctly, ubiquitous communiation (a > term from Xerox PARC) -- may increase total production, but at what > error rate? Does it result in quicker burnout? I like Annlee's direction - just because I as a wireless user am more productive, is there a hidden hit to the productivity of others? I.e. more distractive noise in cubeland, higher support costs, both in terms of infrastructure and help desk support ( more calls about connectivity problems as I roam about, being "more productive"? ) There remains the bandwidth question. How much is enough. Check out the Long Reach Ethernet presentation at http://newsroom.cisco.com/dlls/video_audio_archive/?video about 3/4 down the page. seems like Cisco is saying that 5 megabits is more than adequate for AVVID applications. which means that a reasonably well thought out wireless infrastructure offers all that most organizations require for such apps. > > Working at home, I used to keep a laptop in the bedroom, but I only > do that now if I'm sick enough that I can't get downstairs to my > office. I do keep a notepad, colored pens, etc., in the bedroom. > Indeed, sometimes when I have creative/writers' block, it's very > helpful for me to switch modes -- write rather than type. I will even > do what I'll call Zen Design -- relax on the bed, thinking about a > design problem, and even drift off to sleep -- and I'll very > frequently wake up with a key insight. > > > > >Improving the work environment qualitatively may be as simple as > >giving people the chance to work while getting up and moving > >around--not being chained to their plow, er, desk. If this makes > >people more comfortable overall, the theory goes, all their work > >will improve overall. Theory really is a wonderful thing, and > >qualitative improvements do matter. Whether this is among the > >ones which actually help is moot--I recall being grievously > >annoyed in my cubicle days by people whose conversations made it > >hard for me to work--getting their voicemail over a speakerphone > >was a pet peeve. Someone walking around my work area chatting on > >a portable phone could make me go postal if I need to concentrate > >and can't. > > > >But, as Howard said, this is one of many potential tools in the > >kit. In a sense, it may make our job harder, since we will need > >to be able to recognize the appropriate tool to solve the > >problem, which puts us squarely in the problem-recognition > >business. And we may need to persuade customers they don't need > >the gee-whizziest tool; the same amount of money could provide x, > >y, and z, or they could simply spend less money. Probably not an > >argument that will go over well in your present job, I know. But > >I have a customer who keeps coming back--because I keep finding > >him the most economical solution for the problems I've identified > >which he is choosing to solve this month vs. deferring. Every bit > >of business not deferred further comes back to me. > > > >But that's a long-term view again, and many businesses don't feel > >they can take a long view; the stock market may punish them too > >severely. > > > >Annlee > > > >Chuck Whose Road is Ever Shorter wrote: > >> ""Howard C. Berkowitz""
RE: Bandwidth, QoS, and Contention networks [7:72645]
I think comparing shared wireless to dedicated wired connections is a bit of an apple vs orange contest. You can get shared wireless where you can't get anything else (e.g. walking from one meeting room to the other, or attending a meeting with 10 other people in a room where there are only 4 wired ports), so obviously it is better than all the other choices. If you can choose between a 100Mbps switch port and a 11Mbps shared wireless link without sacrificing anything (e.g. in case of servers or desktop machines), then the 100Mbps switch port is obviously better. Thanks, Zsombor "Chuck Whose Road is Ever Shorte wrote: > > ""Howard C. Berkowitz"" wrote in message > news:[EMAIL PROTECTED] > > > What's the medium cost between the two cities? Can you use > demand > > circuits as a backup? Can you live with one more PVC and > trust the > > physical connection? Is QoS-unpredictable cable or DSL > available? > > > > Funny you should ask this, Howard. I've been struggling for > several weeks > how to pose the question. Have we, the engineering / technical > sales > community oversold the idea of dedicated bandwidth and QoS? > > Take, for example, wireless. > > Wireless is essentially a step backwards. For years we have > been convincing > customers to get rid of their hubs and move into a switched > domain, with > dedicated bandwidth for every user. This is often done in the > name of > productivity. Fewer interruptions of data streams, meaning work > completed > faster.Now all the wireless vendors ( Cisco included ) are > producing studies > showing how wireless is increasing productivity to the tune of > an hour a > day. On a shared contention medium. Cisco will shortly release > their > wireless telephone as part of their AVVID suite of products, > competing with > the SpectraLink product that has been available for a couple of > years. > > All this gives one reason to re-evaluate what we have been told > for the last > couple of years. a contention medium provides the means for > greater > productivity? > > You mention QoS in your response above. QoS is something being > pushed as > necessary for voice, video, and other delay sensitive traffic. > Cisco > wireless AP's offer one way quasi QoS. Wireless, however, > remains a > contention medium, and will remain so until the FCC changes the > rules. I'm > not sure they will be able to release sufficient radio spectrum > to permit > all the bandwidth and services that wired can. But wireless is > so damn > convenient! > > I'm not suggesting that dedicated bandwidth to the desktop is a > bad thing or > that there is not need for QoS. However, I'm wondering how all > of us might > reconcile two seemingly opposed points of view regarding > bandwidth and QoS - > recognizing that wireless, whatever it's limitations, is here > to stay, and > will become and remain essential to any and all networks, > enterprise or > small business, going forward. > > Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72665&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Bandwidth, QoS, and Contention networks [7:72645]
>Howard makes good points. Let me offer another view, perhaps more >mainline-business" oriented, vs. the very special requirements in >medicine (which I do appreciate -- I'm alive, thank you, >radiologist and surgeon). > >The increase in productivity due to wireless is believed to >come--and I can't be more specific than "believed" since I don't >know the quality of the supposed studies--from capturing value in >otherwise wasted time, and/or from making qualitative >improvements in the work environment, leading to more output from >the extant inputs. Whether it is a good idea in the long run to >capture the work that could be done when not at the desk (via >802.11 systems) has not been assessed; we are gathering much >empirical evidence, though ;-). Personally, the time away from >the desk is most useful to me--I decompress, and I think. Perhaps you have some perspective from your Air Force days on whether having constant communications available is, in fact, a good thing. My impression is that air traffic controllers, tactical controllers, etc., have enforced rest breaks, unless emergency conditions require otherwise. Wireless -- or perhaps more correctly, ubiquitous communiation (a term from Xerox PARC) -- may increase total production, but at what error rate? Does it result in quicker burnout? Working at home, I used to keep a laptop in the bedroom, but I only do that now if I'm sick enough that I can't get downstairs to my office. I do keep a notepad, colored pens, etc., in the bedroom. Indeed, sometimes when I have creative/writers' block, it's very helpful for me to switch modes -- write rather than type. I will even do what I'll call Zen Design -- relax on the bed, thinking about a design problem, and even drift off to sleep -- and I'll very frequently wake up with a key insight. > >Improving the work environment qualitatively may be as simple as >giving people the chance to work while getting up and moving >around--not being chained to their plow, er, desk. If this makes >people more comfortable overall, the theory goes, all their work >will improve overall. Theory really is a wonderful thing, and >qualitative improvements do matter. Whether this is among the >ones which actually help is moot--I recall being grievously >annoyed in my cubicle days by people whose conversations made it >hard for me to work--getting their voicemail over a speakerphone >was a pet peeve. Someone walking around my work area chatting on >a portable phone could make me go postal if I need to concentrate >and can't. > >But, as Howard said, this is one of many potential tools in the >kit. In a sense, it may make our job harder, since we will need >to be able to recognize the appropriate tool to solve the >problem, which puts us squarely in the problem-recognition >business. And we may need to persuade customers they don't need >the gee-whizziest tool; the same amount of money could provide x, >y, and z, or they could simply spend less money. Probably not an >argument that will go over well in your present job, I know. But >I have a customer who keeps coming back--because I keep finding >him the most economical solution for the problems I've identified >which he is choosing to solve this month vs. deferring. Every bit >of business not deferred further comes back to me. > >But that's a long-term view again, and many businesses don't feel >they can take a long view; the stock market may punish them too >severely. > >Annlee > >Chuck Whose Road is Ever Shorter wrote: >> ""Howard C. Berkowitz"" wrote in message >> news:[EMAIL PROTECTED] >> >> >>>What's the medium cost between the two cities? Can you use demand >>>circuits as a backup? Can you live with one more PVC and trust the >>>physical connection? Is QoS-unpredictable cable or DSL available? >>> >> >> >> Funny you should ask this, Howard. I've been struggling for several weeks >> how to pose the question. Have we, the engineering / technical sales > > community oversold the idea of dedicated bandwidth and QoS? > > >> Take, for example, wireless. >> >> Wireless is essentially a step backwards. For years we have been convincing >> customers to get rid of their hubs and move into a switched domain, with >> dedicated bandwidth for every user. This is often done in the name of >> productivity. Fewer interruptions of data streams, meaning work completed >> faster.Now all the wireless vendors ( Cisco included ) are producing >studies >> showing how wireless is increasing productivity to the tune of an hour a >> day. On a shared contention m
Re: Bandwidth, QoS, and Contention networks [7:72645]
Howard makes good points. Let me offer another view, perhaps more mainline-business" oriented, vs. the very special requirements in medicine (which I do appreciate -- I'm alive, thank you, radiologist and surgeon). The increase in productivity due to wireless is believed to come--and I can't be more specific than "believed" since I don't know the quality of the supposed studies--from capturing value in otherwise wasted time, and/or from making qualitative improvements in the work environment, leading to more output from the extant inputs. Whether it is a good idea in the long run to capture the work that could be done when not at the desk (via 802.11 systems) has not been assessed; we are gathering much empirical evidence, though ;-). Personally, the time away from the desk is most useful to me--I decompress, and I think. Improving the work environment qualitatively may be as simple as giving people the chance to work while getting up and moving around--not being chained to their plow, er, desk. If this makes people more comfortable overall, the theory goes, all their work will improve overall. Theory really is a wonderful thing, and qualitative improvements do matter. Whether this is among the ones which actually help is moot--I recall being grievously annoyed in my cubicle days by people whose conversations made it hard for me to work--getting their voicemail over a speakerphone was a pet peeve. Someone walking around my work area chatting on a portable phone could make me go postal if I need to concentrate and can't. But, as Howard said, this is one of many potential tools in the kit. In a sense, it may make our job harder, since we will need to be able to recognize the appropriate tool to solve the problem, which puts us squarely in the problem-recognition business. And we may need to persuade customers they don't need the gee-whizziest tool; the same amount of money could provide x, y, and z, or they could simply spend less money. Probably not an argument that will go over well in your present job, I know. But I have a customer who keeps coming back--because I keep finding him the most economical solution for the problems I've identified which he is choosing to solve this month vs. deferring. Every bit of business not deferred further comes back to me. But that's a long-term view again, and many businesses don't feel they can take a long view; the stock market may punish them too severely. Annlee Chuck Whose Road is Ever Shorter wrote: > ""Howard C. Berkowitz"" wrote in message > news:[EMAIL PROTECTED] > > >>What's the medium cost between the two cities? Can you use demand >>circuits as a backup? Can you live with one more PVC and trust the >>physical connection? Is QoS-unpredictable cable or DSL available? >> > > > Funny you should ask this, Howard. I've been struggling for several weeks > how to pose the question. Have we, the engineering / technical sales > community oversold the idea of dedicated bandwidth and QoS? > > Take, for example, wireless. > > Wireless is essentially a step backwards. For years we have been convincing > customers to get rid of their hubs and move into a switched domain, with > dedicated bandwidth for every user. This is often done in the name of > productivity. Fewer interruptions of data streams, meaning work completed > faster.Now all the wireless vendors ( Cisco included ) are producing studies > showing how wireless is increasing productivity to the tune of an hour a > day. On a shared contention medium. Cisco will shortly release their > wireless telephone as part of their AVVID suite of products, competing with > the SpectraLink product that has been available for a couple of years. > > All this gives one reason to re-evaluate what we have been told for the last > couple of years. a contention medium provides the means for greater > productivity? > > You mention QoS in your response above. QoS is something being pushed as > necessary for voice, video, and other delay sensitive traffic. Cisco > wireless AP's offer one way quasi QoS. Wireless, however, remains a > contention medium, and will remain so until the FCC changes the rules. I'm > not sure they will be able to release sufficient radio spectrum to permit > all the bandwidth and services that wired can. But wireless is so damn > convenient! > > I'm not suggesting that dedicated bandwidth to the desktop is a bad thing or > that there is not need for QoS. However, I'm wondering how all of us might > reconcile two seemingly opposed points of view regarding bandwidth and QoS - > recognizing that wireless, whatever it's limitations, is here to stay, and > will become and remain essential to any and all networks, ent
