home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.cell-relay
- Path: sparky!uunet!munnari.oz.au!bunyip.cc.uq.oz.au!miw
- From: miw@cc.uq.oz.au (Mark I. Williams)
- Subject: Re: >>>>Question
- Message-ID: <miw.711958314@cc.uq.oz.au>
- Sender: news@bunyip.cc.uq.oz.au (USENET News System)
- Organization: Prentice Centre, University of Queensland
- References: <1992Jul18.230550.3046@sics.se> <BrM0FA.194@ux1.cso.uiuc.edu>
- Date: Fri, 24 Jul 1992 06:11:54 GMT
- Lines: 24
-
- gardner@ux1.cso.uiuc.edu (Mike Gardner) writes:
-
- >What is even worse is that ATM will possibly throw away more packets due to
- >congestion than errors. Also the 10**8 error rate is for each link. Cross
- >country ATM could cross a dozen or more links, each with possible congestion
- >problems. I wouldn't want to run a ack-less protocol across that!
-
- (1) The 10**-8 error ratio *includes* losses due to congestion. This is
- why there is so much effort being put into ATM resource
- management/control.
-
- (2) The 10**-8 error ratio is defined to be end-end. An ATM link should
- have an error ratio more like 10**-12, and a switch should have losses
- something like 10**-10.
-
- cheers,
-
-
-
- --
- Mark Williams The University of Queensland miw@cc.uq.edu.au
- +61 7 36 54012 (w) Prentice Centre
- +61 7 36 54477 (fax) Qld 4072 Australia
- Nullum magnum ingenium sine mixtura dementiae fuit. -- Seneca
-