RE: Intenet Speed Record: What am I missing? [7:64767]
On Sat, 8 Mar 2003, Mossburg, Geoff (MAN-Corporate) wrote: > I went to Justin's link and followed a few more and found this: > http://lsr.internet2.edu/ > That link not only has the original text that the news article quoted, but > it explains that this is a contest and gives the rules for the contest. I > will leave it to the greater minds in the group to interpret the rules... Wow, I was at SC2002 and I didn't know this was occuring. It should tell you how little publicity was given to the event. But of course for the bandwidth freaks out there (like me) they had it all! I saw a few transatlantic OC-192s, some OC-48s and at the minimum an OC-12. Cisco had a booth playing a movie over its own OC-48 from Germany. And if I remember correctly the video was uncompressed and just shipped across for viewing. Check out http://scinet.supercomp.org/ for pictures of the setup and the primary network layout. (Doesn't include vendor demos like Marconi's impressive ATM net and Cisco's transatlantic demo) Andrew --- http://www.andrewsworld.net/ ICQ: 2895251 Cisco Certified Network Associate "Learn from the mistakes of others. You won't live long enough to make all of them yourself." Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64809&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
Here is the SLAC webpage with more details: http://www-iepm.slac.stanford.edu/lsr/ Justin -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Mossburg, Geoff (MAN-Corporate) Sent: Friday, March 07, 2003 7:24 PM To: [EMAIL PROTECTED] Subject: RE: Intenet Speed Record: What am I missing? [7:64767] John, Would you mind posting a link to the article? I'm curious. Thanks! Geoff Mossburg -Original Message- From: John Neiberger [mailto:[EMAIL PROTECTED] Sent: Friday, March 07, 2003 3:57 PM To: [EMAIL PROTECTED] Subject: Intenet Speed Record: What am I missing? [7:64767] Here's a quote from something I just saw in the news: "Scientists at the Stanford Linear Accelerator Center used fiber-optic cables to transfer 6.7 gigabytes of data -- the equivalent of two DVD movies -- across 6,800 miles in less than a minute. Pushing the tech envelope The team was able to transfer uncompressed data at 923 megabits per second for 58 seconds from Sunnyvale, California, to Amsterdam, Netherlands. That's about 3,500 times faster than a typical Internet broadband connection. " Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 times faster and they're fairly common. What's the big deal about 923 Mbps? I realize that I must be missing something very obvious here but I don't understand the milestone they're claiming to have passed. Admittedly, I'm about to fall asleep in my chair but that's par for the course with me. :-) So, what's the big deal? In a world of OC-192 and up, why is <1Gbps so earth shattering? John Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64808&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
Hey all, I went to Justin's link and followed a few more and found this: http://lsr.internet2.edu/ That link not only has the original text that the news article quoted, but it explains that this is a contest and gives the rules for the contest. I will leave it to the greater minds in the group to interpret the rules... :-) Geoff Mossburg -Original Message- From: Justin Kinney [mailto:[EMAIL PROTECTED] Sent: Friday, March 07, 2003 8:09 PM To: [EMAIL PROTECTED] Subject: RE: Intenet Speed Record: What am I missing? [7:64767] Actually, the application they used was straight tcp. You can get it at http://www.cnd.gatech.edu/rapid/. Again, I think the message is that this was done over a switched network, and they did this over a 196msec, 6 hop path. Justin -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, March 07, 2003 5:31 PM To: [EMAIL PROTECTED] Subject: RE: Intenet Speed Record: What am I missing? [7:64767] s vermill wrote: > > John Neiberger wrote: What you're missing is the cluelessness of news reporters and the public maybe... > > > > Here's a quote from something I just saw in the news: > > > > "Scientists at the Stanford Linear Accelerator Center used > > fiber-optic > > cables to transfer 6.7 gigabytes of data -- the equivalent of > > two DVD > > movies -- across 6,800 miles in less than a minute. Well, nobody actually watched the movies! Now a real feat would be to send it over in that time and have someone watching it. > > > > Pushing the tech envelope > > The team was able to transfer uncompressed data at 923 > megabits > > per > > second for 58 seconds from Sunnyvale, California, to > Amsterdam, > > Netherlands. That's