Re: Bandwidth, QoS, and Contention networks [7:72645]
At 3:25 AM + 7/20/03, " Chuck Whose Road is Ever Shorter " wrote: >""Howard C. Berkowitz"" wrote in message >news:[EMAIL PROTECTED] > >> What's the medium cost between the two cities? Can you use demand >> circuits as a backup? Can you live with one more PVC and trust the >> physical connection? Is QoS-unpredictable cable or DSL available? >> > >Funny you should ask this, Howard. I've been struggling for several weeks >how to pose the question. Have we, the engineering / technical sales >community oversold the idea of dedicated bandwidth and QoS? Perhaps not sufficiently refined the ideas. There is a common assumption that "dedicated bandwidth" means "large amounts of dedicated bandwidth." But for IP telephony, the issue is more having quick and reliable access to a small amount of bandwidth. Remember 802.4 Token Bus, which people (especially General Motors) was essential for process control? It was eventually superceded by carefully tuned 802.3, kept lightly loaded and with small MTUs to encourage interleaving. I deal with medical imaging applications where workstation access at 100 Mbps or more is a valid requirement. Still, the real requirement is often dedicated bandwidth from the desktop to a workgroup disk cache, not network-wide to the cache when it may be several days before the radiologist reads the image. There are applications where the end-to-end bandwidth makes sense. I know of a US radiologist that has made a business of providing night interpretation services for smaller Australian hospitals, and has to get the image in minutes. When one considers the people cost here, transoceanic bandwidth begins to look attractive. Things like telepresence surgery, of course, MUST have dedicated bandwidth. Yet I'm also working with battlefield medical systems where I have to design for the communications links being down, or only providing 2.4 Kbps on the WAN. Clearly, this doesn't fly for telepresence, but still has value for logistics. > >You mention QoS in your response above. QoS is something being pushed as >necessary for voice, video, and other delay sensitive traffic. Cisco >wireless AP's offer one way quasi QoS. Wireless, however, remains a >contention medium, and will remain so until the FCC changes the rules. I'm >not sure they will be able to release sufficient radio spectrum to permit >all the bandwidth and services that wired can. But wireless is so damn >convenient! One of the arguments for QoS features is that it doesn't need a highly skilled capacity planner/performance analyst. True, but there may be alternatives, such as more use of expert systems for capacity planning and less real-time QoS > >I'm not suggesting that dedicated bandwidth to the desktop is a bad thing or >that there is not need for QoS. However, I'm wondering how all of us might >reconcile two seemingly opposed points of view regarding bandwidth and QoS - >recognizing that wireless, whatever it's limitations, is here to stay, and >will become and remain essential to any and all networks, enterprise or >small business, going forward. They are all tools in a toolbox. I am now dealing with applications where wireless simply is not acceptable, either for battlefield or hospital use. Remember that simply being able to locate a military transmitter is sufficient to target it--cryptography doesn't help. In hospitals, you frequently have shielded rooms that block wireless propagation. While the hospital-wide bans on cellular telephones and the like may be overreaction, there are very real reasons to be extremely careful of any free-space electromagnetic energy in areas where electrodes penetrate the skin and provide low-resistance paths to the heart. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72655&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Bandwidth, QoS, and Contention networks [7:72645]
""Howard C. Berkowitz"" wrote in message news:[EMAIL PROTECTED] > What's the medium cost between the two cities? Can you use demand > circuits as a backup? Can you live with one more PVC and trust the > physical connection? Is QoS-unpredictable cable or DSL available? > Funny you should ask this, Howard. I've been struggling for several weeks how to pose the question. Have we, the engineering / technical sales community oversold the idea of dedicated bandwidth and QoS? Take, for example, wireless. Wireless is essentially a step backwards. For years we have been convincing customers to get rid of their hubs and move into a switched domain, with dedicated bandwidth for every user. This is often done in the name of productivity. Fewer interruptions of data streams, meaning work completed faster.Now all the wireless vendors ( Cisco included ) are producing studies showing how wireless is increasing productivity to the tune of an hour a day. On a shared contention medium. Cisco will shortly release their wireless telephone as part of their AVVID suite of products, competing with the SpectraLink product that has been available for a couple of years. All this gives one reason to re-evaluate what we have been told for the last couple of years. a contention medium provides the means for greater productivity? You mention QoS in your response above. QoS is something being pushed as necessary for voice, video, and other delay sensitive traffic. Cisco wireless AP's offer one way quasi QoS. Wireless, however, remains a contention medium, and will remain so until the FCC changes the rules. I'm not sure they will be able to release sufficient radio spectrum to permit all the bandwidth and services that wired can. But wireless is so damn convenient! I'm not suggesting that dedicated bandwidth to the desktop is a bad thing or that there is not need for QoS. However, I'm wondering how all of us might reconcile two seemingly opposed points of view regarding bandwidth and QoS - recognizing that wireless, whatever it's limitations, is here to stay, and will become and remain essential to any and all networks, enterprise or small business, going forward. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72645&t=72645 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QOS manager [7:71786]
When i configure qos in frame relay network, i must set proprieties in core router or in border router ??? I want to grant max bandwith for one specific port. The connections for this port start from both sides. Tanks. Frederico Madeira Coordenador de Suporte N. Landim Comircio Ltda PABX: 81. 3497.3029 e-mail: [EMAIL PROTECTED] Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=71786&t=71786 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS - Enabling Traffic Shaping [7:70880]
Brian, A few minutes after I sent my post I saw that coming. I was taking his statement "The burst size should 1/8 of the bit rate cfg'd" as a scenario requirement. The "1/8" should have clued me into the fact the the problem was not stating a specific scenario requirement but just stating the Bc=CIR*Tc rule. Thanks, Jonathan > -Original Message- > From: Brian McGahan [mailto:[EMAIL PROTECTED] > Sent: Wednesday, June 18, 2003 2:53 PM > To: 'Jonathan V Hays'; 'John Smith'; [EMAIL PROTECTED]; > [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED]; > [EMAIL PROTECTED]; > [EMAIL PROTECTED]; 'CyscoExpert' > Subject: RE: QoS - Enabling Traffic Shaping > > > Jonathan, > > First let's make sure we're on the same page with definitions. > For traffic-shaping (both GTS and FRTS), the terms are as follows: > > AR = Access Rate (physical clock or rate-limit at local loop) > CIR = Average per second > Bc = Bits per interval committed > Be = Bits per interval excess > Tc = Interval in ms > > The following formula should be used to calculate Bc: > > Bc = CIR * Tc/1000 > > The following formula should be used to calculate Be: > > Be = (AR - CIR) * Tc/1000 > > > In this example, our access-rate is 64000bps. We are trying to > average 15% of our access rate for telnet traffic. Average per second > is CIR. > > CIR = 64000 bps X .15 = 9600 bps > > Let's assume for this example that our Tc is 1/8 second, or 125 > ms. Our committed burst size would then be as follows: > > Bc = CIR * Tc/1000 > Bc = 9600 * 125/1000 > Bc = 9600 * .125 > Bc = 1200 > > Now let's double check our calculation. > > We have 125 ms per interval. There are 1000 ms per second. > > 1 interval | 1000 ms > - = 8 intervals per second > 125 ms | 1 second > > If we are sending 1200 bits per interval 8 times per second: > > 1200 bits | 8 intervals > = 1200 * 8 = 9600bps > 1 interval | 1 second > > > As a side note, if you want no excess burst, you want the Be > value to be > zero, not the same as the Bc value. The syntax should then read as > follows: > > traffic-shape group 101 9600 1200 0 > > > > HTH > > Brian McGahan, CCIE #8593 > Director of Design and Implementation > [EMAIL PROTECTED] > > CyscoExpert Corporation > Internetwork Consulting & Training > Toll Free: 866.CyscoXP > Fax: 847.674.2625 > > > > -Original Message- > > From: Jonathan V Hays [mailto:[EMAIL PROTECTED] > > Sent: Wednesday, June 18, 2003 11:39 AM > > To: 'John Smith'; [EMAIL PROTECTED]; [EMAIL PROTECTED] > > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]; > > [EMAIL PROTECTED]; [EMAIL PROTECTED] > > Subject: RE: QoS - Enabling Traffic Shaping > > > > John, > > > > I assume the '25' after each percentage sign in your original post > below > > is a typo (e.g., "15%25 of bandwidth"). Given that assumption, you > want > > to assign traffic to three categories, with 15%, 50%, and 35% of the > 64 > > kbps bandwidth. > > > > 64000 x .15 = 9600 bps > > 64000 x .50 = 32000 bps > > 64000 x .35 = 22400 bps > > > > > burst size should 1/8 of the bit rate cfg'd, no excess burst > > > > 1/8 = .125 > > > > 9600 x .125 = 1200 bps > > 32000 x .125 = 4000 bps > > 22400 x .125 = 2800 bps > > > > However, I question the burst size given in your solution. The > > definition from the Doc CD states that the burst-size is the > "Sustained > > number of bits that can be sent per INTERVAL. On Frame Relay > interfaces, > > this is the Committed Burst size contracted with your service > provider. > > " > > > > See > > > http://www.cisco.com/univercd/cc/td/doc/product/software/ios12 > 2/122cgcr/ > > fqos_r/qrfcmd11.htm#1019905 > > > > If the interval is 1/8 second (and it may not be) I would think the > > correct answer would divide each of 1200, 4000, and 2800 by > 8, to get > > bits per interval. > > > > HTH, > > > > Jonathan > > > > > -Original Message- > > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > > > Behalf Of John Smith > > > Sent: Wednesday, June 18, 2003 10:21 AM > > > To: [EMAIL PROTECTED]; [EMAIL PROTECTED] > > > Cc: [EMAIL PROTECTED]; > > > [EMAIL PROTECTED]; [EMAIL PROTECTED]; > > > [EMAIL PROTECTED] > >
RE: QoS - Enabling Traffic Shaping [7:70869]
John, I assume the '25' after each percentage sign in your original post below is a typo (e.g., "15%25 of bandwidth"). Given that assumption, you want to assign traffic to three categories, with 15%, 50%, and 35% of the 64 kbps bandwidth. 64000 x .15 = 9600 bps 64000 x .50 = 32000 bps 64000 x .35 = 22400 bps > burst size should 1/8 of the bit rate cfg'd, no excess burst 1/8 = .125 9600 x .125 = 1200 bps 32000 x .125 = 4000 bps 22400 x .125 = 2800 bps However, I question the burst size given in your solution. The definition from the Doc CD states that the burst-size is the "Sustained number of bits that can be sent per INTERVAL. On Frame Relay interfaces, this is the Committed Burst size contracted with your service provider. " See http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/ fqos_r/qrfcmd11.htm#1019905 If the interval is 1/8 second (and it may not be) I would think the correct answer would divide each of 1200, 4000, and 2800 by 8, to get bits per interval. HTH, Jonathan > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > Behalf Of John Smith > Sent: Wednesday, June 18, 2003 10:21 AM > To: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED]; > [EMAIL PROTECTED]; [EMAIL PROTECTED]; > [EMAIL PROTECTED] > Subject: QoS - Enabling Traffic Shaping > > > Somehow I have forgotten how to do math... Can anyone explain > to me or point to a doc on figuring out the percentage for > the below lab. > > Enable traffic shaping on int serial 0/0 as follows: > > Shape Telnet and ICMP traffic to 15%25 of bandwidth > Shape FTP traffic to 50%25 of bw > Shape all remianing traffic to 35%25 of bw. > > Remember the total link bw is 64 kbps on this interface. The > burst size should 1/8 of the bit rate cfg'd, no excess burst > should be allowed. > > The answer from the lab help are, below can anyone tell me > how the number 9600, 32000 and 22400 were arrived at? I can > figure out the 1/8th part 1200, 4000 and 2800, but somehow > can't remember how to get the original percentage.. > > traffic-shape group 101 9600 1200 1200 > traffic-shape group 102 32000 4000 4000 > traffic-shape group 103 22400 2800 2800 > > Thanks. > > > > - > Do you Yahoo!? > SBC Yahoo! DSL - Now only $29.95 per month! > > > __ > _ > You are subscribed to the GroupStudy.com CCIE R&S Discussion Group. > > Subscription information may be found at: > http://www.groupstudy.com/list/CCIELab.html Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70869&t=70869 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Traffic generator for Testing QoS features [7:70864]
If your company has the money, check out SmartBits. Even Cisco use it to test their products. I have no association with either SmartBits or Cisco but have used the product to test the Cos/QoS features of Cisco routers. Best regards, Dom Stocqueler SysDom Technologies Visit our website - www.sysdom.org -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of neil K Sent: 18 June 2003 16:05 To: [EMAIL PROTECTED] Subject: Traffic generator for Testing QoS features [7:70864] I want to test QoS features on a router, and am looking for a good traffic generator which can generate different kinds of traffic. Any recommendations? Thanks, neil Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70878&t=70864 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Traffic generator for Testing QoS features [7:70864]
