From nobody Thu May 18 10:02:23 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B9E6D12EAED for ; Thu, 18 May 2017 10:02:21 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0.8 X-Spam-Level: X-Spam-Status: No, score=0.8 tagged_above=-999 required=5 tests=[BAYES_50=0.8, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id U9pyvETKeCYi for ; Thu, 18 May 2017 10:02:19 -0700 (PDT) Received: from mx1.cnes.fr (mx1.cnes.fr [194.199.174.200]) by ietfa.amsl.com (Postfix) with ESMTP id 0190812EBA2 for ; Thu, 18 May 2017 09:56:31 -0700 (PDT) X-IronPort-AV: E=Sophos; i="5.38,359,1491264000"; d="scan'208,217"; a="11404592" X-IPAS-Result: A2EnAQAa0R1Z/wIBeApdGgEBAQECAQEBAQgBAQEBgm5nUhCBDAeNfqIrhTiCDyyFeAKFbz8YAQIBAQEBAQEBA2gdC4UbATBMEgEFEBVWJgEEDg2KGw6xQSaKcAEBAQEBAQEBAQEBAQEBAQEBAQEaBY0UgymDP4IxBZZ0hx9wa4VBjGqKMIZTlEYfOYEKLyAlh01AN4ckAYEMAQEB From: Kuhn Nicolas To: "aqm@ietf.org" CC: "Mohit P. Tahiliani" , Ankit Deepak Thread-Topic: Design and Implementation of RFC 7928 for ns-3 Thread-Index: AdLP0bVTthAzMihyS/KNwd5yMrWQhg== Date: Thu, 18 May 2017 16:56:27 +0000 Deferred-Delivery: Thu, 18 May 2017 16:56:00 +0000 Message-ID: Accept-Language: en-US Content-Language: fr-FR X-MS-Has-Attach: X-MS-TNEF-Correlator: x-tm-as-product-ver: SMEX-11.0.0.4255-8.100.1062-23078.000 x-tm-as-result: No--15.658300-8.000000-31 x-tm-as-user-approved-sender: No x-tm-as-user-blocked-sender: No Content-Type: multipart/alternative; boundary="_000_F3B0A07CFD358240926B78A680E166FFB3D3B0TWMBXP03cnesnetad_" MIME-Version: 1.0 Archived-At: Subject: [aqm] Design and Implementation of RFC 7928 for ns-3 X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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, 18 May 2017 17:02:22 -0000 --_000_F3B0A07CFD358240926B78A680E166FFB3D3B0TWMBXP03cnesnetad_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Dear all, For your information, a paper implementing the AQM evaluation guidelines in= ns-3 has recently been published. The project is described in [1] and [2]; the source code is available in [3= ]. Thanks a lot to the authors! Cheers, Nico [1] https://aqm-eval-suite.github.io/ [2] Ankit Deepak, K. S. Shravya, and Mohit P. Tahiliani. 2017. Design and I= mplementation of AQM Evaluation Suite for ns-3. In Proceedings of the Works= hop on ns-3 (WNS3 '17). ACM, New York, NY, USA, 87-94. DOI: https://doi.org= /10.1145/3067665.3067674 [3] https://github.com/aqm-eval-suite/ns-3-dev-git --_000_F3B0A07CFD358240926B78A680E166FFB3D3B0TWMBXP03cnesnetad_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Dear all,

For your information, a paper i= mplementing the AQM evaluation guidelines in ns-3 has recently been publish= ed.

The project is described in [1]= and [2]; the source code is available in [3].

Thanks a lot to the authors!

Cheers,

Nico

[1] https://aqm-eval-suite.github.i= o/

[2] Ankit Deepak, K. S. Shra= vya, and Mohit P. Tahiliani. 2017. Design and Implementation of AQM Evaluat= ion Suite for ns-3. In Proceedings of the Workshop on ns-3 (WNS3 '17). ACM,= New York, NY, USA, 87-94. DOI: https://doi.org/10.1145/3067665.3067674<= /o:p>