about 3,500 times faster than a typical > > Internet > > broadband connection. " Whoopee! > > > > Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 > > times faster > > and they're fairly common. What's the big deal about 923 > > Mbps? I > > realize that I must be missing something very obvious here but > > I don't > > understand the milestone they're claiming to have passed. > > > > Admittedly, I'm about to fall asleep in my chair but that's > par > > for the > > course with me. :-) > > > > So, what's the big deal? In a world of OC-192 and up, why is > > > earth shattering? > > > > John > > > > John, > > It clearly isn't a bandwidth record in terms of bps. I'm aware > of as many as a few hundred OC-192s being "DWDMed" onto a > single fiber. I suspect this has to do with the > control/reliability mechanisms associated with the file > transfer. I suspect that they ignored the control issues. :-) If they used FTP/TCP/IP, they probably started measuring after the control and data 3-way handshakes and set the TCP window as huge as possible?? They probably rolled their own transfer mechanism, I would guess, but maybe not. TCP could be pretty good for this Priscilla > I read the same article. Nothing was said about the > protocols involved (it was packaged for mass consuption > evidentally). It was likely FTP/TCP/IP or something along > those lines (although a negative ack approach would likely be > the most effective). I noticed that there were several > intermediate hops. Don't know if that was a ploy to reduce the > delay portion of the bandwidth*delay product between any two > points or if there were other reasons. > > Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64807&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Intenet Speed Record: What am I missing? [7:64767]
Sure, here you go: http://news.excite.com/tech/article/id/160874|technology|03-07-2003::19:44|r euters.html ""Mossburg, Geoff (MAN-Corporate)"" wrote in message news:[EMAIL PROTECTED] > John, > Would you mind posting a link to the article? I'm curious. > Thanks! > Geoff Mossburg > > -Original Message- > From: John Neiberger [mailto:[EMAIL PROTECTED] > Sent: Friday, March 07, 2003 3:57 PM > To: [EMAIL PROTECTED] > Subject: Intenet Speed Record: What am I missing? [7:64767] > > > Here's a quote from something I just saw in the news: > > "Scientists at the Stanford Linear Accelerator Center used fiber-optic > cables to transfer 6.7 gigabytes of data -- the equivalent of two DVD > movies -- across 6,800 miles in less than a minute. > > Pushing the tech envelope > The team was able to transfer uncompressed data at 923 megabits per > second for 58 seconds from Sunnyvale, California, to Amsterdam, > Netherlands. That's about 3,500 times faster than a typical Internet > broadband connection. " > > Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 times faster > and they're fairly common. What's the big deal about 923 Mbps? I > realize that I must be missing something very obvious here but I don't > understand the milestone they're claiming to have passed. > > Admittedly, I'm about to fall asleep in my chair but that's par for the > course with me. :-) > > So, what's the big deal? In a world of OC-192 and up, why is earth shattering? > > John Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64806&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
John, Would you mind posting a link to the article? I'm curious. Thanks! Geoff Mossburg -Original Message- From: John Neiberger [mailto:[EMAIL PROTECTED] Sent: Friday, March 07, 2003 3:57 PM To: [EMAIL PROTECTED] Subject: Intenet Speed Record: What am I missing? [7:64767] Here's a quote from something I just saw in the news: "Scientists at the Stanford Linear Accelerator Center used fiber-optic cables to transfer 6.7 gigabytes of data -- the equivalent of two DVD movies -- across 6,800 miles in less than a minute. Pushing the tech envelope The team was able to transfer uncompressed data at 923 megabits per second for 58 seconds from Sunnyvale, California, to Amsterdam, Netherlands. That's about 3,500 times faster than a typical Internet broadband connection. " Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 times faster and they're fairly common. What's the big deal about 923 Mbps? I realize that I must be missing something very obvious here but I don't understand the milestone they're claiming to have passed. Admittedly, I'm about to fall asleep in my chair but that's par for the course with me. :-) So, what's the big deal? In a world of OC-192 and up, why is <1Gbps so earth shattering? John Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64804&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
