From owner-freebsd-bugs Thu May 30 11:13:57 1996 Return-Path: owner-bugs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id LAA05471 for bugs-outgoing; Thu, 30 May 1996 11:13:57 -0700 (PDT) Received: from halloran-eldar.lcs.mit.edu (halloran-eldar.lcs.mit.edu [18.26.0.159]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id LAA05466 for ; Thu, 30 May 1996 11:13:55 -0700 (PDT) Received: by halloran-eldar.lcs.mit.edu; (5.65/1.1.8.2/19Aug95-0530PM) id AA03302; Thu, 30 May 1996 14:13:52 -0400 Date: Thu, 30 May 1996 14:13:52 -0400 From: Garrett Wollman Message-Id: <9605301813.AA03302@halloran-eldar.lcs.mit.edu> To: chen@ipsilon.com Cc: bugs@freebsd.org Subject: a request for TCP enhancement In-Reply-To: <31ADDA8D.7AB8@ipsilon.com> References: <31ADDA8D.7AB8@ipsilon.com> Sender: owner-bugs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk < said: > In FreeBSD, the ratio of the total number of ack's (including window > updates) to the incoming number of TCP packets is about 1 to 2.1. That > is, under heavy traffic, FreeBSD TCP sends out 1 ack or window update > when it receives, on the average, 2.1 packets. In Solaris 2.5, it sends > out 1 ack when it receives, on the average, more than 20 packets. It is generally believed in the research community (at least the part that I work next door to) that at least 1:2 is desirable and there is sentiment in some quarters to get rid of delayed acks altogether and always ack every packet immediately. The way in which acking frequency interacts with congestion control and avoidance is non-trivial, and changing it is not a good idea unless you think you understand TCP better than Van Jacobson. -GAWollman -- Garrett A. Wollman | Shashish is simple, it's discreet, it's brief. ... wollman@lcs.mit.edu | Shashish is the bonding of hearts in spite of distance. Opinions not those of| It is a bond more powerful than absence. We like people MIT, LCS, ANA, or NSA| who like Shashish. - Claude McKenzie + Florent Vollant