home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!yale!gumby!wupost!news.utdallas.edu!corpgate!bnrgate!nrcnet0!cunews!revcan!software.mitel.com!grayt
- From: grayt@Software.Mitel.COM (Tom Gray)
- Newsgroups: comp.dcom.cell-relay
- Subject: Re: > Cell SIze
- Message-ID: <12387@pinard>
- Date: 27 Jul 92 12:39:38 GMT
- References: <1992Jul26.023447.21173@sics.se>
- Organization: Mitel. Kanata (Ontario). Canada.
- Lines: 34
-
- In article <1992Jul26.023447.21173@sics.se> craig@sics.se (Craig Partridge) writes:
- >|> >Note that if you were trying to compromise on these various tradeoffs,
- >|> >you'd probably pick a size that balance interrupt costs and bandwidth.
- >|> >My guess (based on what I've seen of predicted processor performance
- >|> >and estimated packet sizes) is that you'd choose a size of about 1,000
- >|> >bits (c. 128 bytes), which was one of the original proposals to CCITT.
- >|>
- >|> Yes, but...but.... This analysis considers data traffic only! Remember
- >|> that ATM will be expected to provide other services as well. With a cell
- >|> payload of 128 bytes echo cancellation will be even more difficult than
- >|> with 64 bytes.
- >
- >I was not focussing on data traffic. I was focussing on the future,
- >and transmission speeds in 100s of Mbits (yes to the home -- we can run
- >100 Mbit/sec over twisted pair...). At that point, small cells are
- >silly and serialization times are not a problem.
-
- The services are indifferent to the speed of the serial link which
- is carrying them. It will still take 8ms to fill a 128 byte
- packet with speech whether the serial stream is running at
- 64kb/s or 64Mb/s. The problem of echo canellation (and
- its other associated problems such as delay jitter) will
- still remain.
-
- Voice and video service will be the bulk of the traffic
- on the networks of the future. They have specific requirements
- which have prevented current data networks from carrying them
- in any practical situation. Has anyone ever made a voice call
- using TCP/IP?. These requirements will not change and so
- the network standards of tomorrow must adapt to them.
-
-
- --
- i.sinature
-