[3] https://github.com= /aqm-eval-suite/ns-3-dev-git=

--_000_F3B0A07CFD358240926B78A680E166FFB3D3B0TWMBXP03cnesnetad_-- From nobody Wed May 24 15:35:06 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D00B51293D9; Wed, 24 May 2017 15:35:04 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0.701 X-Spam-Level: X-Spam-Status: No, score=0.701 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GzvTgLryXoaL; Wed, 24 May 2017 15:35:02 -0700 (PDT) Received: from server.dnsblock1.com (server.dnsblock1.com [85.13.236.178]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 13E9E128D64; Wed, 24 May 2017 15:35:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=Content-Type:In-Reply-To:MIME-Version:Date: Message-ID:From:References:Cc:To:Subject:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=81Ncocu4YZGvUdeJuLLq3BzB86qq/yQOP0nkw/cdsjM=; b=4NKNCN1aLib0IIlojcqrgomJH vcf2Ro4VWM0Yu2+IK438RwTfg/qUCIzCwQ+ansiWt9ogLtc5ETb/SNPiIi0rsVfHuwbYCZKFvzE7N l6ELGN8BiuHSV8DCMEUlDcwza9FPrDx9qv6bv/bFMQkg2r7S0CUBMBhpMlVO7y4+skhO93sxdNLNF tu3u4+oogtGIacIpICsa3cIoKM6dTd+xOnoLtJKGFBpfI5aRFJR5xcU2eYqqlkdvAoPpyNbhUScmW xA8qAdXWV7oeXDtqVSB1imt+nwEUetqy/KmJaY8QYDGJAtZ5wbW6LjhchXB9h6lMtw4yzLq8crgnM RPYz0ODlw==; Received: from 197.74.9.51.dyn.plus.net ([51.9.74.197]:56094 helo=[192.168.0.2]) by server.dnsblock1.com with esmtpsa (TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128) (Exim 4.89) (envelope-from ) id 1dDerr-0002ol-IH; Wed, 24 May 2017 23:34:59 +0100 To: Michael Menth Cc: "Rong Pan (ropan)" , tsvwg IETF list , AQM IETF list References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de> From: Bob Briscoe Message-ID: Date: Wed, 24 May 2017 23:34:58 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.1 MIME-Version: 1.0 In-Reply-To: <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de> Content-Type: multipart/alternative; boundary="------------2EEAE8EAF749BE29C695A14A" Content-Language: en-GB X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server.dnsblock1.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - bobbriscoe.net X-Get-Message-Sender-Via: server.dnsblock1.com: authenticated_id: in@bobbriscoe.net X-Authenticated-Sender: server.dnsblock1.com: in@bobbriscoe.net Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Wed, 24 May 2017 22:35:05 -0000 This is a multi-part message in MIME format. --------------2EEAE8EAF749BE29C695A14A Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Michael, I just started reading your paper. Two comments so far: *1) PIE rate averaging is an unfortunate example** *In related work, you mention PIE's measurement of the link rate as an example of use of moving averages. This is an unfortunate example, because the PIE code screws up the calculation. It averages a fraction by averaging the denominator, which I pointed out is wrong in Section 5.3.2 of my PIE review . Quoting... "The [PIE] pseudocode continually measures the sequence of dequeue times, t1, t2, t3,... that it takes to transmit constant amounts of bytes (DQ_THRESHOLD). Then the exponentially weighted moving average (EWMA) of the rate should be: ewma(depart_rate) = DQ_THRESHOLD ∗ ewma(1/t1,1/t2,1/t3,...) != DQ_THRESHOLD / ewma(t1,t2,t3,...) " PIE uses the second (incorrect) formula. In the review, I discuss how wrong this could be, with an example. *2) Exponential Moving Average is an Approximation** *In your moving averages paper, you compare your unbiased exponential moving average (UEMA) algorithm with this commonly used exponential moving average (EMA) algorithm: ewma(X) <- (1-a)*X + a*ewma(X) A few years ago I wondered about this comparison, and wrote a 2-page memo for myself to record the proof. I've temporarily uploaded it here: Moving Averages . I used a Taylor series expansion to show that the iterative formula tends to the exact formula (what you call the unbiased EMA or UEMA) as the number of readings (n) becomes large. When n is not large, you could use the formula at the top of my p2 (just before the Taylor series approximation) to calculate the ratio between the actual (unbiased) UEMA and my regular iterative EMA (REMA). This ratio can be pre-calculated, because its independent of the readings: REMA/UEMA = (1-a)*(Sum_{j=0}^{n-1} a^j) where "theta" in my formula is "a" in yours "Sum_{j=0}^{n-1}" means "the sum from j=0 to j=n-1" Your iterative formula (EMA) starts with an exception, whereas I use an iterative formula that just starts the same as it carries on. To distinguish between them, I have called mine "regular EMA" or REMA. For instance, if a = 0.75 as in your Fig 1(c), then the correction factors for various small numbers of readings, n, are: n UEMA/REMA _____________________ 1 2.29 2 1.73 3 1.46 4 1.31 5 1.22 6 1.15 7 1.11 8 1.08 9 1.06 10 1.04 11 1.03 12 1.02 13 1.02 14 1.01 15 1.01 16 1.01 17 1.01 18 1.00 19 1.00 20 1.00 Cheers Bob On 30/03/17 10:57, Michael Menth wrote: > Hi all, > > Am 30.03.2017 um 11:08 schrieb Jonathan Morton: >>> On 30 Mar, 2017, at 10:46, Bob Briscoe wrote: >>> >>> For PI2 we removed all but 2 and it worked the same or better than PIE in all our tests. I have been assessing each of the other 7 one by one for reinstatement. So far I've rejected 6. I think I can reject this last one by making the sampling time of the base PI algo dependent on the max link rate. Then when the queue goes idle, the base PI algo will decay drop down to zero no slower than the queue drains, without needing this extra heuristic. >> That’s fair enough. >> >> It sounds like the fairly coarsely discrete time intervals in PIE are the main justification for this particular heuristic, so it might be sufficient to document that WRT PIE itself. Using finer time intervals is clearly a better choice for the future. > PIE uses time intervals for measurement purposes and several parameters > contribute. We've recently done some basic work on measurement > methodology that facilitates a comparison of different measurement > approaches and better-informed parametrization by introduction of the > "memory" concept. > https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf > PIE essentially implements TDRM-DTWMA-UEMA illustrated in Fig. 6d. > > The concept of "memory" can also be applied to moving averages which are > also used in PIE for several purposes. Configuration via a "memory" can > make some heuristics more intuitive. > > Best wishes, > > Michael > > > >> - Jonathan Morton >> -- ________________________________________________________________ Bob Briscoe http://bobbriscoe.net/ --------------2EEAE8EAF749BE29C695A14A Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit Michael,

I just started reading your paper. Two comments so far:

1) PIE rate averaging is an unfortunate example
In related work, you mention PIE's measurement of the link rate as an example of use of moving averages. This is an unfortunate example, because the PIE code screws up the calculation. It averages a fraction by averaging the denominator, which I pointed out is wrong in Section 5.3.2 of my PIE review. Quoting...

"The [PIE] pseudocode continually measures the sequence of dequeue times, t1, t2, t3,... that it takes to transmit constant amounts of bytes (DQ_THRESHOLD). Then the exponentially weighted moving average (EWMA) of the rate should be:

    ewma(depart_rate) = DQ_THRESHOLD ∗ ewma(1/t1,1/t2,1/t3,...)                     != DQ_THRESHOLD / ewma(t1,t2,t3,...)"
PIE uses the second (incorrect) formula. In the review, I discuss how wrong this could be, with an example.