Actually, the application they used was straight tcp. You can get it at http://www.cnd.gatech.edu/rapid/. Again, I think the message is that this was done over a switched network, and they did this over a 196msec, 6 hop path. Justin -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, March 07, 2003 5:31 PM To: [EMAIL PROTECTED] Subject: RE: Intenet Speed Record: What am I missing? [7:64767] s vermill wrote: > > John Neiberger wrote: What you're missing is the cluelessness of news reporters and the public maybe... > > > > Here's a quote from something I just saw in the news: > > > > "Scientists at the Stanford Linear Accelerator Center used > > fiber-optic > > cables to transfer 6.7 gigabytes of data -- the equivalent of > > two DVD > > movies -- across 6,800 miles in less than a minute. Well, nobody actually watched the movies! Now a real feat would be to send it over in that time and have someone watching it. > > > > Pushing the tech envelope > > The team was able to transfer uncompressed data at 923 > megabits > > per > > second for 58 seconds from Sunnyvale, California, to > Amsterdam, > > Netherlands. That's about 3,500 times faster than a typical > > Internet > > broadband connection. " Whoopee! > > > > Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 > > times faster > > and they're fairly common. What's the big deal about 923 > > Mbps? I > > realize that I must be missing something very obvious here but > > I don't > > understand the milestone they're claiming to have passed. > > > > Admittedly, I'm about to fall asleep in my chair but that's > par > > for the > > course with me. :-) > > > > So, what's the big deal? In a world of OC-192 and up, why is > > > earth shattering? > > > > John > > > > John, > > It clearly isn't a bandwidth record in terms of bps. I'm aware > of as many as a few hundred OC-192s being "DWDMed" onto a > single fiber. I suspect this has to do with the > control/reliability mechanisms associated with the file > transfer. I suspect that they ignored the control issues. :-) If they used FTP/TCP/IP, they probably started measuring after the control and data 3-way handshakes and set the TCP window as huge as possible?? They probably rolled their own transfer mechanism, I would guess, but maybe not. TCP could be pretty good for this Priscilla > I read the same article. Nothing was said about the > protocols involved (it was packaged for mass consuption > evidentally). It was likely FTP/TCP/IP or something along > those lines (although a negative ack approach would likely be > the most effective). I noticed that there were several > intermediate hops. Don't know if that was a ploy to reduce the > delay portion of the bandwidth*delay product between any two > points or if there were other reasons. > > Scott Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64803&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
On Fri, 7 Mar 2003, Priscilla Oppenheimer wrote: > I suspect that they ignored the control issues. :-) If they used FTP/TCP/IP, > they probably started measuring after the control and data 3-way handshakes > and set the TCP window as huge as possible?? > > They probably rolled their own transfer mechanism, I would guess, but maybe > not. TCP could be pretty good for this Ok, based on what I've heard from NANOG they used a portion of the I2 and created a PVC across it for only this single test. They then probably increased the TCP Window size to make this work. I'm not saying this is all factual, I'm just recalling what I've read there. Andrew --- http://www.andrewsworld.net/ ICQ: 2895251 Cisco Certified Network Associate "Learn from the mistakes of others. You won't live long enough to make all of them yourself." Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64801&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
s vermill wrote: > > John Neiberger wrote: What you're missing is the cluelessness of news reporters and the public maybe... > > > > Here's a quote from something I just saw in the news: > > > > "Scientists at the Stanford Linear Accelerator Center used > > fiber-optic > > cables to transfer 6.7 gigabytes of data -- the equivalent of > > two DVD > > movies -- across 6,800 miles in less than a minute. Well, nobody actually watched the movies! Now a real feat would be to send it over in that time and have someone watching it. > > > > Pushing the tech envelope > > The team was able to transfer uncompressed data at 923 > megabits > > per > > second for 58 seconds from Sunnyvale, California, to > Amsterdam, > > Netherlands. That's about 