From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail2.tohojo.dk ([77.235.48.147]:42662 "EHLO mail2.tohojo.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752656AbcFEKwE (ORCPT ); Sun, 5 Jun 2016 06:52:04 -0400 From: =?utf-8?Q?Toke_H=C3=B8iland-J=C3=B8rgensen?= To: Luca Muscariello Cc: "linux-wireless\@vger.kernel.org" , "make-wifi-fast\@lists.bufferbloat.net" , "ath9k-devel\@lists.ath9k.org" Subject: Re: [Make-wifi-fast] [RFC/RFT 0/5] Adding an airtime fairness scheduler to ath9k References: <20160603165144.17356-1-toke@toke.dk> Date: Sun, 05 Jun 2016 12:51:58 +0200 In-Reply-To: (Luca Muscariello's message of "Sat, 4 Jun 2016 17:24:42 +0200") Message-ID: <877fe3ucvl.fsf@toke.dk> (sfid-20160605_125221_878750_380462D6) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-wireless-owner@vger.kernel.org List-ID: Luca Muscariello writes: > I don't fully understand your plots but it would be useful to report > the physical rate of the stations. Yes, well, there's not really one rate to report for each station, since Minstrel jumps about a bit and tries different ones. > As a benchmark, if you know the physical rates assuming they are also > optimally chosen (by minstrel for instance ) and stations don't move, > the long term throughout can be computed ( e.g. for TCP ) assuming air > time fairness. Than you can understand if your gain is what you should > expect or if the implementation is not yet done. So far I've just been looking at the figures for airtime (the first graph in the blog post). These are the same numbers that the scheduler uses to make scheduling decisions. It seems like the scheduler does help somewhat, but is not perfect yet. Am definitely lacking a good ground truth to compare against, though. Computing the expected throughput might be possible, since minstrel does report statistics for how many packets were transmitted at each rate. Will look into it; thanks for the suggestion :) -Toke