Re: [aqm] Gathering Queue Length Statistics

grenville armitage <> Thu, 26 February 2015 04:19 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id B39F21A19EC for <>; Wed, 25 Feb 2015 20:19:20 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -0.506
X-Spam-Status: No, score=-0.506 tagged_above=-999 required=5 tests=[BAYES_05=-0.5, HELO_EQ_AU=0.377, HOST_EQ_AU=0.327, RCVD_IN_DNSWL_LOW=-0.7, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id nSt1PTW1Z0HD for <>; Wed, 25 Feb 2015 20:19:18 -0800 (PST)
Received: from ( []) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 260D31A01AA for <>; Wed, 25 Feb 2015 20:19:17 -0800 (PST)
Received: from [] ( []) by (8.14.3/8.14.3) with ESMTP id t1Q4JF6V010274 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 26 Feb 2015 15:19:16 +1100
Message-ID: <>
Date: Thu, 26 Feb 2015 15:19:15 +1100
From: grenville armitage <>
User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:16.0) Gecko/20121107 Thunderbird/16.0.2
MIME-Version: 1.0
References: <BAY176-W35105790F5BC047AA12309A5170@phx.gbl>
In-Reply-To: <BAY176-W35105790F5BC047AA12309A5170@phx.gbl>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: <>
Subject: Re: [aqm] Gathering Queue Length Statistics
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Discussion list for active queue management and flow isolation." <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Thu, 26 Feb 2015 04:19:20 -0000

On 02/26/2015 04:53, Ryan Doyle wrote:
> I am using the fq_codel, codel, and pie qdiscs for my research and am
> wondering if there is a way to collect statistics regarding the
> average queue length since a qdisc was enabled? I have looked at tc's
> "-s" flag for statistics, but they show nothing about queue length
> and I have been unable to find anything else that might help me get
> queue length statistics.

Perhaps off on a tangent, but you might find a useful complement to your measurements. If you tcpdump the packets going in and out either side of your router, spp can take the two dump files and estimate the instantaneous RTT (itself a reflection of instantaneous queue depth) over time between said router ports. Use this to validate/contrast with RTT estimates you extract from TCP timestamps.