home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!network.ucsd.edu!sdcrsi!xlnt!xlnt.com
- From: rasp@xlnt.com (Ron Perloff)
- Newsgroups: comp.dcom.cell-relay
- Subject: Re: >>>>Future of IP routers
- Message-ID: <1179@xlnt.COM>
- Date: 1 Sep 92 16:43:37 GMT
- References: <Btv855.11w@usenet.ucs.indiana.edu> <p9k3qsc@rhyolite.wpd.sgi.com>
- Sender: postmaster@xlnt.COM
- Organization: xdi
- Lines: 25
-
- In article <p9k3qsc@rhyolite.wpd.sgi.com> vjs@rhyolite.wpd.sgi.com (Vernon
- Schryver) writes:
- >That's a lot of confusing words to describe something that is obvious
- >with a few diagrams. It is a surprising notion, that every TRT
- >expiration implies a "wasted, deadhead" rotation of the token around
- >the ring.
-
- Vernon is absolutely right, including the part about drawing some diagrams
- to convince yourself.
-
- A very convenient way to model this is to assume a 'dummy' node
- in the ring that transmits a frame of length equal to the latency on every
- rotation of the token. This helps visualize the performance implications
- of some of the configuration parameters.
-
- E.g., When you lower TTRT, you force more token rotations in any given
- period of time and, therefore, the 'latency node' burns more bandwidth.
- In a ring with a lot of latency, the amount of lost bandwidth may surprise you.
-
- Ronald S. Perloff rasp@xlnt.com
- XLNT Designs, Inc. (XDI)
- 15050 Avenue of Science
- Suite 106
- San Diego, CA 92128
- 619 487 9320 - FAX: 619 487 9768
-