2) Exponential Moving Average is an Approximation
In your moving averages paper, you compare your unbiased exponential moving average (UEMA) algorithm with this commonly used exponential moving average (EMA) algorithm:
    ewma(X) <- (1-a)*X + a*ewma(X)
A few years ago I wondered about this comparison, and wrote a 2-page memo for myself to record the proof. I've temporarily uploaded it here: Moving Averages.

I used a Taylor series expansion to show that the iterative formula tends to the exact formula (what you call the unbiased EMA or UEMA) as the number of readings (n) becomes large.

When n is not large, you could use the formula at the top of my p2 (just before the Taylor series approximation) to calculate the ratio between the actual (unbiased) UEMA and my regular iterative EMA (REMA). This ratio can be pre-calculated, because its independent of the readings:
    REMA/UEMA = (1-a)*(Sum_{j=0}^{n-1} a^j)

where
"theta" in my formula is "a" in yours
"Sum_{j=0}^{n-1}" means "the sum from j=0 to j=n-1"
Your iterative formula (EMA) starts with an exception, whereas I use an iterative formula that just starts the same as it carries on. To distinguish between them, I have called mine "regular EMA" or REMA.

For instance, if a = 0.75 as in your Fig 1(c), then the correction factors for various small numbers of readings, n, are:

n       UEMA/REMA_____________________1            2.29 2            1.73 3            1.46 4            1.31 5            1.22 6            1.15 7            1.11 8            1.08 9            1.06 10            1.04 11            1.03 12            1.02 13            1.02 14            1.01 15            1.01 16            1.01 17            1.01 18            1.00 19            1.00 20            1.00
Cheers

Bob

On 30/03/17 10:57, Michael Menth wrote:

Hi all,

Am 30.03.2017 um 11:08 schrieb Jonathan Morton:

On 30 Mar, 2017, at 10:46, Bob Briscoe <ietf@bobbriscoe.net> wrote:

For PI2 we removed all but 2 and it worked the same or better than PIE in all our tests. I have been assessing each of the other 7 one by one for reinstatement. So far I've rejected 6. I think I can reject this last one by making the sampling time of the base PI algo dependent on the max link rate. Then when the queue goes idle, the base PI algo will decay drop down to zero no slower than the queue drains, without needing this extra heuristic.

That’s fair enough.

It sounds like the fairly coarsely discrete time intervals in PIE are the main justification for this particular heuristic, so it might be sufficient to document that WRT PIE itself.  Using finer time intervals is clearly a better choice for the future.

PIE uses time intervals for measurement purposes and several parameters
contribute. We've recently done some basic work on measurement
methodology that facilitates a comparison of different measurement
approaches and better-informed parametrization by introduction of the
"memory" concept.
https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf
PIE essentially implements TDRM-DTWMA-UEMA illustrated in Fig. 6d.

The concept of "memory" can also be applied to moving averages which are
also used in PIE for several purposes. Configuration via a "memory" can
make some heuristics more intuitive.