Try NetIQ chariot. --luan -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Wednesday, June 18, 2003 12:41 PM To: [EMAIL PROTECTED] Subject: RE: Traffic generator for Testing QoS features [7:70864] Check out Nantechthey have a BGP route generator for windows and it includes a traffic generator for testing with different service types. We use it in our lab and it works pretty good for $75. Thanks, Mario Puras SoluNet Technical Support -Original Message- From: neil K [mailto:[EMAIL PROTECTED] Sent: Wednesday, June 18, 2003 11:05 AM To: [EMAIL PROTECTED] Subject: Traffic generator for Testing QoS features [7:70864] I want to test QoS features on a router, and am looking for a good traffic generator which can generate different kinds of traffic. Any recommendations? Thanks, neil Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70876&t=70864 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS - Enabling Traffic Shaping [7:70877]
Jonathan, First let's make sure we're on the same page with definitions. For traffic-shaping (both GTS and FRTS), the terms are as follows: AR = Access Rate (physical clock or rate-limit at local loop) CIR = Average per second Bc = Bits per interval committed Be = Bits per interval excess Tc = Interval in ms The following formula should be used to calculate Bc: Bc = CIR * Tc/1000 The following formula should be used to calculate Be: Be = (AR - CIR) * Tc/1000 In this example, our access-rate is 64000bps. We are trying to average 15% of our access rate for telnet traffic. Average per second is CIR. CIR = 64000 bps X .15 = 9600 bps Let's assume for this example that our Tc is 1/8 second, or 125 ms. Our committed burst size would then be as follows: Bc = CIR * Tc/1000 Bc = 9600 * 125/1000 Bc = 9600 * .125 Bc = 1200 Now let's double check our calculation. We have 125 ms per interval. There are 1000 ms per second. 1 interval | 1000 ms - = 8 intervals per second 125 ms | 1 second If we are sending 1200 bits per interval 8 times per second: 1200 bits | 8 intervals = 1200 * 8 = 9600bps 1 interval | 1 second As a side note, if you want no excess burst, you want the Be value to be zero, not the same as the Bc value. The syntax should then read as follows: traffic-shape group 101 9600 1200 0 HTH Brian McGahan, CCIE #8593 Director of Design and Implementation [EMAIL PROTECTED] CyscoExpert Corporation Internetwork Consulting & Training Toll Free: 866.CyscoXP Fax: 847.674.2625 > -Original Message- > From: Jonathan V Hays [mailto:[EMAIL PROTECTED] > Sent: Wednesday, June 18, 2003 11:39 AM > To: 'John Smith'; [EMAIL PROTECTED]; [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]; > [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: RE: QoS - Enabling Traffic Shaping > > John, > > I assume the '25' after each percentage sign in your original post below > is a typo (e.g., "15%25 of bandwidth"). Given that assumption, you want > to assign traffic to three categories, with 15%, 50%, and 35% of the 64 > kbps bandwidth. > > 64000 x .15 = 9600 bps > 64000 x .50 = 32000 bps > 64000 x .35 = 22400 bps > > > burst size should 1/8 of the bit rate cfg'd, no excess burst > > 1/8 = .125 > > 9600 x .125 = 1200 bps > 32000 x .125 = 4000 bps > 22400 x .125 = 2800 bps > > However, I question the burst size given in your solution. The > definition from the Doc CD states that the burst-size is the "Sustained > number of bits that can be sent per INTERVAL. On Frame Relay interfaces, > this is the Committed Burst size contracted with your service provider. > " > > See > http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/ > fqos_r/qrfcmd11.htm#1019905 > > If the interval is 1/8 second (and it may not be) I would think the > correct answer would divide each of 1200, 4000, and 2800 by 8, to get > bits per interval. > > HTH, > > Jonathan > > > -Original Message- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > > Behalf Of John Smith > > Sent: Wednesday, June 18, 2003 10:21 AM > > To: [EMAIL PROTECTED]; [EMAIL PROTECTED] > > Cc: [EMAIL PROTECTED]; > > [EMAIL PROTECTED]; [EMAIL PROTECTED]; > > [EMAIL PROTECTED] > > Subject: QoS - Enabling Traffic Shaping > > > > > > Somehow I have forgotten how to do math... Can anyone explain > > to me or point to a doc on figuring out the percentage for > > the below lab. > > > > Enable traffic shaping on int serial 0/0 as follows: > > > > Shape Telnet and ICMP traffic to 15%25 of bandwidth > > Shape FTP traffic to 50%25 of bw > > Shape all remianing traffic to 35%25 of bw. > > > > Remember the total link bw is 64 kbps on this interface. The > > burst size should 1/8 of the bit rate cfg'd, no excess burst > > should be allowed. > > > > The answer from the lab help are, below can anyone tell me > > how the number 9600, 32000 and 22400 were arrived at? I can > > figure out the 1/8th part 1200, 4000 and 2800, but somehow > > can't remember how to get the original percentage.. > > > > traffic-shape group 101 9600 1200 1200 > > traffic-shape group 102 32000 4000 4000 > > traffic-shape group 103 22400 2800 2800 > > > > Thanks. > > > > > > > > - > > Do you Yahoo!? > > SBC Yahoo! DSL - Now only $29.95 per month! > > > > > > __ > > _ > > You are subscribed to the GroupStudy.com CCIE R&S Discussion Group. > > > > Subscription information may be found at: > > http://www.groupstudy.com/list/CCIELab.html Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70877&t=70877 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Traffic generator for Testing QoS features [7:70864]
Check out Nantechthey have a BGP route generator for windows and it includes a traffic generator for testing with different service types. We use it in our lab and it works pretty good for $75. Thanks, Mario Puras SoluNet Technical Support -Original Message- From: neil K [mailto:[EMAIL PROTECTED] Sent: Wednesday, June 18, 2003 11:05 AM To: [EMAIL PROTECTED] Subject: Traffic generator for Testing QoS features [7:70864] I want to test QoS features on a router, and am looking for a good traffic generator which can generate different kinds of traffic. Any recommendations? Thanks, neil Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70870&t=70864 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Traffic generator for Testing QoS features [7:70864]
Adtech Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70866&t=70864 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Traffic generator for Testing QoS features [7:70864]
I want to test QoS features on a router, and am looking for a good traffic generator which can generate different kinds of traffic. Any recommendations? Thanks, neil Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70864&t=70864 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS - Enabling Traffic Shaping [7:70861]
Somehow I have forgotten how to do math... Can anyone explain to me or point to a doc on figuring out the percentage for the below lab. Enable traffic shaping on int serial 0/0 as follows: Shape Telnet and ICMP traffic to 15%25 of bandwidth Shape FTP traffic to 50%25 of bw Shape all remianing traffic to 35%25 of bw. Remember the total link bw is 64 kbps on this interface. The burst size should 1/8 of the bit rate cfg'd, no excess burst should be allowed. The answer from the lab help are, below can anyone tell me how the number 9600, 32000 and 22400 were arrived at? I can figure out the 1/8th part 1200, 4000 and 2800, but somehow can't remember how to get the original percentage.. traffic-shape group 101 9600 1200 1200 traffic-shape group 102 32000 4000 4000 traffic-shape group 103 22400 2800 2800 Thanks. - Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70861&t=70861 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: New BCRAN - VPN, QoS and Traffic Shaping [7:70226]
I'm really worried about these three topics - is there anybody out there who has some idea of how deep they cover these, and maybe some good links? Thanks in advance for any help WB On Thu, 2003-06-05 at 16:31, Weaselboy wrote: > I'm trying to get ready to take the new BCRAN when it comes out, but all > the study guides are focused on the old exam. For those who took the > beta, can you give me any guidance on these three topics - VPN, QoS and > Traffic Shaping. I'm not looking for anybody to break the NDA, I just > want to know how deep I need to go, and if there are any good links on > the CCO. Thanks. > > The WB Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70382&t=70226 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
New BCRAN - VPN, QoS and Traffic Shaping [7:70226]
I'm trying to get ready to take the new BCRAN when it comes out, but all the study guides are focused on the old exam. For those who took the beta, can you give me any guidance on these three topics - VPN, QoS and Traffic Shaping. I'm not looking for anybody to break the NDA, I just want to know how deep I need to go, and if there are any good links on the CCO. Thanks. The WB Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=70226&t=70226 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
LLQ Modular QoS on VPN/Tunnel/or POS [7:69792]
All, I'd try this myself if I had a router at my fingertips... 1. I'd like to apply LLQ for my VoIP traffic over a site to site VPN using Cisco's Modular QoS technique. Will the router "bark" if I try to apply my output service policy to the tunnel interface? Or do I need to apply it to the regular serial interface? 2. Also, in a different network, can I apply my output service policy to a POS interface, or will the router bark? I've applied service policies to serial interfaces in the past with no problem, but you never know what will happen when it's a new kind of interface. - Jennifer Mellone Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=69792&t=69792 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS and Multicast question [7:66554]
I need to protect regular data across a WAN link against excessive Multicast streams (video, etc.). We are using "ip multicast rate-limiting", but this is flow specific so we have to create specific ACL's to catch each stream. I tried CBWFQ on a 7206, but I need to do an aggregate limit as well as flow specific limiting. I saw a service-policy command nested under policy-map>class. Would this be an option? I need suggestions. Example: Mcast 1 gets 1mb Mcast 2 gets 700k All Mcast together cannot exceed 3mb. Thanks Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66554&t=66554 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Basic QOS Frame MPLS question [7:66210]
If your traffic (voice or data) is within the contract, the Provider must transport it without drops. That is quality of service. It is why it is recommended to set up Be =0 when transporting voice over a PVC. If you need to use the Frame Relay characteristic of transmitting bursts higher than the contracted, the Frame Relay network will discard your packets when congestion occurs. You can set up the CLP bit for lower priority traffic so the Provider will discard those packets first; but it will discard the other packets if the network is still congested. I can not see a relation between change from Frame Relay to MPLS to provide QoS. The Provider must offer the guaranteed quality contracted, use it Frame Relay, ATM or Dedicated Lines. For dedicated circuits you can just transmit the bandwidth; for Frame Relay you have some flexibility. I do not know if there is a Service Provider that offers MPLS circuits with guaranteed bandwidth just for high priority traffic. Is there? >From what I know, the advantage of implementing MPLS on the Provider4s network is to offer isolated services based on layer 3. Without MPLS, the Provider needs layer 2 technologies to transport isolated networks. On private networks, MPLS helps you choose links and routers without the limitation of routing protocols, based on the best metric. Am I missing something ? "Paul" em 26/03/2003 18:15:30 Para: cc: Assunto: Re: Basic QOS Frame MPLS question [7:66210] Wow .. thanks very much for the links . We just got someone trying to sell us an MPLS solution and I don't know much about it . I got a lot of reading to do :) Thanks again . - Original Message - From: To: Sent: Wednesday, March 26, 2003 2:08 PM Subject: Re: Basic QOS Frame MPLS question [7:66210] > I don4t think so. > > There are many QoS tool that you can use without MPLS. > > For example, you can use "ip rtp priority", so the priority traffic will go > to a high priority queue. Also, the fragmentation options will help you to > avoid 'big' frames from starving the voice frames. > > Low Latency Queueing for Frame Relay > http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121newft/121 t/121t2/dtfrpqfq.htm#wp1033474 > > > Link Fragmentation and Interleaving with Frame-Relay > http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos _c/fqcprt6/qcflfifr.htm > > Frame Relay Header compression > http://www.cisco.com/univercd/cc/td/doc/product/software/ios112/rtphead.htm# xtocid63548 > > > > > > > > "Paul" @groupstudy.com em 25/03/2003 19:59:20 > > Favor responder a "Paul" > > Enviado Por: [EMAIL PROTECTED] > > > Para: [EMAIL PROTECTED] > cc: > > Assunto:Basic QOS Frame MPLS question [7:66210] > > > Hi, Quick question to everyone > > At work I have a Frame Cloud that links all our sites together in a hub and > spoke manner. > > At some of the sites I would like to extend our IP Telephony and perhaps > introduce Video Conferencing. > > Assume I have adequate bandwidth throughout for video and IP telephony. > > I would like to implement QOS. Am I correct in assuming that I can only > prioritise voice/video over the frame circuit, and that if I want to > implement > QOS I would have to 'swap' Frame for MPLS/Layer 4 Switching ??? > > Kind regards > > Paul Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66327&t=66210 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS - match MPLS Exp [7:66311]
