Not to shift the subject, but if you need the consistent latency and
reliability for multimedia, why not go with ATM? The bandwidth available
is better than Token ring and the cost isn't *that* much more.
In any case, SNA over Ethernet is no problem in a well-designed
network. As Priscilla stated, IBM has the market for most of the high-end
hardware that utilizes SNA, IBM likes(d) token ring, and change is a hard
thing to come by in the world of Big Blue.
Just my $0.02...
Craig
At 04:20 PM 10/12/2000 -0400, you wrote:
> > At 08:19 AM 10/12/00, Alldread AK2 Robert J wrote:
> >>I am just curious as to why SNA still runs on tokenring today.
> >
> > IBM got their foot in the door at many large companies and then closed the
> > door behind them. The end result is that many companies deployed all
> > IBM-developed technologies. If a company uses their network for
> > mission-critical applications, it takes a long time for the network to
> > evolve, since changes risk bringing the network down.
>
>It's actually more than that though. Token Ring and FDDI are token passing
>in nature and can have predictable response times. Ethernet is bursty by
>nature and, although good for most everybody needs, still has issues in the
>QoS or what I call Best Effort QoS, for what Token Ring already offers,
>which is consistent, predictable network response times.
>With SNA, that is very important.
>
>I have seen companies go through a lot of planning into putting Video
>Conferencing using H.323 over IP and it's still a headache. Where in the
>Token Ring world of 100Mb Token Ring ( yes it exists ), this is a lot easier
>as it already offers QoS-ing ( made that word up <g> ) and has done so for
>years.
>
> >> Is there any
> >>reason that it cannot just hook right into an ethernet network??
>
>-->You can use Ethernet and although, a little harder to setup, but works OK
>provided you don't have an over saturated network and your support staff
>knows how it works. It still rears up every now and then.
>
> > SNA can run on Ethernet. IBM and other vendors support this. It's actually
> > somewhat common.
> >
> >> I have
> >>read a few white papers on SNA, and I assume that because tokenring was the
> >>major LAN media back in the day, and because SNA uses RIF's to determine
> >>paths to other hosts, that SNA was built requiring the use of RIF's. Is
> >>this correct??
> > Historically this is not correct. SNA has its own complex path
> > determination methods that were used on large "internetworks" long before
> > Token Ring and source-route bridging were ever invented. Long before LANs
> > were invented, come to think of it. SNA shipped in 1974, (though it didn't
> > catch on right away.) Token Ring's birth date is more like 1984.
> >
> > The Internetworking Technologies Handbook does a pretty good job with SNA.
> > SNA is surprisingly still common, as is Token Ring. (At least in Southern
> > Oregon where I live!? &;-)
>
>-->I do a lot of work for a Movie production company that uses Token Ring
>because of all of the Multimedia they do. We tried a lot of different
>Ethernet solutions but just couldn't get it to perform well.
>For most people, Ethernet works fine as most of the usual traffic is E-Mail,
>Web, file and print, which aren't latency sensitive.
>
>However, to multimedia rendering, video conferencing, Voice, SNA, etc, that
>are very latency sensitive, it makes a big difference.
>
>Sorry for the Token Ring speech. <g> and no, I don't sell it.
>
>Madge has a pretty decent paper on it:
>
>http://www.madge.com/Connect/Downloads/Documents/content.asp?Article=536&Sub
>Area=
>
>IMHO, it is a lot better than Ethernet but the price killed it.
>Sometimes, you do get what you paid for though.... ;-)
>
>It's funny, I have seen companies put a lot of money into QoS strategies,
>that, had they just done Token Ring in the first place, would have cost them
>less and had less support issues in the long run. <g>
>
>Scotty
>
>
>
> > Priscilla
> >
> >
> >>thanx,
> >>
> >>skin-e
> >>
> >>_________________________________
> >>FAQ, list archives, and subscription info:
> >>http://www.groupstudy.com/list/cisco.html
> >>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> >
> >
> > ________________________
> >
> > Priscilla Oppenheimer
> > http://www.priscilla.com
> >
> > _________________________________
> > FAQ, list archives, and subscription info:
> > http://www.groupstudy.com/list/cisco.html
> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
>
>----------------------------------------------------------
>Scott Nelson - Network Engineer
>Wash DC +1202-270-8968 & +1202-352-6646
>Los Angeles +1310-367-6646
>mailto:[EMAIL PROTECTED]
>http://www.bnmnetworks.net
>
>PGP Public Key:
>http://home.earthlink.net/~scottnelson/keys/srnbnm.txt
>----------------------------------------------------------
>
>"The better the customer service, the sooner you get to speak
>with someone who can't help you."
>----------------------------------------------
>
>_________________________________
>FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