Best wishes,

Michael

 - Jonathan Morton

-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/

--------------2EEAE8EAF749BE29C695A14A-- From nobody Thu May 25 00:55:52 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5D03012E036; Thu, 25 May 2017 00:55:51 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yFsqfRxwRDNv; Thu, 25 May 2017 00:55:48 -0700 (PDT) Received: from mx04.uni-tuebingen.de (mx04.uni-tuebingen.de [134.2.5.214]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5F56812DFE0; Thu, 25 May 2017 00:55:47 -0700 (PDT) Received: from [192.168.1.101] (hsi-kbw-5-56-217-255.hsi17.kabel-badenwuerttemberg.de [5.56.217.255]) by mx04.uni-tuebingen.de (Postfix) with ESMTPSA id 033615C420; Thu, 25 May 2017 09:55:43 +0200 (CEST) To: Bob Briscoe References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de> Cc: "Rong Pan (ropan)" , tsvwg IETF list , AQM IETF list From: Michael Menth Message-ID: Date: Thu, 25 May 2017 09:55:40 +0200 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.8.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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, 25 May 2017 07:55:51 -0000 Hi Bob, hi Rong, Am 25.05.2017 um 00:34 schrieb Bob Briscoe: > Michael, > > I just started reading your paper. Two comments so far: > > *1) PIE rate averaging is an unfortunate example** > *In related work, you mention PIE's measurement of the link rate as an > example of use of moving averages. This is an unfortunate example, > because the PIE code screws up the calculation. It averages a fraction > by averaging the denominator, which I pointed out is wrong in Section > 5.3.2 of my PIE review . > Quoting... > > "The [PIE] pseudocode continually measures the sequence of dequeue > times, t1, t2, t3,... that it takes to transmit constant amounts of > bytes (DQ_THRESHOLD). Then the exponentially weighted moving average > (EWMA) of the rate should be: > > ewma(depart_rate) = DQ_THRESHOLD ∗ ewma(1/t1,1/t2,1/t3,...) > != DQ_THRESHOLD / ewma(t1,t2,t3,...) > " > PIE uses the second (incorrect) formula. In the review, I discuss how > wrong this could be, with an example. Thanks, Bob, for pointing this out to me. Rong, is PIE doing this by intent (if so, what's the reason?) or is this a flaw? > > *2) Exponential Moving Average is an Approximation** > *In your moving averages paper, you compare your unbiased exponential > moving average (UEMA) algorithm with this commonly used exponential > moving average (EMA) algorithm: > ewma(X) <- (1-a)*X + a*ewma(X) > > A few years ago I wondered about this comparison, and wrote a 2-page > memo for myself to record the proof. I've temporarily uploaded it here: > Moving Averages . > > I used a Taylor series expansion to show that the iterative formula > tends to the exact formula (what you call the unbiased EMA or UEMA) as > the number of readings (n) becomes large. > > When n is not large, you could use the formula at the top of my p2 (just > before the Taylor series approximation) to calculate the ratio between > the actual (unbiased) UEMA and my regular iterative EMA (REMA). This > ratio can be pre-calculated, because its independent of the readings: > REMA/UEMA = (1-a)*(Sum_{j=0}^{n-1} a^j) > > where > "theta" in my formula is "a" in yours > "Sum_{j=0}^{n-1}" means "the sum from j=0 to j=n-1" > Your iterative formula (EMA) starts with an exception, whereas I use > an iterative formula that just starts the same as it carries on. To > distinguish between them, I have called mine "regular EMA" or REMA. > > For instance, if a = 0.75 as in your Fig 1(c), then the correction > factors for various small numbers of readings, n, are: > > n UEMA/REMA > _____________________ > 1 2.29 > 2 1.73 > 3 1.46 > 4 1.31 > 5 1.22 > 6 1.15 > 7 1.11 > 8 1.08 > 9 1.06 > 10 1.04 > 11 1.03 > 12 1.02 > 13 1.02 > 14 1.01 > 15 1.01 > 16 1.01 > 17 1.01 > 18 1.00 > 19 1.00 > 20 1.00 Yes, the bias of E(W)MA is caused by the first sample and vanishes over time. If EMA runs continuously, it has no impact. If EMA is continuously restarted, it may have an impact. However, this is only a minor aspect of the paper. https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf The major contribution of the paper is a framework that brings together various types of moving averages, moving histograms and rate measurement methods. In particular, it relates E(W)MA's "magic weight parameter" to a memory which is also implemented by other methods (with other parameters). The concept of a memory simplifies many engineering problems. Regards, Michael > > > Cheers > > > Bob > > On 30/03/17 10:57, Michael Menth wrote: >> Hi all, >> >> Am 30.03.2017 um 11:08 schrieb Jonathan Morton: >>>> On 30 Mar, 2017, at 10:46, Bob Briscoe wrote: >>>> >>>> For PI2 we removed all but 2 and it worked the same or better than PIE in all our tests. I have been assessing each of the other 7 one by one for reinstatement. So far I've rejected 6. I think I can reject this last one by making the sampling time of the base PI algo dependent on the max link rate. Then when the queue goes idle, the base PI algo will decay drop down to zero no slower than the queue drains, without needing this extra heuristic. >>> That’s fair enough. >>> >>> It sounds like the fairly coarsely discrete time intervals in PIE are the main justification for this particular heuristic, so it might be sufficient to document that WRT PIE itself. Using finer time intervals is clearly a better choice for the future. >> PIE uses time intervals for measurement purposes and several parameters >> contribute. We've recently done some basic work on measurement >> methodology that facilitates a comparison of different measurement >> approaches and better-informed parametrization by introduction of the >> "memory" concept. >> https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf >> PIE essentially implements TDRM-DTWMA-UEMA