I'm testing QoS over a MPLS network on a lab. I reset QoS on the CE (R1) inbound LAN interface, set the DSCP bit outbound to the WAN to af11. On the PE (R5) router inbound I match the DSCP bit af11 and set the MPLS Exp bit to 1, and get match as below. The I have a outbound policy on the PE (R5) FE0/0.1 sub-interface to the MPLS P (R4) node which 'match ip mpls exp 1' and no packets match this policy as swon below(2): The input policy (set-exp-in) does mark the packet, but the output policy (test1) does't see marked mpls packets. Why? r5#sh pol int s4/0 Serial4/0 Service-policy input: set-exp-in Class-map: class-af11 (match-any) 117 packets, 5912 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: ip dscp af11 117 packets, 5912 bytes 5 minute rate 0 bps QoS Set mpls experimental 1 Packets marked 117 Class-map: class-default (match-any) 63 packets, 28040 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: any r5# r5#sh pol int f0/0.1 FastEthernet0/0.1 Service-policy output: test1 Class-map: test1 (match-any) 0 packets, 0 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: mpls experimental 1 0 packets, 0 bytes 5 minute rate 0 bps QoS Set ip precedence 1 Packets marked 0 Class-map: class-default (match-any) 66224 packets, 6229493 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: any r5# The lab with MPLS as follows : R1 R5 R4R2 R7 R8 CE --- PE --- P --- P --- PE --- CE on R1 : class-map match-any class-af11 match access-group name acl-af11 match ip precedence 1 policy-map class-wan-out class class-af11 set ip dscp af11 ip access-list extended acl-af11 permit tcp any any eq domain permit tcp any any eq echo permit tcp any any eq telnet permit tcp any any eq tacacs on R5: class-map match-any test1 match mpls experimental 1 match ip dscp af11 class-map match-any class-af11 match ip dscp af11 policy-map set-exp-in class class-af11 set mpls experimental 1 policy-map test1 class test1 set ip precedence 1 Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66311&t=66311 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Basic QOS Frame MPLS question [7:66210]
Paul wrote: > > > I would like to implement QOS. Am I correct in assuming that I > can only > prioritise voice/video over the frame circuit, and that if I > want to implement > QOS I would have to 'swap' Frame for MPLS/Layer 4 Switching ??? > > Kind regards > > Paul > > In a traditional FR type network, the FR switches cannot prioritize your traffic because it cannot tell the difference between a high priority packet.. So the QoS you would apply only gets applied to your router's WAN interface. There might be 10 FR switches in between your 2 routers, and none of them can prioritize because it cannot distinguish traffic. With MPLS, you can do QoS even within the cloud because for example, if you set your VOIP to be prec 5 and require high priority, your MPLS cloud (routers) can tell the difference and will treat your prec 5 traffic better (if the provider has it congiured this way). That is the key difference. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66271&t=66210 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Basic QOS Frame MPLS question [7:66210]
At 02:08 PM 3/26/2003 +, [EMAIL PROTECTED] wrote: >I don4t think so. > >There are many QoS tool that you can use without MPLS. For what it's worth, MPLS is not a QOS tool. It can be used as a component in a QOS strategy, but by itself, provides no QOS. >For example, you can use "ip rtp priority", so the priority traffic will go >to a high priority queue. Also, the fragmentation options will help you to >avoid 'big' frames from starving the voice frames. > >Low Latency Queueing for Frame Relay >http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121newft/121t/121t2/dtfrpqfq.htm#wp1033474 > > >Link Fragmentation and Interleaving with Frame-Relay >http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos_c/fqcprt6/qcflfifr.htm > >Frame Relay Header compression >http://www.cisco.com/univercd/cc/td/doc/product/software/ios112/rtphead.htm#xtocid63548 > > > > > > > >"Paul" @groupstudy.com em 25/03/2003 19:59:20 > >Favor responder a "Paul" > >Enviado Por: [EMAIL PROTECTED] > > >Para: [EMAIL PROTECTED] >cc: > >Assunto:Basic QOS Frame MPLS question [7:66210] > > >Hi, Quick question to everyone > >At work I have a Frame Cloud that links all our sites together in a hub and >spoke manner. > >At some of the sites I would like to extend our IP Telephony and perhaps >introduce Video Conferencing. > >Assume I have adequate bandwidth throughout for video and IP telephony. > >I would like to implement QOS. Am I correct in assuming that I can only >prioritise voice/video over the frame circuit, and that if I want to >implement >QOS I would have to 'swap' Frame for MPLS/Layer 4 Switching ??? > >Kind regards > >Paul Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66261&t=66210 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Basic QOS Frame MPLS question [7:66210]
I don4t think so. There are many QoS tool that you can use without MPLS. For example, you can use "ip rtp priority", so the priority traffic will go to a high priority queue. Also, the fragmentation options will help you to avoid 'big' frames from starving the voice frames. Low Latency Queueing for Frame Relay http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121newft/121t/121t2/dtfrpqfq.htm#wp1033474 Link Fragmentation and Interleaving with Frame-Relay http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos_c/fqcprt6/qcflfifr.htm Frame Relay Header compression http://www.cisco.com/univercd/cc/td/doc/product/software/ios112/rtphead.htm#xtocid63548 "Paul" @groupstudy.com em 25/03/2003 19:59:20 Favor responder a "Paul" Enviado Por: [EMAIL PROTECTED] Para: [EMAIL PROTECTED] cc: Assunto:Basic QOS Frame MPLS question [7:66210] Hi, Quick question to everyone At work I have a Frame Cloud that links all our sites together in a hub and spoke manner. At some of the sites I would like to extend our IP Telephony and perhaps introduce Video Conferencing. Assume I have adequate bandwidth throughout for video and IP telephony. I would like to implement QOS. Am I correct in assuming that I can only prioritise voice/video over the frame circuit, and that if I want to implement QOS I would have to 'swap' Frame for MPLS/Layer 4 Switching ??? Kind regards Paul Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66252&t=66210 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Basic QOS Frame MPLS question [7:66210]
Hi, Quick question to everyone At work I have a Frame Cloud that links all our sites together in a hub and spoke manner. At some of the sites I would like to extend our IP Telephony and perhaps introduce Video Conferencing. Assume I have adequate bandwidth throughout for video and IP telephony. I would like to implement QOS. Am I correct in assuming that I can only prioritise voice/video over the frame circuit, and that if I want to implement QOS I would have to 'swap' Frame for MPLS/Layer 4 Switching ??? Kind regards Paul Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66210&t=66210 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Configure QoS on Cat3550 [7:66055]
Hi,All Configure WRR queuing strategy on Cat3550 as the following: mls qos interface fa0/1 switch mode access mls qos band 2 3 2 3 Then "show queueing interface fa 0/1",the system always show me "queueing strategy: none" instead of "weighted-round-robin" as expected.But I'm sure the WRR queuing strategy is working.So I wonder if the "show queueing" command is still valid on Cat3550.(on Cat6500,it's ok) By the way,how to show the traffic distribution among the queues?The command"show mls qos inter fa 0/1 statics" can only show you the traffic of seperate DSCP,instead of seperate queues. Thanks in advance. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=66055&t=66055 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: PHBs in QoS [7:65947]
In English (well almost), a PHB is a Per Hop Behaviour. A quick google search give the following document - www.sfc.wide.ad.jp/~uhyo/html/japan/STREAM/ppt/RTMach99-kiri.pdf It is the DSCP (Differentiated Services Code Point) that needs to and identified. See RFCs 2474 and 2475 for more details. HTH Dom Stocqueler -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: 21 March 2003 19:07 To: [EMAIL PROTECTED] Subject: PHBs in QoS [7:65947] Can somone please (in English) tell me what a PHB actually is and how I can identify or set one. Differentiated Services Field: 0x00 (DSCP 0x00: Default; ECN: 0x00) 00.. = Differentiated Services Codepoint: Default (0x00) ..0. = ECN-Capable Transport (ECT): 0 ...0 = ECN-CE: 0 Many thx :)) Ken Farrington Global Networks, Barclays Capital, 5 The North Colonnade, Canary Wharf, London, E14 4BB * Tel : 020 7773 3550 * Mob : 07768-866655 * [EMAIL PROTECTED] For more information about Barclays Capital, please visit our web site at http://www.barcap.com. Internet communications are not secure and therefore the Barclays Group does not accept legal responsibility for the contents of this message. Although the Barclays Group operates anti-virus programmes, it does not accept responsibility for any damage whatsoever that is caused by viruses being passed. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Barclays Group. Replies to this email may be monitored by the Barclays Group for operational or business reasons. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=65975&t=65947 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
PHBs in QoS [7:65947]
Can somone please (in English) tell me what a PHB actually is and how I can identify or set one. Differentiated Services Field: 0x00 (DSCP 0x00: Default; ECN: 0x00) 00.. = Differentiated Services Codepoint: Default (0x00) ..0. = ECN-Capable Transport (ECT): 0 ...0 = ECN-CE: 0 Many thx :)) Ken Farrington Global Networks, Barclays Capital, 5 The North Colonnade, Canary Wharf, London, E14 4BB * Tel : 020 7773 3550 * Mob : 07768-866655 * [EMAIL PROTECTED] For more information about Barclays Capital, please visit our web site at http://www.barcap.com. Internet communications are not secure and therefore the Barclays Group does not accept legal responsibility for the contents of this message. Although the Barclays Group operates anti-virus programmes, it does not accept responsibility for any damage whatsoever that is caused by viruses being passed. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Barclays Group. Replies to this email may be monitored by the Barclays Group for operational or business reasons. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=65947&t=65947 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Modular QoS CLI [7:65074]
>I am not sure how I should word my access-list or the police statements >(bps, normal burst) I am not sure if I undertood what you are asking. On the access-list you specify the flow that you want to give the bandwidth specified on the policy-map. >interface vlan 90 >service-policy input limit-traffic >I want to limit traffic coming in and going out. With this command you are applying the traffic entering the interface. Maybe CAR would be better if you want to always limit the internal traffic from reaching the Internet. For example, if your internal network is connected to FA 0/0 and you external network is connected to ATM0/0/0, you would apply CAR on you FA0/0 interface If I remember right, with Policy map you are only limiting the traffic when congestion occur; if there is no congestion, the users will use the available bandwidth. What platform are you going to apply this? I am asking because you may be sure that the traffic is going to the interface (not the case when using MLS). "Naomi James" @groupstudy.com em 11/03/2003 15:09:51 Favor responder a "Naomi James" Enviado Por: [EMAIL PROTECTED] Para: [EMAIL PROTECTED] cc: Assunto:Modular QoS CLI [7:65074] I am trying to limit the traffic from two of my subnet using modular QoS CLI. I created a traffic class for my two subnets. class-map match-all res-traffic match access-group 101 class-map match-all wire-traffic match access-group 102 I defined a service policy name. policy-map limit-traffic class res-traffic police "bps" "normal burst" exceed-action drop policy-map limit-traffic class wire-traffic police "bps" "normal burst" exceed-action drop I am going to apply this service policy to my interface connecting to the Internet. interface vlan 90 service-policy input limit-traffic I am not sure how I should word my access-list or the police statements (bps, normal burst) so that I can stop these two subnets from saturating our bandwidth (6 MB) to the Internet. I want to limit traffic coming in and going out. access-list 101 deny ip "ip address" any access-list 101 permit ip any any access-list 102 deny ip "ip address" any access-list 102 permit ip any any Thanks for your help. Naomi James Computer Services and Information Technology Savannah State University 912-356-2509 [GroupStudy removed an attachment of type image/gif which had a name of Mabelt.gif] [GroupStudy removed an attachment of type image/gif which had a name of Mabelb.gif] Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=65160&t=65074 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Modular QoS CLI [7:65074]
I am trying to limit the traffic from two of my subnet using modular QoS CLI. I created a traffic class for my two subnets. class-map match-all res-traffic match access-group 101 class-map match-all wire-traffic match access-group 102 I defined a service policy name. policy-map limit-traffic class res-traffic police "bps" "normal burst" exceed-action drop policy-map limit-traffic class wire-traffic police "bps" "normal burst" exceed-action drop I am going to apply this service policy to my interface connecting to the Internet. interface vlan 90 service-policy input limit-traffic I am not sure how I should word my access-list or the police statements (bps, normal burst) so that I can stop these two subnets from saturating our bandwidth (6 MB) to the Internet. I want to limit traffic coming in and going out. access-list 101 deny ip "ip address" any access-list 101 permit ip any any access-list 102 deny ip "ip address" any access-list 102 permit ip any any Thanks for your help. Naomi James Computer Services and Information Technology Savannah State University 912-356-2509 [GroupStudy removed an attachment of type image/gif which had a name of Mabelt.gif] [GroupStudy removed an attachment of type image/gif which had a name of Mabelb.gif] Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=65074&t=65074 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: mpls, QoS, etc... [7:64810]
Also this one. http://www.netcraftsmen.net/welcher/ ""Router Kid"" wrote in message news:[EMAIL PROTECTED] > http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos > _c/qcfbook.pdf > > Let me know if you like it :) > > > > ""Xy Hien Le"" wrote in message > news:[EMAIL PROTECTED] > > Hi All, > > > > Can anyone show me a few links to some good QoS, MPLS practice labs? > > Thanks in advance. > > Xy Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64845&t=64810 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: mpls, QoS, etc... [7:64810]
