TCP Performance

New Message	Reply	About this list	Date view	Thread view	Subject view	Author view	Attachment view

From: Ethan Solomita (ethan@aristophanes.engr.sgi.com)
Date: 07/30/98-07:07:55 PM Z

Next message: Spencer Shepler: "NFSv4 meeting times for Chicago"
Previous message: spencer.shepler: "NFSv4 working group status (approved)"

Date: Thu, 30 Jul 1998 17:07:55 -0700
Message-Id: <199807310007.RAA02603@aristophanes.engr.sgi.com>
From: Ethan Solomita <ethan@aristophanes.engr.sgi.com>
Subject: TCP Performance

	I don't mean to distract people from the pressing issue
of timestamp representation, but I was hoping to ask a somewhat
historical question about TCP performance.

	Have there been efforts to permit TCP rsizes and wsizes
of larger than 48K, and if so have they helped the performance
disparity between UDP and TCP? Are these issues of I/O sizes
addressed or changed in v4? For all I know this could either have
been widely tried, or have never been tried.

	Thanks!
	-- Ethan

Next message: Spencer Shepler: "NFSv4 meeting times for Chicago"
Previous message: spencer.shepler: "NFSv4 working group status (approved)"

New Message	Reply	About this list	Date view	Thread view	Subject view	Author view	Attachment view

This archive was generated by hypermail 2.1.2 : 03/04/05-01:46:07 AM Z CST