3,500 times faster than a typical > > Internet > > broadband connection. " Whoopee! > > > > Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 > > times faster > > and they're fairly common. What's the big deal about 923 > > Mbps? I > > realize that I must be missing something very obvious here but > > I don't > > understand the milestone they're claiming to have passed. > > > > Admittedly, I'm about to fall asleep in my chair but that's > par > > for the > > course with me. :-) > > > > So, what's the big deal? In a world of OC-192 and up, why is > > > earth shattering? > > > > John > > > > John, > > It clearly isn't a bandwidth record in terms of bps. I'm aware > of as many as a few hundred OC-192s being "DWDMed" onto a > single fiber. I suspect this has to do with the > control/reliability mechanisms associated with the file > transfer. I suspect that they ignored the control issues. :-) If they used FTP/TCP/IP, they probably started measuring after the control and data 3-way handshakes and set the TCP window as huge as possible?? They probably rolled their own transfer mechanism, I would guess, but maybe not. TCP could be pretty good for this Priscilla > I read the same article. Nothing was said about the > protocols involved (it was packaged for mass consuption > evidentally). It was likely FTP/TCP/IP or something along > those lines (although a negative ack approach would likely be > the most effective). I noticed that there were several > intermediate hops. Don't know if that was a ploy to reduce the > delay portion of the bandwidth*delay product between any two > points or if there were other reasons. > > Scott > > > > > Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64792&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Intenet Speed Record: What am I missing? [7:64767]
>Big deal because it's 6800 miles for a full minute. An OC-48 can't push data >that fast that far or that long. > I don't understand...there are OC-192s (WDM, multiples on a fiber) in operational use on quite a number of undersea cables. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64784&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Intenet Speed Record: What am I missing? [7:64767]
At 8:56 PM + 3/7/03, John Neiberger wrote: >Here's a quote from something I just saw in the news: > >"Scientists at the Stanford Linear Accelerator Center used fiber-optic >cables to transfer 6.7 gigabytes of data -- the equivalent of two DVD >movies -- across 6,800 miles in less than a minute. > >Pushing the tech envelope >The team was able to transfer uncompressed data at 923 megabits per >second for 58 seconds from Sunnyvale, California, to Amsterdam, >Netherlands. That's about 3,500 times faster than a typical Internet >broadband connection. " > >Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 times faster >and they're fairly common. What's the big deal about 923 Mbps? I >realize that I must be missing something very obvious here but I don't >understand the milestone they're claiming to have passed. > >Admittedly, I'm about to fall asleep in my chair but that's par for the >course with me. :-) > >So, what's the big deal? In a world of OC-192 and up, why is earth shattering? > I'm baffled by it as well. I suppose it might be a fair accomplishment for a host to sink that much traffic, but the network part is straightforward -- it's just OC-192, and any number of routers can handle multiple OC-192. OC-768 has been demonstrated, and DWDM of many OC-192s isn't a big deal -- the big deal is how many lambdas you can get on a fiber. I've seen lab experiments with hundreds. The optical physics people tell me that you're starting to approach fundamental chip technologies much over OC-768, although I have talked to people that think they can double it to 80 Gbps. Lots of parallel OC-192s, however, may be just as useful. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64783&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
A reply from another list: "I'm going to launch a couple DAT tapes across the parking lot with a spud gun and see if I can achieve 923 Mb/s!" Classic!!! Symon -Original Message- From: s vermill [mailto:[EMAIL PROTECTED] Sent: 07 March 2003 21:11 To: [EMAIL PROTECTED] Subject: Re: Intenet Speed Record: What am I missing? [7:64767] John Hutchison wrote: > > Big deal because it's 6800 miles for a full minute. An OC-48 can't > push data that fast that far or that long. > > Yes it can. SONET is 60x60x24x7x52 at whatever rate the optics are running at. Distance is not an issue. Upper-layer protocols (such as TCP) that might make use of a long-delay OC-48 might have an issue, but it has nothing to do with the OC-48 itself. I deal with international SONET from time to time. We even put SONET over satellite sometimes (although one could argue it isn't truly SONET at that point, the STS frame structure is maintained). = This email has been content filtered and subject to spam filtering. If you consider this email is unsolicited please forward the email to [EMAIL PROTECTED] and request that the sender's domain be blocked from sending any further emails. = Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64780&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Intenet Speed Record: What am I missing? [7:64767]