illustrated in Fig. 6d. >> >> The concept of "memory" can also be applied to moving averages which are >> also used in PIE for several purposes. Configuration via a "memory" can >> make some heuristics more intuitive. >> >> Best wishes, >> >> Michael >> >> >> >>> - Jonathan Morton >>> > > -- > ________________________________________________________________ > Bob Briscoe http://bobbriscoe.net/ > -- Prof. Dr. habil. Michael Menth University of Tuebingen Faculty of Science Department of Computer Science Chair of Communication Networks Sand 13, 72076 Tuebingen, Germany phone: (+49)-7071/29-70505 fax: (+49)-7071/29-5220 mailto:menth@uni-tuebingen.de http://kn.inf.uni-tuebingen.de From nobody Fri May 26 07:47:02 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 09CB5129C2F; Fri, 26 May 2017 07:47:01 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.602 X-Spam-Level: X-Spam-Status: No, score=-0.602 tagged_above=-999 required=5 tests=[BAYES_05=-0.5, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=toke.dk Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9HLihRQl0p3i; Fri, 26 May 2017 07:46:59 -0700 (PDT) Received: from mail.toke.dk (mail.toke.dk [52.28.52.200]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 239BE129AB3; Fri, 26 May 2017 07:46:59 -0700 (PDT) From: =?utf-8?Q?Toke_H=C3=B8iland-J=C3=B8rgensen?= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=toke.dk; s=20161023; t=1495810010; bh=GGSRqGBQAm3smJvcXPOcPb14ZR88xcSuFRoQErd3R28=; h=From:To:Cc:Subject:References:Date:In-Reply-To:From; b=ZXx58TpjTp91LrYQwZEm+TIc1i7he23j+UV5L9x0/SYaFzy8bhltTEjUqS3Vido09 spTpOJFzK51YkuVrabMwsJKHhU0N0ie4PSqwqvV+eGcNbBK+2D1hv5ioXKBTOVthM4 DcS8Uf1JdaTomdFrPG9Rb9ahA+1GpdA270Y1XR1vtSyItnCDWd6YYaOMK2v7XpfwZD V7irUEqmO7HQUvXPZfz6dZIg4C6KOfpjh4kGZadhkmW0ZgyOk4D9Y48ajMGucv+ASx xMzY13O+g/ps7TVaGwaavMHdgJuE1yEaGfTAt9j1TvSa1tmhECrz02Gl2pk81wYGbK Uf0u3ZXL1ufrA== To: Jana Iyengar Cc: "Mirja Kuehlewind \(IETF\)" , Wesley Eddy , "aqm-chairs\@ietf.org" , aqm@ietf.org, The IESG , Warren Kumari , draft-ietf-aqm-codel@ietf.org References: <149158704960.11211.6016773961643040981.idtracker@ietfa.amsl.com> <6A67DED7-961E-4966-BA3D-FD465CBF32EE@kuehlewind.net> Date: Fri, 26 May 2017 16:46:48 +0200 In-Reply-To: (Jana Iyengar's message of "Sun, 9 Apr 2017 12:06:50 -0700") X-Clacks-Overhead: GNU Terry Pratchett Message-ID: <87shjrbrqf.fsf@alrua-kau> MIME-Version: 1.0 Content-Type: text/plain Archived-At: Subject: Re: [aqm] Warren Kumari's Yes on draft-ietf-aqm-codel-07: (with COMMENT) X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Fri, 26 May 2017 14:47:01 -0000 Jana Iyengar writes: > +1 to Mirja's response. I'll do some restructuring. Thanks for your > comments! Any progress on said restructuring? :) -Toke From nobody Fri May 26 10:23:43 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 75AE8126B6E for ; Fri, 26 May 2017 10:23:42 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2 X-Spam-Level: X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=unavailable autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id u4ewF6W4rDXF for ; Fri, 26 May 2017 10:23:41 -0700 (PDT) Received: from mail-pf0-x229.google.com (mail-pf0-x229.google.com [IPv6:2607:f8b0:400e:c00::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BAF6E1294EC for ; Fri, 26 May 2017 10:23:39 -0700 (PDT) Received: by mail-pf0-x229.google.com with SMTP id e193so17292750pfh.0 for ; Fri, 26 May 2017 10:23:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20161025; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=tub6mRdA5nqyIvxSolWo2UeNctCvAXGU9FrhvhsiW5Y=; b=GwUz2BqcmzcDSgFUFkDqFJxxFjZwSo4M3BjkHoius9NBlA2u/Ze3cuOcQ8HlhLiPZZ AJDw5d/Ch3IO0fxUx+G5IGotR+s6LBUQtKJKTV6mrKNZEK/lvGfG+mMy8ApL7PWHWHn7 0nvn5h9M6wfR1qbTaJM2zv9YKv+dPWOm1rNVFfiZ0nBmp+93grKFWaYnHX6j2ICeQrT0 tNFaRLlrDEtb6n5NFTGycjWwOTop/SJR3JjYsnSx5F3+YKF3bl8af5lh/oUS4ZASYJ8j H/ovfLemrPgfxy8mCVrRAfC7hX2Dl+t9ORHvVuiQDD7kGzik8PJCc20xY1yIL3Ex3Ks2 vfLQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=tub6mRdA5nqyIvxSolWo2UeNctCvAXGU9FrhvhsiW5Y=; b=GGKS0SP61o4gNE7oPQMYcdhRH4QpsUlPOKeW4MLQeomofFHVkqonwt29Ax2wRYjv71 xWz7/QyqMPc3/w/idnqT9LEvW80mPMZZeRtDsFQYfpMpe4SAtPSX7MYZvv6PTpFQToSE VUllTQTnmE9f9RafvLkXvrjyuDgpi7NI7bbeM3xeH8X95kVSwxWI5ltsJDzHUz5WNr7Q H5PuBGPuMltC1MmJuRoBa/IR0wHxc/Zp86GhJKhAYCJkkDdPNok0yWRsogQWH2UwTaOd QkcQFszFcMzcSFXyouRMrAd9H+MJ2iY746hbDNJJgXlMmpbjbCPgoIB3kEkl7fp8Sf9G 8tjg== X-Gm-Message-State: AODbwcBYd05Vhvi3ssjV/UxIc5BFDJ0quytTJGkFVKj6MRSzGgIAaaZ8 4m/bwq8I4Ry2nnKzpP4djBDhjLNJ6M6l X-Received: by 10.84.212.8 with SMTP id d8mr59791569pli.112.1495819419235; Fri, 26 May 2017 10:23:39 -0700 (PDT) MIME-Version: 1.0 Received: by 10.100.181.165 with HTTP; Fri, 26 May 2017 10:23:38 -0700 (PDT) In-Reply-To: <87shjrbrqf.fsf@alrua-kau> References: <149158704960.11211.6016773961643040981.idtracker@ietfa.amsl.com> <6A67DED7-961E-4966-BA3D-FD465CBF32EE@kuehlewind.net> <87shjrbrqf.fsf@alrua-kau> From: Jana Iyengar Date: Fri, 26 May 2017 10:23:38 -0700 Message-ID: To: =?UTF-8?B?VG9rZSBIw7hpbGFuZC1Kw7hyZ2Vuc2Vu?= Cc: "Mirja Kuehlewind (IETF)" , Wesley Eddy , "aqm-chairs@ietf.org" , aqm@ietf.org, The IESG , Warren Kumari , draft-ietf-aqm-codel@ietf.org Content-Type: multipart/alternative; boundary="f403045d205622217d05507099b3" Archived-At: Subject: Re: [aqm] Warren Kumari's Yes on draft-ietf-aqm-codel-07: (with COMMENT) X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Fri, 26 May 2017 17:23:42 -0000 --f403045d205622217d05507099b3 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Yup, coming up next week. On Fri, May 26, 2017 at 7:46 AM, Toke H=C3=B8iland-J=C3=B8rgensen wrote: > Jana Iyengar writes: > > > +1 to Mirja's response. I'll do some restructuring. Thanks for your > > comments! > > Any progress on said restructuring? :) > > -Toke > --f403045d205622217d05507099b3 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Yup, coming up next week.

