Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!apple!metaphor!xymox!philf
From: philf@xymox.metaphor.com (Phil Fernandez)
Newsgroups: comp.protocols.tcp-ip
Subject: Re: ether collision backoff
Keywords: non-standard
Message-ID: <840@metaphor.Metaphor.COM>
Date: 13 Oct 89 06:01:36 GMT
References: <42686@sgi.sgi.com> <14015@well.UUCP>
Sender: news@metaphor.Metaphor.COM
Reply-To: philf@xymox.metaphor.com (Phil Fernandez)
Organization: Metaphor Computer Systems, Mountain View, CA
Lines: 28

In article <14015@well.UUCP> nagle@well.UUCP (John Nagle) writes:
> (...re. collision backoff pros and cons)
>      General advice: DON'T DO THIS unless you can establish by both
>mathematical calculation and tests on large networks that you are not
>introducing instability...

This general area has been a topic of much conversation around
Metaphor lately, where we're looking to some excessive collision
problems.  Let me ask:

Just how does one change the Ethernet backoff algorithm when it's
implemented completely within an IC such as the AMD Lance?  Sun, for
example, uses the Lance on their desktop workstations, and I don't
know how they might have changed the backoff scheme even if they
wanted to.

How often does one even get the opportunity to reimplement this
algorithm these days?

pmf


+-----------------------------+----------------------------------------------+
| Phil Fernandez              |             philf@metaphor.com               |
|                             |     ...!{apple|decwrl}!metaphor!philf        |
| Metaphor Computer Systems   |"Does the body rule the mind, or does the mind|
| Mountain View, CA           | rule the body?  I dunno..." - Morrissey      |
+-----------------------------+----------------------------------------------+