John Hutchison wrote: > > Big deal because it's 6800 miles for a full minute. An OC-48 > can't push data > that fast that far or that long. > > Yes it can. SONET is 60x60x24x7x52 at whatever rate the optics are running at. Distance is not an issue. Upper-layer protocols (such as TCP) that might make use of a long-delay OC-48 might have an issue, but it has nothing to do with the OC-48 itself. I deal with international SONET from time to time. We even put SONET over satellite sometimes (although one could argue it isn't truly SONET at that point, the STS frame structure is maintained). Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64772&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: Intenet Speed Record: What am I missing? [7:64767]
John Neiberger wrote: > > Here's a quote from something I just saw in the news: > > "Scientists at the Stanford Linear Accelerator Center used > fiber-optic > cables to transfer 6.7 gigabytes of data -- the equivalent of > two DVD > movies -- across 6,800 miles in less than a minute. > > Pushing the tech envelope > The team was able to transfer uncompressed data at 923 megabits > per > second for 58 seconds from Sunnyvale, California, to Amsterdam, > Netherlands. That's about 3,500 times faster than a typical > Internet > broadband connection. " > > Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 > times faster > and they're fairly common. What's the big deal about 923 > Mbps? I > realize that I must be missing something very obvious here but > I don't > understand the milestone they're claiming to have passed. > > Admittedly, I'm about to fall asleep in my chair but that's par > for the > course with me. :-) > > So, what's the big deal? In a world of OC-192 and up, why is > earth shattering? > > John > John, It clearly isn't a bandwidth record in terms of bps. I'm aware of as many as a few hundred OC-192s being "DWDMed" onto a single fiber. I suspect this has to do with the control/reliability mechanisms associated with the file transfer. I read the same article. Nothing was said about the protocols involved (it was packaged for mass consuption evidentally). It was likely FTP/TCP/IP or something along those lines (although a negative ack approach would likely be the most effective). I noticed that there were several intermediate hops. Don't know if that was a ploy to reduce the delay portion of the bandwidth*delay product between any two points or if there were other reasons. Scott > Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64771&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Re: Intenet Speed Record: What am I missing? [7:64767]
Big deal because it's 6800 miles for a full minute. An OC-48 can't push data that fast that far or that long. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64769&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Intenet Speed Record: What am I missing? [7:64767]
Here's a quote from something I just saw in the news: "Scientists at the Stanford Linear Accelerator Center used fiber-optic cables to transfer 6.7 gigabytes of data -- the equivalent of two DVD movies -- across 6,800 miles in less than a minute. Pushing the tech envelope The team was able to transfer uncompressed data at 923 megabits per second for 58 seconds from Sunnyvale, California, to Amsterdam, Netherlands. That's about 3,500 times faster than a typical Internet broadband connection. " Okay, 923 Mbps is a speed record? An OC-48 is roughly 2.6 times faster and they're fairly common. What's the big deal about 923 Mbps? I realize that I must be missing something very obvious here but I don't understand the milestone they're claiming to have passed. Admittedly, I'm about to fall asleep in my chair but that's par for the course with me. :-) So, what's the big deal? In a world of OC-192 and up, why is <1Gbps so earth shattering? John Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64767&t=64767 -- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]