On Fri, May 26, 2017 at 7:46 AM, Toke H=C3= =B8iland-J=C3=B8rgensen <toke@toke.dk> wrote:

Jana Iyengar <jri@google.com> writes:

> +1 to Mirja's response. I'll do some restructuring. Thanks for= your
> comments!

Any progress on said restructuring? :)

-Toke

--f403045d205622217d05507099b3-- From nobody Fri May 26 10:26:13 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 723D1129459; Fri, 26 May 2017 10:26:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0.7 X-Spam-Level: X-Spam-Status: No, score=0.7 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5hg1kJRXCQzc; Fri, 26 May 2017 10:26:02 -0700 (PDT) Received: from server.dnsblock1.com (server.dnsblock1.com [85.13.236.178]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 19C3F129329; Fri, 26 May 2017 10:26:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=Content-Transfer-Encoding:Content-Type: In-Reply-To:MIME-Version:Date:Message-ID:From:References:Cc:To:Subject:Sender :Reply-To:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id:List-Help: List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=ZufaqL8ZdoT9KPu9divz2I/6rT7leMQEWBXqbxnnG/o=; b=B7C7NUnUVOqrDAcBNZDjbgS5kE LGDCZED4nwUX4uHn8EVg00o2Dnl7s3TbQWQdKfveEQtt1OVgaO52nv7Gh4isYXrGwKHUhaRn6WZow JIIe5jEuxQl1gnlgEwE6/wO+f9DGILTbjT1Y7C4qQn6T7/6iEn7xe4zP4FORChDu7dU6TX+W6UOkl hI39b6JWD2gqNkfAYyyqkCSp+twEEiZX172vN4CbzjKXmzo8FBa3eDGJ4oCO+99NN0mJMDPh1vBZN yHYBHimgMgVA9HRoCqhkvf/jC2hDIOzz+/2h6LkIBevfT+U56KC9Olw1gUF3hHex2E7a6EqtLOGwb FiQb6b2w==; Received: from 197.74.9.51.dyn.plus.net ([51.9.74.197]:37116 helo=[192.168.0.2]) by server.dnsblock1.com with esmtpsa (TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128) (Exim 4.89) (envelope-from ) id 1dEIzv-0003Pt-Vg; Fri, 26 May 2017 18:26:00 +0100 To: Michael Menth Cc: "Rong Pan (ropan)" , tsvwg IETF list , AQM IETF list References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de>