http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos _c/qcfbook.pdf Let me know if you like it :) ""Xy Hien Le"" wrote in message news:[EMAIL PROTECTED] > Hi All, > > Can anyone show me a few links to some good QoS, MPLS practice labs? > Thanks in advance. > Xy Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64844&t=64810 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
mpls, QoS, etc... [7:64810]
Hi All, Can anyone show me a few links to some good QoS, MPLS practice labs? Thanks in advance. Xy Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64810&t=64810 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Newbie QoS question [7:64466]
Its so simple you are using Weighted Fair Queue in your network if you want different download rate in different systems so use Priority Queueing for assuing Priority to different ip address. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64531&t=64466 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Newbie QoS question [7:64466]
Hello, I'm trying to setup the most simple configuration, in order to show how QoS works on a 827H. This is what I do: access-list 120 permit ip host 192.168.0.13 any access-list 130 permit ip host 192.168.0.7 any class-map one match access-group 120 class-map two match access-group 130 policy-map test class one bandwidth 8 class two bandwidth 16 int e0 service-policy output test // only way to apply this policy Router#sh pol Policy Map test Class one Weighted Fair Queueing Bandwidth 8 (kbps) Max Threshold 64 (packets) Class two Weighted Fair Queueing Bandwidth 16 (kbps) Max Threshold 64 (packets) Router#sh pol int Ethernet0 Service-policy output: test Class-map: one (match-all) 0 packets, 0 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: access-group 120 Weighted Fair Queueing Output Queue: Conversation 265 Bandwidth 8 (kbps) Max Threshold 64 (packets) (pkts matched/bytes matched) 0/0 (depth/total drops/no-buffer drops) 0/0/0 Class-map: two (match-all) 0 packets, 0 bytes 5 minute offered rate 0 bps, drop rate 0 bps Match: access-group 130 Weighted Fair Queueing Output Queue: Conversation 266 Bandwidth 16 (kbps) Max Threshold 64 (packets) (pkts matched/bytes matched) 0/0 (depth/total drops/no-buffer drops) 0/0/0 Class-map: class-default (match-any) 10265 packets, 1107280 bytes 5 minute offered rate 4000 bps, drop rate 0 bps Match: any What's the problem? I would expect to see a different download rate from 192.168.0.13 than 192.168.0.7 But, both of them have the same download rate. I would highly appreciate any suggestion. Thanks to all of you. Regards. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64466&t=64466 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QOS BOOK FOR CCIE WRITTEN EXAM [7:64052]
sorry guys. I did paste a link on my post but no idea what happened. I hope it works this time. http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fqos _c/qcfbook.pdf Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64053&t=64052 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QOS BOOK FOR CCIE WRITTEN EXAM [7:64052]
_c/qcfbook.pdf People who are studying for ccie written will find alot of usefull info on QOS. Can anyone share a solid document on MPLS with me? Regards! Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64052&t=64052 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS difficult (?) question [7:63633]
I would use the diff serve bit to modify the packet header to give class A traffic a higher priority. That way, that packet gets sent first from the output queue. -Original Message- From: ira [mailto:[EMAIL PROTECTED] Sent: 25 February 2003 19:35 To: [EMAIL PROTECTED] Subject: Re: QoS difficult (?) question [7:63633] No. If there is availabe bandw over the guaranteed bandw that is used, I want to assign it with priority to class A. If class A doesn-t need the bandw then to class B and so on. CQ assigns the remaining band to all the traffic patterns (it can not use priorities, as far as I know) It must be a frankestein method to use... Tks --- Ajit wrote: > custom q it !! > - Original Message - > From: "ira" > To: > Sent: Monday, February 24, 2003 11:35 PM > Subject: QoS difficult (?) question [7:63633] > > > > hallo Qos ! > > > > I have 5 types of traffic: A, B, C, D, and E. > > I want to reserve 20% of bandw for each type. > > If I have no traffic (or less than the reserved > > 20%)for some of the types, I want that this > remaining > > bandw to be allocated to the other types of > traffic, > > but in a priority way : A > > > Any suggestions ? > > > > Thanks! > > > > __ > > Do you Yahoo!? > > Yahoo! Tax Center - forms, calculators, tips, more > > http://taxes.yahoo.com/ > [EMAIL PROTECTED] > > > > __ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/ Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63840&t=63633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS difficult (?) question [7:63633]
No. If there is availabe bandw over the guaranteed bandw that is used, I want to assign it with priority to class A. If class A doesn-t need the bandw then to class B and so on. CQ assigns the remaining band to all the traffic patterns (it can not use priorities, as far as I know) It must be a frankestein method to use... Tks --- Ajit wrote: > custom q it !! > - Original Message - > From: "ira" > To: > Sent: Monday, February 24, 2003 11:35 PM > Subject: QoS difficult (?) question [7:63633] > > > > hallo Qos ! > > > > I have 5 types of traffic: A, B, C, D, and E. > > I want to reserve 20% of bandw for each type. > > If I have no traffic (or less than the reserved > > 20%)for some of the types, I want that this > remaining > > bandw to be allocated to the other types of > traffic, > > but in a priority way : A > > > Any suggestions ? > > > > Thanks! > > > > __ > > Do you Yahoo!? > > Yahoo! Tax Center - forms, calculators, tips, more > > http://taxes.yahoo.com/ > [EMAIL PROTECTED] > > > > __ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/ Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63750&t=63633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS difficult (?) question [7:63633]
With CQ you can reserve 20% for each class of traffic. As far as I remember, if one class do not have traffic, the router will not lose time serving this queue, so it will go back, faster, to the other queue. Also, this treatment will occur in case of congestion. With this approach there is no guarantee that a class traffic will use just 20%. From what I understood, that is what you want. In the other hand, if you want to limit the traffic, use CAR. "Jake" @groupstudy.com em 24/02/2003 16:29:42 Favor responder a "Jake" Enviado Por: [EMAIL PROTECTED] Para: [EMAIL PROTECTED] cc: Assunto:Re: QoS difficult (?) question [7:63633] How are you prioritizing traffic? LLQ, priority queueing ect...? Are yor prioritizing over frame, point to point, lan ect...? ""ira"" wrote in message news:[EMAIL PROTECTED] > hallo Qos ! > > I have 5 types of traffic: A, B, C, D, and E. > I want to reserve 20% of bandw for each type. > If I have no traffic (or less than the reserved > 20%)for some of the types, I want that this remaining > bandw to be allocated to the other types of traffic, > but in a priority way : A > Any suggestions ? > > Thanks! > > __ > Do you Yahoo!? > Yahoo! Tax Center - forms, calculators, tips, more > http://taxes.yahoo.com/ Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63719&t=63633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS difficult (?) question [7:63633]
How are you prioritizing traffic? LLQ, priority queueing ect...? Are yor prioritizing over frame, point to point, lan ect...? ""ira"" wrote in message news:[EMAIL PROTECTED] > hallo Qos ! > > I have 5 types of traffic: A, B, C, D, and E. > I want to reserve 20% of bandw for each type. > If I have no traffic (or less than the reserved > 20%)for some of the types, I want that this remaining > bandw to be allocated to the other types of traffic, > but in a priority way : A > Any suggestions ? > > Thanks! > > __ > Do you Yahoo!? > Yahoo! Tax Center - forms, calculators, tips, more > http://taxes.yahoo.com/ Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63640&t=63633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS difficult (?) question [7:63633]
hallo Qos ! I have 5 types of traffic: A, B, C, D, and E. I want to reserve 20% of bandw for each type. If I have no traffic (or less than the reserved 20%)for some of the types, I want that this remaining bandw to be allocated to the other types of traffic, but in a priority way : Ahttp://taxes.yahoo.com/ Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63633&t=63633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS Issue - voice experts only please [7:63562]
Hello voice and QoS guru's. We are currently finishing up on installing a Nortel (I know, sorry) Option11c at our headquarters site and Nortel BCM 3.0's at remote sites. Every site is connected to headquarters via a leased T1 w/ PPP as encap. The routers aren't doing any of the voice over IP, all that is happening in the BCM. But we've noticed between one of our sites we've got a little hissing sound and I myself believe 100% it's not the link, it's the certified Nortel contractors. We're currently getting averages of 8/10/12 ms on the link with 255/255 reliability. So just to make the corporate guys happy, we're going to put QoS on the link. Since the BCM is doing all the Voice over IP I was just curious if my additions would be effective, please let me know, I will only be applying this to the serial interfaces: ip rtp header-compression ip rtp reserve 16384 100 784 (The second line is where I'm confused, will this work with my scenario?) Sorry for my lack of QoS. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63562&t=63562 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS 3550 burst size [7:63324]
I am having trouble calculating some real world burst sizes for QoS. My goal is simply rate limit TCP as closely to 1.0 Mb as possible. I understand the sawtooth will make this difficult and the fact that packet sizes are ever changing, but I am basing my caluclation on 1500 byte Ethernet. the command specifically is: police 100 burst-size exceed-action drop It is this burst size that I am not sure how to calculate. Using Cisco's formula I tend to get very small numbers ( 250 bits etc...the default minimum is 8KB!) I have played around with a traffic generator and kind eyeballed that for 1.0 Mb of traffic a burst size of 125000 seems to work. Is this a reasonable number? What have other poeple used? Are there any guidelines to what this should be set to? Below is a config of what I have: switch# sh class Class Map match-all ANY (id 2) Match access-group 101 switch# sh access-li Extended IP access list 101 permit ip any any switch# sh poli Policy Map test3 class ANY police 100 125000 exceed-action drop switch# sh run int f0/1 Current configuration : 109 bytes ! interface FastEthernet0/18 switchport mode access no ip address service-policy input test3 end Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63324&t=63324 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS on 3550 Aargh! [7:63164]
uh yup, but let me preface it with, "I'M AN IDIOT!" to spare myself further embarrassment I will repeat the age-old adage "Check Layer 1 first" I have it working now. The config posted was correct and functional just had to remember which port to put service-policy on!!! SmaatBits on 0/48 test machine on 0/18...well, putting the service-policy on 0/18 input does not help...needs to be on 0/48. Doh! Check layer 1, Check layer 1 - Original Message - From: "The Long and Winding Road" To: Sent: Monday, February 17, 2003 2:13 PM Subject: Re: QoS on 3550 Aargh! [7:63164] > can you provide a sanitized config for the access-list in question and for > the interface in question? > > -- > TANSTAAFL > "there ain't no such thing as a free lunch" > > > > > ""Jim Devane"" wrote in message > [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > > I am completely frustrated. > > > > I am trying to do something very simple but am having considerable > trouble. > > I wish only to rate-limit ALL packets coming into a particular interface > on > > a 3550 > > It does have EMI and Qos is anabled. This is the config that I have tried > so > > far and the packets just blast right through... I know the burst is larger > > tahn the max speed, should not matter. Incidentally, I entered 500 000 for > > both values but the switch auto-changed the first value( I believe since > it > > is in values of 8 Kbps) > > > > Any ideas? I have read the CCO doco on this over and over and I cannot see > > what I am missing. I suspect somethign in my class map is wrong, but I am > > not sure how to manipulate it... > > > > any help appreciated. > > > > > > pwps-esw01#sh class > > pwps-esw01#sh class-map > > Class Map match-all test2 (id 3) > >Match access-group 123 > > > > Class Map match-all test1 (id 2) > >Match any > > Class Map match-any class-default (id 0) > >Match any > > pwps-esw01#sh poli > > pwps-esw01#sh policy-map > > Policy Map int18 > > class test2 > >police 496000 50 exceed-action drop > > > > pwps-esw01#sh mls qos int f0/18 > > FastEthernet0/18 > > Attached policy-map for Ingress: int18 > > trust state: not trusted > > trust mode: not trusted > > COS override: dis > > default COS: 0 > > DSCP Mutation Map: Default DSCP Mutation Map > > trust device: none > > > > pwps-esw01#sh mls qos int f0/18 st > > FastEthernet0/18 > > Ingress > > dscp: incoming no_change classified policeddropped (in bytes) > > Others: 14938711 14938711 0 0 0 > > Egress > > dscp: incoming no_change classified policeddropped (in bytes) > > Others: 691426721 n/a n/a 0 0 > > > > pwps-esw01# Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63184&t=63164 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS on 3550 Aargh! [7:63164]
