From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ed W Subject: Re: Raise initial congestion window size / speedup slow start? Date: Thu, 15 Jul 2010 00:01:37 +0100 Message-ID: <4C3E41D1.1090609@wildgooses.com> References: <4C3D94E3.9080103@wildgooses.com> <4C3DD5EB.9070908@tmr.com> <20100714.111553.104052157.davem@davemloft.net> <4C3E0684.5060409@wildgooses.com> <4C3E1B54.40604@hp.com> <20100714203919.GD6682@nuttenaction> <4C3E34AB.2060405@wildgooses.com> <20100714223633.GJ6682@nuttenaction> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Rick Jones , David Miller , davidsen@tmr.com, linux-kernel@vger.kernel.org, netdev@vger.kernel.org To: Hagen Paul Pfeifer Return-path: Received: from mail1.nippynetworks.com ([212.227.250.41]:59492 "EHLO mail1.nippynetworks.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754185Ab0GNXBj (ORCPT ); Wed, 14 Jul 2010 19:01:39 -0400 In-Reply-To: <20100714223633.GJ6682@nuttenaction> Sender: netdev-owner@vger.kernel.org List-ID: > Do you cite "An Argument for Increasing TCP's Initial Congestion Window"? > People at google stated that a CWND of 10 seems to be fair in their > measurements. 10 because the test setup was equipped with a reasonable large > link capacity? Do they analyse their modification in environments with a small > BDP (e.g. multihop MANET setup, ...)? I am curious, but We will see what > happens if TCPM adopts this. > Well, I personally would shoot for starting from the position of assuming better than zero knowledge about our link and incorporating that into the initial cwnd estimate... We know something about the RTT from the syn/ack times, speed of the local link and quickly we will learn about median window sizes to other destinations, plus additionally the kernel has some knowledge of other connections currently in progress. With all that information perhaps we can make a more informed option than just a hard coded magic number? (Oh and lets make the option pluggable so that we can soon have 10 different kernel options...) Seems like there is evidence that networks are starting to cluster into groups that would benefit from a range of cwnd options (higher/lower) - perhaps there is some way to choose a reasonable heuristic to cluster these and choose a better starting option? Cheers Ed W