From: Bob Briscoe Message-ID: <404874be-32a9-4425-b2ff-5fbff804b5ee@bobbriscoe.net> Date: Fri, 26 May 2017 18:25:59 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Content-Language: en-GB X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server.dnsblock1.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - bobbriscoe.net X-Get-Message-Sender-Via: server.dnsblock1.com: authenticated_id: in@bobbriscoe.net X-Authenticated-Sender: server.dnsblock1.com: in@bobbriscoe.net Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Fri, 26 May 2017 17:26:05 -0000 Michael, Yes, I should have said that the most useful aspect of your paper is the model that ties all the different algos together allowing comparison. Bob On 25/05/17 08:55, Michael Menth wrote: > Hi Bob, hi Rong, > > Am 25.05.2017 um 00:34 schrieb Bob Briscoe: >> Michael, >> >> I just started reading your paper. Two comments so far: >> >> *1) PIE rate averaging is an unfortunate example** >> *In related work, you mention PIE's measurement of the link rate as an >> example of use of moving averages. This is an unfortunate example, >> because the PIE code screws up the calculation. It averages a fraction >> by averaging the denominator, which I pointed out is wrong in Section >> 5.3.2 of my PIE review . >> Quoting... >> >> "The [PIE] pseudocode continually measures the sequence of dequeue >> times, t1, t2, t3,... that it takes to transmit constant amounts of >> bytes (DQ_THRESHOLD). Then the exponentially weighted moving average >> (EWMA) of the rate should be: >> >> ewma(depart_rate) = DQ_THRESHOLD ∗ ewma(1/t1,1/t2,1/t3,...) >> != DQ_THRESHOLD / ewma(t1,t2,t3,...) >> " >> PIE uses the second (incorrect) formula. In the review, I discuss how >> wrong this could be, with an example. > Thanks, Bob, for pointing this out to me. > > Rong, is PIE doing this by intent (if so, what's the reason?) or is this > a flaw? > >> *2) Exponential Moving Average is an Approximation** >> *In your moving averages paper, you compare your unbiased exponential >> moving average (UEMA) algorithm with this commonly used exponential >> moving average (EMA) algorithm: >> ewma(X) <- (1-a)*X + a*ewma(X) >> >> A few years ago I wondered about this comparison, and wrote a 2-page >> memo for myself to record the proof. I've temporarily uploaded it here: >> Moving Averages . >> >> I used a Taylor series expansion to show that the iterative formula >> tends to the exact formula (what you call the unbiased EMA or UEMA) as >> the number of readings (n) becomes large. >> >> When n is not large, you could use the formula at the top of my p2 (just >> before the Taylor series approximation) to calculate the ratio between >> the actual (unbiased) UEMA and my regular iterative EMA (REMA). This >> ratio can be pre-calculated, because its independent of the readings: >> REMA/UEMA = (1-a)*(Sum_{j=0}^{n-1} a^j) >> >> where >> "theta" in my formula is "a" in yours >> "Sum_{j=0}^{n-1}" means "the sum from j=0 to j=n-1" >> Your iterative formula (EMA) starts with an exception, whereas I use >> an iterative formula that just starts the same as it carries on. To >> distinguish between them, I have called mine "regular EMA" or REMA. >> >> For instance, if a = 0.75 as in your Fig 1(c), then the correction >> factors for various small numbers of readings, n, are: >> >> n UEMA/REMA >> _____________________ >> 1 2.29 >> 2 1.73 >> 3 1.46 >> 4 1.31 >> 5 1.22 >> 6 1.15 >> 7 1.11 >> 8 1.08 >> 9 1.06 >> 10 1.04 >> 11 1.03 >> 12 1.02 >> 13 1.02 >> 14 1.01 >> 15 1.01 >> 16 1.01 >> 17 1.01 >> 18 1.00 >> 19 1.00 >> 20 1.00 > Yes, the bias of E(W)MA is caused by the first sample and vanishes over > time. If EMA runs continuously, it has no impact. If EMA is continuously > restarted, it may have an impact. However, this is only a minor aspect > of the paper. > https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf > > The major contribution of the paper is a framework that brings together > various types of moving averages, moving histograms and rate measurement > methods. In particular, it relates E(W)MA's "magic weight parameter" to > a memory which is also implemented by other methods (with other > parameters). The concept of a memory simplifies many engineering problems. > > Regards, > > Michael > >> >> Cheers >> >> >> Bob >> >> On 30/03/17 10:57, Michael Menth wrote: >>> Hi all, >>> >>> Am 30.03.2017 um 11:08 schrieb Jonathan Morton: >>>>> On 30 Mar, 2017, at 10:46, Bob Briscoe wrote: >>>>> >>>>> For PI2 we removed all but 2 and it worked the same or better than PIE in all our tests. I have been assessing each of the other 7 one by one for reinstatement. So far I've rejected 6. I think I can reject this last one by making the sampling time of the base PI algo dependent on the max link rate. Then when the queue goes idle, the base PI algo will decay drop down to zero no slower than the queue drains, without needing this extra heuristic. >>>> That’s fair enough. >>>> >>>> It sounds like the fairly coarsely discrete time intervals in PIE are the main justification for this particular heuristic, so it might be sufficient to document that WRT PIE itself. Using finer time intervals is clearly a better choice for the future. >>> PIE uses time intervals for measurement purposes and several parameters >>> contribute. We've recently done some basic work on measurement >>> methodology that facilitates a comparison of different measurement >>> approaches and better-informed parametrization by introduction of the >>> "memory" concept. >>> https://atlas.informatik.uni-tuebingen.de/~menth/papers/Menth17c.pdf >>> PIE essentially implements TDRM-DTWMA-UEMA illustrated in Fig. 6d. >>> >>> The concept of "memory" can also be applied to moving averages which are >>> also used in PIE for several purposes. Configuration via a "memory" can >>> make some heuristics more intuitive. >>> >>> Best wishes, >>> >>> Michael >>> >>> >>> >>>> - Jonathan Morton >>>> >> -- >> ________________________________________________________________ >> Bob Briscoe http://bobbriscoe.net/ >> -- ________________________________________________________________ Bob Briscoe http://bobbriscoe.net/ From nobody Fri May 26 11:30:03 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B935F1293F9; Fri, 26 May 2017 11:30:01 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.523 X-Spam-Level: X-Spam-Status: No, score=-14.523 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vYqGOJ-zcKNN; Fri, 26 May 2017 11:30:00 -0700 (PDT) Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B6EBD1271DF; Fri, 26 May 2017 11:29:59 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1332; q=dns/txt; s=iport; t=1495823400; x=1497033000; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-id:content-transfer-encoding: mime-version; bh=rMRZ8PYaZr3N7EXTGtaU9+y5jIEo4Ib9J0sdRLg8f2g=; b=XfVZdjaE77dCkyb9zxTaVDh4vgQpsXDb5L3VrriaRQPEjrNU7LzidJAM tJYR3WETi6FfpDimHDj9jVDkNBx6VHrdvm7FZWu46sa+Uo2TJznmud/7d J4r4ta3jnzPjVvphw8AUB/3NbKhjucK1GqIVHvM7/zemb+AUIuE9ADHxh c=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BlAQCdcyhZ/4cNJK1cGgEBAQECAQEBA?= =?us-ascii?q?QgBAQEBg1WBdoNoihinXoIPhiQCGoJwPxgBAgEBAQEBAQFrKIUZBiMEDUUQAgE?= =?us-ascii?q?IDgwCJgICAjAVEAIEAQ2KLgGrJoFsOotTAQEBAQEBAQEBAQEBAQEBAQEBAQEBH?= =?us-ascii?q?YELhVaBYCuCdId7L4IxAQSeIwGTJ5F3lE0BHziBCnQVWAGGb4hyAYEMAQEB?= X-IronPort-AV: E=Sophos;i="5.38,398,1491264000"; d="scan'208";a="432076793" Received: from alln-core-2.cisco.com ([173.36.13.135]) by alln-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 26 May 2017 18:29:58 +0000 Received: from XCH-RCD-018.cisco.com (xch-rcd-018.cisco.com [173.37.102.28]) by alln-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id v4QITwMP014763 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 26 May 2017 18:29:58 GMT Received: from xch-aln-017.cisco.com (173.36.7.27) by XCH-RCD-018.cisco.com (173.37.102.28) with Microsoft SMTP Server (TLS) id 15.0.1210.3; Fri, 26 May 2017 13:29:57 -0500 Received: from xch-aln-017.cisco.com ([173.36.7.27]) by XCH-ALN-017.cisco.com ([173.36.7.27]) with mapi id 15.00.1210.000; Fri, 26 May 2017 13:29:57 -0500 From: "Rong Pan (ropan)" To: Bob Briscoe , Michael Menth CC: tsvwg IETF list , AQM IETF list Thread-Topic: [aqm] Questioning each PIE heuristic - moving averages and rate measurement Thread-Index: AQHSqTwWerDmuF5Yhk2304NzwwextKIEvUsAgACcqQCAAjGtgP//nIaA Date: Fri, 26 May 2017 18:29:57 +0000 Message-ID: <9E1B8C70-B6C7-456C-B938-F2C08532FB69@cisco.com> References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de>