can you provide a sanitized config for the access-list in question and for the interface in question? -- TANSTAAFL "there ain't no such thing as a free lunch" ""Jim Devane"" wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > I am completely frustrated. > > I am trying to do something very simple but am having considerable trouble. > I wish only to rate-limit ALL packets coming into a particular interface on > a 3550 > It does have EMI and Qos is anabled. This is the config that I have tried so > far and the packets just blast right through... I know the burst is larger > tahn the max speed, should not matter. Incidentally, I entered 500 000 for > both values but the switch auto-changed the first value( I believe since it > is in values of 8 Kbps) > > Any ideas? I have read the CCO doco on this over and over and I cannot see > what I am missing. I suspect somethign in my class map is wrong, but I am > not sure how to manipulate it... > > any help appreciated. > > > pwps-esw01#sh class > pwps-esw01#sh class-map > Class Map match-all test2 (id 3) >Match access-group 123 > > Class Map match-all test1 (id 2) >Match any > Class Map match-any class-default (id 0) >Match any > pwps-esw01#sh poli > pwps-esw01#sh policy-map > Policy Map int18 > class test2 >police 496000 50 exceed-action drop > > pwps-esw01#sh mls qos int f0/18 > FastEthernet0/18 > Attached policy-map for Ingress: int18 > trust state: not trusted > trust mode: not trusted > COS override: dis > default COS: 0 > DSCP Mutation Map: Default DSCP Mutation Map > trust device: none > > pwps-esw01#sh mls qos int f0/18 st > FastEthernet0/18 > Ingress > dscp: incoming no_change classified policeddropped (in bytes) > Others: 14938711 14938711 0 0 0 > Egress > dscp: incoming no_change classified policeddropped (in bytes) > Others: 691426721 n/a n/a 0 0 > > pwps-esw01# Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63177&t=63164 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS on 3550 Aargh! [7:63164]
I am completely frustrated. I am trying to do something very simple but am having considerable trouble. I wish only to rate-limit ALL packets coming into a particular interface on a 3550 It does have EMI and Qos is anabled. This is the config that I have tried so far and the packets just blast right through... I know the burst is larger tahn the max speed, should not matter. Incidentally, I entered 500 000 for both values but the switch auto-changed the first value( I believe since it is in values of 8 Kbps) Any ideas? I have read the CCO doco on this over and over and I cannot see what I am missing. I suspect somethign in my class map is wrong, but I am not sure how to manipulate it... any help appreciated. pwps-esw01#sh class pwps-esw01#sh class-map Class Map match-all test2 (id 3) Match access-group 123 Class Map match-all test1 (id 2) Match any Class Map match-any class-default (id 0) Match any pwps-esw01#sh poli pwps-esw01#sh policy-map Policy Map int18 class test2 police 496000 50 exceed-action drop pwps-esw01#sh mls qos int f0/18 FastEthernet0/18 Attached policy-map for Ingress: int18 trust state: not trusted trust mode: not trusted COS override: dis default COS: 0 DSCP Mutation Map: Default DSCP Mutation Map trust device: none pwps-esw01#sh mls qos int f0/18 st FastEthernet0/18 Ingress dscp: incoming no_change classified policeddropped (in bytes) Others: 14938711 14938711 0 0 0 Egress dscp: incoming no_change classified policeddropped (in bytes) Others: 691426721 n/a n/a 0 0 pwps-esw01# Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63164&t=63164 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Qos [7:63014]
You could configure LLQ for audio; but for video, I have seem some articles saying that video could starve other traffic. One approach would be limit your video, for example 200 Kbps, using CAR, and after that apply LLQ. Regards, -- Alaerte Gladston Vidali Phone: 55 11 3268 - 2879 Cel.: 55 11 9619-3382 "Projet AIM" @groupstudy.com em 14/02/2003 06:31:25 Favor responder a "Projet AIM" Enviado Por: [EMAIL PROTECTED] Para: [EMAIL PROTECTED] cc: Assunto:Qos [7:63014] Hi all, I have a project on a cisco 3620 router and I want to know how can i configure the quality of service on this router to priviledge the video traffic over the other sincerly PIPPOO _ MSN Search, le moteur de recherche qui pense comme vous ! http://search.msn.fr/worldwide.asp Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63026&t=63014 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Qos [7:63014]
Hi all, I have a project on a cisco 3620 router and I want to know how can i configure the quality of service on this router to priviledge the video traffic over the other sincerly PIPPOO _ MSN Search, le moteur de recherche qui pense comme vous ! http://search.msn.fr/worldwide.asp Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=63014&t=63014 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS Question [7:62351]
Hi, CBWFQ will meet your requirement. If no VPN traffic, the rest will fill the line. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62565&t=62351 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QOS Question [7:62351]
Using one T1 IP circuit, I need to allocate a certain percentage for VPN and the rest for generic Internet traffic. Would CBWFQ be the best solution to reserve 60% for VPN? And if it wasn't used for VPN, could it be used for other? Thanks Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62351&t=62351 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS - Migration Path from MC3810 and 7000 Routers [7:62230]
When you have ip phones, QoS should be addressed starting from layer 2. It is good thing let the swithes do QoS for your voice packets before they get to the routers. I think you can take a look at 2950. As for the core, unless you will have some kind of MAN connections in the future, any reason you can not get by with something like 7200s? You have 3810 and no voice, I do not think you should repeat the pattern with 7600 and no gigabit. Any reason you want to replace FR with ATM in the future? Since you rely on IP QoS, what layer 2 you are using at WAN should not matter, ip/atm/frame/, as long as the carrier gives you the SLA you need for the voice traffic. If you do not have redundancy at your central WAN aggregation point, you may consider it right now, since voice is likely way critical than data for many compaines. Just my .02 --kent - Original Message - From: "Firesox" To: Sent: Thursday, January 30, 2003 9:07 PM Subject: QoS - Migration Path from MC3810 and 7000 Routers [7:62230] > I am in need of any suggestions/comments on the following migration path. > > Currently 7000 Router in the Core over 30 Frame PVC to remote sites with > MC3810. > Running only data right now, but deploying VoIP with Avaya ECLIPS solution. > I would like to stay with Cisco Routers and MC3810 will be end of life soon > if not already and 7000 as well. > In the future ATM will replace the entire frame-relay cloud. There will be > about 1000 IP phones deloyed. > I need good QoS and scalable solution for future ATM migration. > I am thinking 7600 series in the Core and either or 2600 on the edge. > > Any comments would be appreciated. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62263&t=62230 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS - Migration Path from MC3810 and 7000 Routers [7:62230]
I think it is a good option. Maybe 3600 for remote sites that require more usage. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62245&t=62230 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS - Migration Path from MC3810 and 7000 Routers [7:62230]
I am in need of any suggestions/comments on the following migration path. Currently 7000 Router in the Core over 30 Frame PVC to remote sites with MC3810. Running only data right now, but deploying VoIP with Avaya ECLIPS solution. I would like to stay with Cisco Routers and MC3810 will be end of life soon if not already and 7000 as well. In the future ATM will replace the entire frame-relay cloud. There will be about 1000 IP phones deloyed. I need good QoS and scalable solution for future ATM migration. I am thinking 7600 series in the Core and either or 2600 on the edge. Any comments would be appreciated. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=62230&t=62230 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Read "Administering QoS in Cisco IP Networks"? [7:61826]
I've offered my opinion on the Cisco Press title on QoS on more than one occasion. Anyone care to offer their own on the above title from Syngress? I've not yet finished chapter 1 and I already have suspicions. And why such extensive coverage of EIGRP (something like 70 or 80 pages)? I wasn't aware that EIGRP was so QoS-capable that it would deserve such a showcasing. QoS coverage doesn't even begin until around page 120. I would also be very grateful for any suggestions. I see that CP is coming out with a new title but no release date (actually, it isn't even mentioned on the CP website but it's listed as pending on barnesandnoble.com). I see that Sybex has a CCIP QoS / multicast study guide, but I've never been a huge fan of Sybex. Maybe this one is worthwhile? Thanks all. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61826&t=61826 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS on 2621xm [7:61353]
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Nope, The only engineering you can do is at the cpe, where your traffic goes out and comes in. Thais makes that you at best can configure QOS at the BOTTLENECK, that may be your remote office router. If not applicable, than the agregation point (HQ) will be the next best. I still would say that you carefully analyse the traffic patterns and look at the bottlenecks. That is the no 1 point to do business. Martijn - -Oorspronkelijk bericht- Van: Julian Pentermann [mailto:[EMAIL PROTECTED]] Verzonden: dinsdag 21 januari 2003 6:58 Aan: "mjans001" Onderwerp: Re: QOS on 2621xm [7:61353] would the isp have to do anything or would i just impliment the qos on my router? Thanks for the help - - Original Message - From: ""mjans001"" Newsgroups: groupstudy.cisco Sent: Tuesday, January 21, 2003 12:53 AM Subject: RE: QOS on 2621xm [7:61353] > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > You may need to use Priority Queueing, and hardcode telnet High prio > based on an access-list. > > Normal traffic despools after telnet queue is empty. If you are sure > that there will always be bandwitfh left for other traffic, PQ will do > fine. > > That is one way of using it. > > >>> > During transmission, PQ gives priority queues absolute preferential > treatment over low priority queues; important traffic, given the > highest priority, always takes precedence over less important traffic. > Packets are classified based on user-specified criteria and placed > into one of the four > output queues-high, medium, normal, and low-based on the assigned priority. > Packets that are not classified by priority fall into the normal > queue. Figure 7 illustrates this process. > > Congestion Management Overview > http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/12cgcr/qos_c /qcpart2/qcconman.htm > > > Why Use Priority Queueing? > PQ provides absolute preferential treatment to high priority traffic, > ensuring that mission-critical traffic traversing various WAN links > gets priority treatment. In addition, PQ provides a faster response > time than do > other methods of queueing. > > Although you can enable priority output queueing for any interface, it > is best used for low-bandwidth, congested serial interfaces. > > Considerations > When choosing to use PQ, consider that because lower priority traffic > is often denied bandwidth in favor of higher priority traffic, use of > PQ could, > in the worst case, result in lower priority traffic never being transmitted. > To avoid inflicting these conditions on lower priority traffic, you > can use > traffic shaping or CAR to rate-limit the higher priority traffic. > > PQ introduces extra overhead that is acceptable for slow interfaces, > but may > not be acceptable for higher speed interfaces such as Ethernet. With > PQ enabled, the system takes longer to switch packets because the > packets are classified by the processor card. > > PQ uses a static configuration and does not adapt to changing network > conditions. > > > > > >>>>>> > > Martijn > > - -Oorspronkelijk bericht- > Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Namens > Julian P > Verzonden: maandag 20 januari 2003 9:02 > Aan: [EMAIL PROTECTED] > Onderwerp: QOS on 2621xm [7:61353] > > > Hi > > > We would like to prioritize incoming traffic on our 256k internet link > to uunet .We need to give telnet at least 64k incoming bandwidth. > > Any ideas on the best way to do this ? > > Thanks in advance > > Julian > Version: PGP 8.0 > > iQA/AwUBPix7Bndq56XWk+VyEQJ+/ACfS2LZO44i+6Y+cRg37a/ApiovJtgAoLvz > kS6ZvDnOtSXEqAAi/6u1v+p4 > =nXJB > -END PGP SIGNATURE- Version: PGP 8.0 iQA/AwUBPi2VbHdq56XWk+VyEQIPOQCfTguOnPMduMdxWbRuzbadddit3esAn3/6 vmrK61ZimecTbrS2DXPX3Jwo =FsQK -END PGP SIGNATURE- Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61482&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS on 2621xm [7:61353]
What would he do if he had a FR link to a remote site and he wanted to ensure his high priority traffic was sent without DE, whilst low priority traffic can burst and be sent as DE - does FRTS (for Telnet only) solve this problem (if that can be done)? Rik -Original Message- From: mjans001 [mailto:[EMAIL PROTECTED]] Sent: Tuesday, 21 January 2003 6:54 AM To: [EMAIL PROTECTED] Subject: RE: QOS on 2621xm [7:61353] -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 You may need to use Priority Queueing, and hardcode telnet High prio based on an access-list. Normal traffic despools after telnet queue is empty. If you are sure that there will always be bandwitfh left for other traffic, PQ will do fine. That is one way of using it. >>> During transmission, PQ gives priority queues absolute preferential treatment over low priority queues; important traffic, given the highest priority, always takes precedence over less important traffic. Packets are classified based on user-specified criteria and placed into one of the four output queues-high, medium, normal, and low-based on the assigned priority. Packets that are not classified by priority fall into the normal queue. Figure 7 illustrates this process. Congestion Management Overview http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/12cgcr/qos_c /qcpart2/qcconman.htm Why Use Priority Queueing? PQ provides absolute preferential treatment to high priority traffic, ensuring that mission-critical traffic traversing various WAN links gets priority treatment. In addition, PQ provides a faster response time than do other methods of queueing. Although you can enable priority output queueing for any interface, it is best used for low-bandwidth, congested serial interfaces. Considerations When choosing to use PQ, consider that because lower priority traffic is often denied bandwidth in favor of higher priority traffic, use of PQ could, in the worst case, result in lower priority traffic never being transmitted. To avoid inflicting these conditions on lower priority traffic, you can use traffic shaping or CAR to rate-limit the higher priority traffic. PQ introduces extra overhead that is acceptable for slow interfaces, but may not be acceptable for higher speed interfaces such as Ethernet. With PQ enabled, the system takes longer to switch packets because the packets are classified by the processor card. PQ uses a static configuration and does not adapt to changing network conditions. >>>>>> Martijn - -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Namens Julian P Verzonden: maandag 20 januari 2003 9:02 Aan: [EMAIL PROTECTED] Onderwerp: QOS on 2621xm [7:61353] Hi We would like to prioritize incoming traffic on our 256k internet link to uunet .We need to give telnet at least 64k incoming bandwidth. Any ideas on the best way to do this ? Thanks in advance Julian Version: PGP 8.0 iQA/AwUBPix7Bndq56XWk+VyEQJ+/ACfS2LZO44i+6Y+cRg37a/ApiovJtgAoLvz kS6ZvDnOtSXEqAAi/6u1v+p4 =nXJB -END PGP SIGNATURE- Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61425&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS on 2621xm [7:61353]
