home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!hal.com!decwrl!access.usask.ca!skorpio!reidk
- From: reidk@skorpio.usask.ca (Karen Reid)
- Newsgroups: comp.dcom.cell-relay
- Subject: Acceptable delays
- Message-ID: <1992Jul29.164441.29276@access.usask.ca>
- Date: 29 Jul 92 16:44:41 GMT
- Sender: reidk@skorpio (Karen Reid)
- Organization: University of Saskatchewan, Saskatoon, Canada
- Lines: 17
- Nntp-Posting-Host: skorpio.usask.ca
-
- Can anyone point me towards references that have a good indication
- of the end-to-end delays that various services on an ATM (BISDN) network
- can tolerate (particularly interactive video and voice)?
- What proportion of the delay will beattributed to building packets,
- queueing delay, and propagation delay?
-
- My question stems from wondering how big buffers at internal
- switching nodes can be before the delay incurred by cells arriving
- at an almost full buffer exceeds some quality of service limit. I've
- seen figures for the delay at a switch ranging from 10's of microseconds
- to 100's of milliseconds.
-
- Thanks,
-
- Karen Reid
- reidk@cs.usask.ca
- reidk@skorpio.usask.ca
-