From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: [PATCH] CUBIC v2.3 with new improved slow start Date: Wed, 29 Oct 2008 16:40:45 -0700 Message-ID: <20081029164045.7422b68d@extreme> References: <006001c93a0d$477d4e30$4a580e98@ncsu2cc0c3fa00> <20081029153928.3d47e26f@extreme> <000b01c93a1c$090222c0$6c01a8c0@RHEELAPTOP> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "Injong Rhee" , To: "Injong Rhee" Return-path: Received: from mail.vyatta.com ([76.74.103.46]:35284 "EHLO mail.vyatta.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752029AbYJ2Xkz (ORCPT ); Wed, 29 Oct 2008 19:40:55 -0400 In-Reply-To: <000b01c93a1c$090222c0$6c01a8c0@RHEELAPTOP> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 29 Oct 2008 19:14:03 -0400 "Injong Rhee" wrote: > > > > This looks like a good optimization, it obviously needs more testing > > because Linux always seems to find new broken hardware. The areas > > that need to be tested should include: > > * MacOs has a broken version of delayed ack that might cause > > HyStart to radically underestimate. > > We tested with FreeBSD. I presume that it covers MacOS. We will look into > that. No Darwin added some stupid code that holds off acks for up to 2 seconds. > > * Applications that dribble out packets might get better (or > > worse) performance. This include dumb web servers. > > Not really. It does not affect those flows that do not use their full > capacity of the network. HyStart kicks in only when the flow are using more > than avail bw during slow start. If the applications do not pump data in, it > would still be in slow start -- while in slow start, HyStart works like the > traditional slow start (double cwnd per rtt). > > > * Does this increase or reduce latency when using TCP for > > applications which never fill the congestion window? (games, financial, > > etc). > > > > For the same reasons as above, No, HyStart does not engage in this > situation. > Ok