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 You may need to use Priority Queueing, and hardcode telnet High prio based on an access-list. Normal traffic despools after telnet queue is empty. If you are sure that there will always be bandwitfh left for other traffic, PQ will do fine. That is one way of using it. >>> During transmission, PQ gives priority queues absolute preferential treatment over low priority queues; important traffic, given the highest priority, always takes precedence over less important traffic. Packets are classified based on user-specified criteria and placed into one of the four output queues-high, medium, normal, and low-based on the assigned priority. Packets that are not classified by priority fall into the normal queue. Figure 7 illustrates this process. Congestion Management Overview http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/12cgcr/qos_c/qcpart2/qcconman.htm Why Use Priority Queueing? PQ provides absolute preferential treatment to high priority traffic, ensuring that mission-critical traffic traversing various WAN links gets priority treatment. In addition, PQ provides a faster response time than do other methods of queueing. Although you can enable priority output queueing for any interface, it is best used for low-bandwidth, congested serial interfaces. Considerations When choosing to use PQ, consider that because lower priority traffic is often denied bandwidth in favor of higher priority traffic, use of PQ could, in the worst case, result in lower priority traffic never being transmitted. To avoid inflicting these conditions on lower priority traffic, you can use traffic shaping or CAR to rate-limit the higher priority traffic. PQ introduces extra overhead that is acceptable for slow interfaces, but may not be acceptable for higher speed interfaces such as Ethernet. With PQ enabled, the system takes longer to switch packets because the packets are classified by the processor card. PQ uses a static configuration and does not adapt to changing network conditions. >>>>>> Martijn - -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Namens Julian P Verzonden: maandag 20 januari 2003 9:02 Aan: [EMAIL PROTECTED] Onderwerp: QOS on 2621xm [7:61353] Hi We would like to prioritize incoming traffic on our 256k internet link to uunet .We need to give telnet at least 64k incoming bandwidth. Any ideas on the best way to do this ? Thanks in advance Julian Version: PGP 8.0 iQA/AwUBPix7Bndq56XWk+VyEQJ+/ACfS2LZO44i+6Y+cRg37a/ApiovJtgAoLvz kS6ZvDnOtSXEqAAi/6u1v+p4 =nXJB -END PGP SIGNATURE- Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61420&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS on 2621xm [7:61353]
In QoS now, there is more than just setting the bandwidth allocated to each class. What you need to do is combine this with the diffserv bit. This is basically the newer version of the TOS bit in the IP header that defines what sort of service happens to the packet. In a nut shell, this means which packets gets sent first in the output queue. As telnet is very delay sensitive, you should give it a higher diffserv bit. What you basically do is create class maps for each traffic type and apply it to the Ethernet interface so that the router modifies the packets as it enters the router. As the packet leaves the interface, you use normal service policies. Sample config below: class-map match-all Gold match access-group 170 class-map match-all Bronze match access-group 172 class-map match-all Telnet_SSH match access-group 131 class-map match-all Silver match access-group 171 ! policy-map SET_DIFFSERV class Telnet_SSH set ip dscp cs5 policy-map class Gold bandwidth percent (%) class Bronze bandwidth percent (%) ! interface FastEthernet0/0 description Link to LAN ip address x.x.x.x y.y.y.y speed 100 full-duplex service-policy input SET_DIFFSERV ! interface Serial0/0 bandwidth 512 ip address a.a.a.a b.b.b.b service-policy output access-list 131 remark Prioritise Telnet and SSH access-list 131 permit tcp any any eq telnet access-list 131 permit tcp any any eq 22 access-list 170 remark Gold Class QoS access-list 170 permit ip host xx host xxx access-list 170 permit ip host xxx host xxx access-list 170 permit tcp host eq 443 any access-list 170 permit tcp host eq www any access-list 170 permit udp any any eq domain access-list 172 remark Bronze Class QoS access-list 172 permit ip host host x > > -Original Message- > From: Julian P [mailto:[EMAIL PROTECTED]] > Sent: 20 January 2003 10:02 > To: [EMAIL PROTECTED] > Subject: QOS on 2621xm [7:61353] > > > Hi > > > We would like to prioritize incoming traffic on our 256k internet link to > uunet .We need to give telnet at least 64k incoming bandwidth. > > Any ideas on the best way to do this ? > > Thanks in advance > > Julian Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61365&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QOS on 2621xm [7:61353]
You will need to QoS on both sides of the link. Your side should be fine, but good luck with the ISP side. I would use the diffserve bits as well as bandwidth allocation using policy maps. Andrew -Original Message- From: Julian P [mailto:[EMAIL PROTECTED]] Sent: 20 January 2003 10:02 To: [EMAIL PROTECTED] Subject: QOS on 2621xm [7:61353] Hi We would like to prioritize incoming traffic on our 256k internet link to uunet .We need to give telnet at least 64k incoming bandwidth. Any ideas on the best way to do this ? Thanks in advance Julian Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61357&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QOS on 2621xm [7:61353]
Hi We would like to prioritize incoming traffic on our 256k internet link to uunet .We need to give telnet at least 64k incoming bandwidth. Any ideas on the best way to do this ? Thanks in advance Julian Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61353&t=61353 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QOS on IOS 12.1.10 2610 Router [7:61269]
I can only apply service-policy out "name" to the main interface but not to sub interfaces. Can some one tell me if it automatically applyes to sub interfaces. Thanks Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61269&t=61269 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS suggestion [7:60994]
Hi, I have the following testing setup but it looks like the LLQ does not work. Can you have a look on it? When the 256k link was congested. Why I ping the prec. 5 packet behind the 256k line it only have the same response time with default ping? 128k--- FR 256k Attached 256k router configuration below. class-map match-all voice-traffic match ip precedence 5 ! policy-map voice-policy class voice-traffic priority percent 75 class class-default fair-queue interface Serial0 bandwidth 256 no ip address encapsulation frame-relay IETF load-interval 30 no fair-queue frame-relay traffic-shaping frame-relay lmi-type ansi ! interface Serial0.10 point-to-point bandwidth 256 ip address 10.114.0.6 255.255.255.252 frame-relay interface-dlci 100 class llq ! map-class frame-relay llq frame-relay cir 256000 frame-relay bc 2560 frame-relay be 0 frame-relay mincir 256000 service-policy output voice-policy TIA rgds, ivan Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61074&t=60994 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS suggestion [7:60994]
i don't see any obvious problems with your configuration. I can, however,
offer a couple of troubleshooting tips. I would start by checking out the
access list ("show access-list") to make sure you have packets that qualify.
Second (and this is where I think your problem is), I would lose the
"match-all" in your class-map. Since you're only searching one criterion,
there's no need for the match-all (which is the default match clause
anyways). Here's a quote from Cisco's web site to confirm:
"The match all and match any options need to be specified only if more than
one match criterion is configured in the traffic class. The class-map
match-all command is used when all of the match criteria in the traffic
class must be met in order for a packet to match the specified traffic
class. The class-map match-any command is used when only one of the match
criterion in the traffic class must be met in order for a packet to match
the specified traffic class. If neither the match-all nor match-any keyword
is specified, the traffic class will behave in a manner consistent with
class-map match-all command"
Let me know what you find out. Hope this helps
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=61030&t=60994
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS suggestion [7:60994]
Do you have experience with LLQ and MSFC that you can share? I configured LLQ but it seems packets are not going to the priority queue: class-map match-all Priority-Queue match access-group name TV ! policy-map Policy class Priority-Queue priority 200 class class-default random-detect fair-queue ! interface ATM6/0/0.213 point-to-point bandwidth 1 ip address 192.168.255.177 255.255.255.252 ip pim version 1 ip pim sparse-dense-mode ip ospf cost 8 atm pvc 100 1 201 aal5snap service-policy output Policy ! ip access-list extended TV permit udp host 1.1.1.1 host 239.192.10.22 eq 6 router#sh policy-map int atm 6/0/0.213 output ATM6/0/0.213 service-policy output: Policy queue stats for all priority classes: queue size 0, queue limit 50 packets output 0, packet drops 0 tail/random drops 0, no buffer drops 0, other drops 0 class-map: Priority-Queue (match-all) 0 packets, 0 bytes 5 minute offered rate 0 bps, drop rate 0 bps match: access-group name TV 0 packets, 0 bytes 5 minute rate 0 bps Priority: kbps 200, burst bytes 5000, b/w exceed drops: 0 class-map: class-default (match-any) 474896 packets, 516105147 bytes 5 minute offered rate 1623000 bps, drop rate 0 bps match: any 474896 packets, 516105147 bytes 5 minute rate 1623000 bps queue size 0, queue limit 5838 packets output 477559, packet drops 4 tail/random drops 0, no buffer drops 0, other drops 4 random-detect: Exp-weight-constant: 9 (1/512) Mean queue depth: 0 Class Random Tail Minimum Maximum Mark Output drop drop threshold threshold probability packets 0 0 0 1459 2919 1/10 429315 1 0 0 1641 2919 1/100 2 0 0 1823 2919 1/100 3 0 0 2006 2919 1/100 4 0 0 2188 2919 1/100 5 0 0 2370 2919 1/1048467 6 0 0 2553 2919 1/100 7 0 0 2735 2919 1/100 fair-queue: per-flow queue limit 1459 Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61029&t=60994 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS suggestion [7:60994]
forgot to add one thing you probably already know this but if you decide to use LLQ for a PPP serial connection (like a t1 or frac t1) you will want to implement LFI (link fragmentation and interleave). this means that your config will be implemented on a "multilink1" interface rather than a physical interface. LFI allows you circumvent excessive serialization delays on slow WAN connections. This does not apply to frame relay interfaces. Cisco has some really good docs on this topic. Let me know if you would like more info. Hope this helps Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61025&t=60994 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: QoS suggestion [7:60994]
You've got a few options. The most basic (and most limited) is using IP RTP Priority. The will prioritize all RTP traffic on the applied interface. The best solution (IMHO) is to use LLQ. Low Latency Queueing can be thought of as CB-WFQ with the added benefit of a priority queue. This is probably what you want to do. Create a class-map (or map-class if it's a frame relay interface) and apply the voice traffic to the priority queue with the "priority" command, and then assign all your other traffic to a "fair queue". CB-WFQ does provide minimum bandwidth guarantee but it does not give you the priority queue that voice traffic likes so much. Hope this helps. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=61024&t=60994 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS suggestion [7:60994]
Hi, If I want to set the voice traffic have high priority and the rest will use fair queue, which cisco queuing method can achieve it? The hardware is 17xx or 26xx. Thanks. It looks like the LLQ or CBWFQ will have guaranteed bandwidth feature during congestion but seems not the solution I am looking for. TIA. rgds, ivan Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60994&t=60994 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS Charts [7:60710]
this link has it all:
http://www.cisco.com/en/US/tech/tk543/tech_topology_and_network_serv_and_pro
tocol_suite_home.html
Amar
Network Engineer
MCSE, CCNP
""Howard C. Berkowitz"" a icrit dans le message de news:
[EMAIL PROTECTED]
> >Howard,
> >
> >We'd all like to see a comprehensive QOS chart.
> >
> >Thanks,
> >
> >Kym
>
> They are Word tables in several different papers of mine. Just trying
> to paste them into email text results in something unreadable.
> Anyone good at
> reformatting tables to ASCII, when table cells have multiple lines?
>
> >
> >
> >
> >>From: "Jonathan V Hays"
> >>Reply-To: "Jonathan V Hays"
> >>To:
> >>Subject: RE: Small Minded QOS
> >>Date: Wed, 8 Jan 2003 22:17:53 -0500
> >>
> >>Howard,
> >>
> >>I, for one, would be quite interested in such a comparative table.
> >>
> >>Thanks in advance.
> >>
> >>Jonathan
> >>
> >>-Original Message-
> >>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
> >>Howard C. Berkowitz
> >>Sent: Wednesday, January 08, 2003 3:47 PM
> >>To: [EMAIL PROTECTED]
> >>Subject: Re: Small Minded QOS
> >>
> >>
> >>At 7:39 PM + 1/8/03, Jennifer Bellucci wrote:
> >>>Hello
> >>>
> >>>Can someone tell me where I can find every single QOS formula for all
> >>>the QOS techniques in one document, prefer it if it was one page.
> >>>Please do not send me the link to QOS on cco.
> >>
> >>Could you clarify what you mean by "formula"? When I hear that, I
> >>think of the derivation of the algorithms, and, certainly for things
> >>like RED and WFQ, these were published in scientific papers of many
> >>pages and equations. Indeed, if you go into the literature, you'll
> >>find that what we may call things are actually approximations ("true"
> >>WFQ is a theoretical abstraction; what we call WFQ is more like
> >>Self-Clocked WFQ).
> >>
> >>Now, if you are thinking of a comparative table of what the various
> >>IOS QOS mechanisms do, that's something I might be able to dig up. .
> >>.
> >rd
> >
> >_
> >STOP MORE SPAM with the new MSN 8 and get 2 months FREE*
> >http://join.msn.com/?page=features/junkmail
> >.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=60913&t=60710
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: QoS Charts [7:60710]
Could you send it to me? I could setup a ftp server open to anybody to
request them (and all docs which we find useful).
Francisco Sedano
Informatica Pronet
CCIE Student :-)
"Howard C. Berkowitz"
Enviado por: [EMAIL PROTECTED]
09/01/2003 17:39
Por favor, responda a "Howard C. Berkowitz"
Para: [EMAIL PROTECTED]
cc:
Asunto: QoS Charts [7:60710]
>Howard,
>
>We'd all like to see a comprehensive QOS chart.
>
>Thanks,
>
>Kym
They are Word tables in several different papers of mine. Just trying
to paste them into email text results in something unreadable.
Anyone good at
reformatting tables to ASCII, when table cells have multiple lines?
>
>
>
>>From: "Jonathan V Hays"
>>Reply-To: "Jonathan V Hays"
>>To:
>>Subject: RE: Small Minded QOS
>>Date: Wed, 8 Jan 2003 22:17:53 -0500
>>
>>Howard,
>>
>>I, for one, would be quite interested in such a comparative table.
>>
>>Thanks in advance.
>>
>>Jonathan
>>
>>-Original Message-
>>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
>>Howard C. Berkowitz
>>Sent: Wednesday, January 08, 2003 3:47 PM
>>To: [EMAIL PROTECTED]
>>Subject: Re: Small Minded QOS
>>
>>
>>At 7:39 PM + 1/8/03, Jennifer Bellucci wrote:
>>>Hello
>>>
>>>Can someone tell me where I can find every single QOS formula for all
>>>the QOS techniques in one document, prefer it if it was one page.
>>>Please do not send me the link to QOS on cco.
>>
>>Could you clarify what you mean by "formula"? When I hear that, I
>>think of the derivation of the algorithms, and, certainly for things
>>like RED and WFQ, these were published in scientific papers of many
>>pages and equations. Indeed, if you go into the literature, you'll
>>find that what we may call things are actually approximations ("true"
>>WFQ is a theoretical abstraction; what we call WFQ is more like
>>Self-Clocked WFQ).
>>
>>Now, if you are thinking of a comparative table of what the various
>>IOS QOS mechanisms do, that's something I might be able to dig up. .
>>.
>rd
>
>_
>STOP MORE SPAM with the new MSN 8 and get 2 months FREE*
>http://join.msn.com/?page=features/junkmail
>.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=60753&t=60710
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
QoS Charts [7:60710]
>Howard,
>
>We'd all like to see a comprehensive QOS chart.
>
>Thanks,
>
>Kym
They are Word tables in several different papers of mine. Just trying
to paste them into email text results in something unreadable.
Anyone good at
reformatting tables to ASCII, when table cells have multiple lines?
>
>
>
>>From: "Jonathan V Hays"
>>Reply-To: "Jonathan V Hays"
>>To:
>>Subject: RE: Small Minded QOS
>>Date: Wed, 8 Jan 2003 22:17:53 -0500
>>
>>Howard,
>>
>>I, for one, would be quite interested in such a comparative table.
>>
>>Thanks in advance.
>>
>>Jonathan
>>
>>-Original Message-
>>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
>>Howard C. Berkowitz
>>Sent: Wednesday, January 08, 2003 3:47 PM
>>To: [EMAIL PROTECTED]
>>Subject: Re: Small Minded QOS
>>
>>
>>At 7:39 PM + 1/8/03, Jennifer Bellucci wrote:
>>>Hello
>>>
>>>Can someone tell me where I can find every single QOS formula for all
>>>the QOS techniques in one document, prefer it if it was one page.
>>>Please do not send me the link to QOS on cco.
>>
>>Could you clarify what you mean by "formula"? When I hear that, I
>>think of the derivation of the algorithms, and, certainly for things
>>like RED and WFQ, these were published in scientific papers of many
>>pages and equations. Indeed, if you go into the literature, you'll
>>find that what we may call things are actually approximations ("true"
>>WFQ is a theoretical abstraction; what we call WFQ is more like
>>Self-Clocked WFQ).
>>
>>Now, if you are thinking of a comparative table of what the various
>>IOS QOS mechanisms do, that's something I might be able to dig up. .
>>.
>rd
>
>_
>STOP MORE SPAM with the new MSN 8 and get 2 months FREE*
>http://join.msn.com/?page=features/junkmail
>.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=60710&t=60710
--
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Great Cisco Resourse - Mcast, Qos, MPLS, etc.` [7:60081]
I wanted to share this resource with you: http://www.cisco.com/warp/public/732/Tech/ Perhaps I have been living under a CCO rock all of my life, but this was a great revelation to me. First, it's public. Second, it includes internal Cisco training materials. There have been several inquiries over the last few months for good CCIP info (esp. mcast and QoS). Well, from the above, follow the link to "Multicast" and then to "Learn About Cisco IOS Multicast" and finally to "IP Multicast Training Materials." You will find about 1,000 pages of training material on all things mcast. It appears to be at least, if not more, detailed than the CP book "Developing IP Multicast Networks, Vol I" that was discussed recently. I suspect the author was heavily involved with the development of the training material. I haven't looked into the QoS site much, but I suspect it is on the same level of detail. Since I've had a CCO login for some time, I've always used the non-public: http://www.cisco.com/cgi-bin/Support/PSP/index.pl?i=Technologies Frankly, I'm not sure that I don't prefer the public site. Sorry if this is old hat, but I was amazed. Now, on to that 1,000 pages sitting on the printer... Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60081&t=60081 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Good QoS and Mcast Book (something like that) [7:60043]
Vicuna, Mark wrote: > > Williamson is also coming out with vol II of that title.. > anyone heard > when? > > -Original Message- > From: s vermill [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, December 31, 2002 1:23 PM > To: [EMAIL PROTECTED] > Subject: RE: Good QoS and Mcast Book (something like that) > [7:60043] > > Don't know for sure. I expected Vol II to cover inter-domain mcast in depth. In July, a book entitled "Interdomain Multicast Solution Guide" was published. No author is listed on the CP website. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60069&t=60043 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Good QoS and Mcast Book (something like that) [7:60043]
Williamson is also coming out with vol II of that title.. anyone heard when? -Original Message- From: s vermill [mailto:[EMAIL PROTECTED]] Sent: Tuesday, December 31, 2002 1:23 PM To: [EMAIL PROTECTED] Subject: RE: Good QoS and Mcast Book (something like that) [7:60043] James Ramsay wrote: > > > Which books can peopel recommend for this CCIP exam. > > I only want to buy one QoS book and one MCast book - spent > enough with Cisco Press for one life time ;) > > Any recommendations? > > Many thanks > > James I can't recommend a good QoS book yet because I haven't found one (going to try my first Global Knowledge book soon so I'll get back to you). I don't recommend the Cisco Press book on the subject at all. As for mcast, I would strongly recommend the CP book "Developing IP Multicast Networks, Volume I" by Beau Williamson. I always think that good instructors make better authors than about anyone else. Mr. Williamson apparently teaches the internal Cisco course on the subject and is some kind of mcast advisor to the CTO or something along those lines. Reading the book is somewhat like sitting a course. The style is very conversational. The reason that I think instructors make great authors is that they have a pretty good idea of the problems you are going to have as a learner. They've seen the dumb looks and fielded all of the silly questions. This author clearly takes advantage of his background teaching the subject. Sorry for the delayed response but I wanted to finish the book before answering. There is some errata, so be sure to download the errata sheet at www.ciscopress.com. Regards, Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60049&t=60043 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Good QoS and Mcast Book (something like that) [7:60043]
James Ramsay wrote: > > > Which books can peopel recommend for this CCIP exam. > > I only want to buy one QoS book and one MCast book - spent > enough with Cisco Press for one life time ;) > > Any recommendations? > > Many thanks > > James I can't recommend a good QoS book yet because I haven't found one (going to try my first Global Knowledge book soon so I'll get back to you). I don't recommend the Cisco Press book on the subject at all. As for mcast, I would strongly recommend the CP book "Developing IP Multicast Networks, Volume I" by Beau Williamson. I always think that good instructors make better authors than about anyone else. Mr. Williamson apparently teaches the internal Cisco course on the subject and is some kind of mcast advisor to the CTO or something along those lines. Reading the book is somewhat like sitting a course. The style is very conversational. The reason that I think instructors make great authors is that they have a pretty good idea of the problems you are going to have as a learner. They've seen the dumb looks and fielded all of the silly questions. This author clearly takes advantage of his background teaching the subject. Sorry for the delayed response but I wanted to finish the book before answering. There is some errata, so be sure to download the errata sheet at www.ciscopress.com. Regards, Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60043&t=60043 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Which Books for 640-905 MCAST+QOS [7:58633]
James Ramsay wrote: > > > Which books can peopel recommend for this CCIP exam. > > I only want to buy one QoS book and one MCast book - spent > enough with Cisco Press for one life time ;) > > Any recommendations? > > Many thanks > > James I can't recommend a good QoS book yet because I haven't found one (going to try my first Global Knowledge book soon so I'll get back to you). I don't recommend the Cisco Press book on the subject at all. As for mcast, I would strongly recommend the CP book "Developing IP Multicast Networks, Volume I" by Beau Williamson. I always think that good instructors make better authors than about anyone else. Mr. Williamson apparently teaches the internal Cisco course on the subject and is some kind of mcast advisor to the CTO or something along those lines. Reading the book is somewhat like sitting a course. The style is very conversational. The reason that I think instructors make great authors is that they have a pretty good idea of the problems you are going to have as a learner. They've seen the dumb looks and fielded all of the silly questions. This author clearly takes advantage of his background teaching the subject. Sorry for the delayed response but I wanted to finish the book before answering. There is some errata, so be sure to download the errata sheet at www.ciscopress.com. Regards, Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=60042&t=58633 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Cat 3550 + modular QoS CLI - does it work? [7:59926]
""neal r"" wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > I've been looking at the pages on the Cat 3550 and I can't tell if > this thing will handle the entire range of possibilites that the modular > QoS CLI provides, or if there are some hardware limits on numbers of > queues, etc, etc. it can get a bit confusing at times, but in general, all Cisco euipment eventually has hardware (and software)limitations that eventually come into play. Yes MQC is supported. So far as I can tell, all functionality is supported with the obvious exceptions. No frame relay traffic shaping, for example. The 3550 is a fixed configuration box. No upping the RAM or flash. What you see is what you get. You can use one of the four ( five now? This mayu have changed with the new IOS ) SDM's. Switch_48#sh sdm prefer The current template is the default template. The selected template optimizes the resources in the switch to support this level of features for 8 routed interfaces and 1K VLANs. number of unicast mac addresses: 5K number of igmp groups: 1K number of qos aces:1K number of security aces: 1K number of unicast routes: 8K number of multicast routes:1K Switch_48# Switch_48(config)#sdm ? prefer Config TCAM and ASIC RAM size. Warning: need to reset switch for configuration to take effect. Switch_48(config)#sdm prefer ? access multicast and qos/acl bias, drop unicast extended-match Using extended match for unicast routing routing unicast bias, drop qos/acl vlanvlan bias, drop routing There is a section in the documentation that covers this. As far as queues and things, gotta read the docs carefully. IIRC, it is not all in the same place, so you have to dig around a little. contact me off line. perhaps something can be worked out. Chuck > > >Is there someone with a 3550 in a lab environment that would be > willing to let me log in and check it out before I shell out the $$$ to > get my hands on one of these things? Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=59928&t=59926 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Cat 3550 + modular QoS CLI - does it work? [7:59926]
I've been looking at the pages on the Cat 3550 and I can't tell if this thing will handle the entire range of possibilites that the modular QoS CLI provides, or if there are some hardware limits on numbers of queues, etc, etc. Is there someone with a 3550 in a lab environment that would be willing to let me log in and check it out before I shell out the $$$ to get my hands on one of these things? Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=59926&t=59926 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