<404874be-32a9-4425-b2ff-5fbff804b5ee@bobbriscoe.net> In-Reply-To: <404874be-32a9-4425-b2ff-5fbff804b5ee@bobbriscoe.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/f.1a.0.160910 x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [171.71.130.238] Content-Type: text/plain; charset="utf-8" Content-ID: Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Fri, 26 May 2017 18:30:02 -0000 TWljaGFlbCBhbmQgQm9iLA0KDQpUaGUgZGVwYXJ0X3JhdGUgaXMgaW52ZXJzZWQgaW4gY2FsY3Vs YXRpb24gZGVsYXnigKYuDQpEZWxheSA9IHF1ZXVlX2xlbmd0aC9kZXBhcnRfcmF0ZTsgDQpIZW5j ZSwgY3VycmVudF9xZGVsYXkgPSBxdWV1ZV8uYnl0ZV9sZW5ndGgoKSAqIFBJRS0gPmF2Z19kcV90 aW1lXy9EUV9USFJFU0hPTEQ7DQoNCkJhc2ljYWxseSB0aGUgYXZlcmFnZSBkcV90aW1lIGZvciBk ZXF1ZXVlaW5nIERRX1RIUkVTSE9MRCBpcyBQSUUtPmRxX3RpbWU7IFdoYXQgaXMgdGhlIGFwcHJv eGltYXRlIHRpbWUgdG8gZGVxdWUgdGhlIGN1cnJlbnRfcWxlbj8NCkN1cnJlbnRfcWxlbi9EUV9U SFJFU0hPTEQod2hhdCBwb3J0aW9uIGlzIGN1cnJlbnQgcXVldWUgbGVuZ3RoIHJlbGF0aXZlIHRv IERRX1RIUkVTSE9MRCk/ICogYXZnX2RxX3RpbWUuDQoNClRoYXQgaXMgdGhlIHJhdGlvbmFsZSBi ZWhpbmQgaXQuDQoNClRoYW5rcywNCg0KUm9uZw0KDQogICAgPj4gKEVXTUEpIG9mIHRoZSByYXRl IHNob3VsZCBiZToNCiAgICA+Pg0KICAgID4+ICAgICAgZXdtYShkZXBhcnRfcmF0ZSkgPSBEUV9U SFJFU0hPTEQg4oiXIGV3bWEoMS90MSwxL3QyLDEvdDMsLi4uKQ0KICAgID4+ICAgICAgICAgICAg ICAgICAgICAgICAhPSBEUV9USFJFU0hPTEQgLyBld21hKHQxLHQyLHQzLC4uLikNCiAgICA+PiAi DQogICAgPj4gUElFIHVzZXMgdGhlIHNlY29uZCAoaW5jb3JyZWN0KSBmb3JtdWxhLiBJbiB0aGUg cmV2aWV3LCBJIGRpc2N1c3MgaG93DQogICAgPj4gd3JvbmcgdGhpcyBjb3VsZCBiZSwgd2l0aCBh biBleGFtcGxlLg0KICAgID4gVGhhbmtzLCBCb2IsIGZvciBwb2ludGluZyB0aGlzIG91dCB0byBt ZS4NCiAgICA+DQogICAgPiBSb25nLCBpcyBQSUUgZG9pbmcgdGhpcyBieSBpbnRlbnQgKGlmIHNv LCB3aGF0J3MgdGhlIHJlYXNvbj8pIG9yIGlzIHRoaXMNCiAgICA+IGEgZmxhdz8NCiAgICA+DQoN Cg0K From nobody Fri May 26 13:27:23 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9DBA71287A5; Fri, 26 May 2017 13:27:21 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hk9Q7WJjDr4D; Fri, 26 May 2017 13:27:20 -0700 (PDT) Received: from rcdn-iport-2.cisco.com (rcdn-iport-2.cisco.com [173.37.86.73]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4CEC9127ABE; Fri, 26 May 2017 13:27:20 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2038; q=dns/txt; s=iport; t=1495830440; x=1497040040; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-id:content-transfer-encoding: mime-version; bh=M3UznZZUbUK87vqen9UGlFYCGojcxR6zA5eOfLNJWhU=; b=adzr3U9JyFjZ5rdwdeIpGlUsM5DIr22cspSyj83rYVxjYaasKk8+0HAq m5rwOJu1naNBwRt/s+7qMei8BcE5thwWf3z+N4inq3Bi7QCr5NhqOWplA 89EEer2fVlK2PSTmAK3uaxmfrmO2phQ533ZUidSN18TJlLFs7QGygYvEI I=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AwAQBqjihZ/4wNJK1cGQEBAQEBAQEBA?= =?us-ascii?q?QEBBwEBAQEBg1VigQ0Hg2iKGJFElhqCDyELhXgCGoJwPxgBAgEBAQEBAQFrKIU?= =?us-ascii?q?ZAgEDAQEhBA06CxACAQgODAImAgICJQsVEAIEAQ0FiikBEKp7gWw6i1EBAQEBA?= =?us-ascii?q?QEBAQEBAQEBAQEBAQEBAQEYBYELhVaBYCsLgmmEaYMSL4IxAQSeIwGTJ5F3lE0?= =?us-ascii?q?BHziBCnQVRhIBhm92h3wBgQwBAQE?= X-IronPort-AV: E=Sophos;i="5.38,399,1491264000"; d="scan'208";a="253371082" Received: from alln-core-7.cisco.com ([173.36.13.140]) by rcdn-iport-2.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 26 May 2017 20:27:19 +0000 Received: from XCH-RCD-017.cisco.com (xch-rcd-017.cisco.com [173.37.102.27]) by alln-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id v4QKRJ66014116 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 26 May 2017 20:27:19 GMT Received: from xch-aln-017.cisco.com (173.36.7.27) by XCH-RCD-017.cisco.com (173.37.102.27) with Microsoft SMTP Server (TLS) id 15.0.1210.3; Fri, 26 May 2017 15:27:18 -0500 Received: from xch-aln-017.cisco.com ([173.36.7.27]) by XCH-ALN-017.cisco.com ([173.36.7.27]) with mapi id 15.00.1210.000; Fri, 26 May 2017 15:27:18 -0500 From: "Rong Pan (ropan)" To: Bob Briscoe , Michael Menth CC: tsvwg IETF list , AQM IETF list Thread-Topic: [aqm] Questioning each PIE heuristic - moving averages and rate measurement Thread-Index: AQHSqTwWerDmuF5Yhk2304NzwwextKIEvUsAgACcqQCAAjGtgP//nIaAgAAgygA= Date: Fri, 26 May 2017 20:27:18 +0000 Message-ID: <1D08E138-629C-4551-AC0E-F0F82E536A6B@cisco.com> References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de>

<404874be-32a9-4425-b2ff-5fbff804b5ee@bobbriscoe.net> <9E1B8C70-B6C7-456C-B938-F2C08532FB69@cisco.com> In-Reply-To: <9E1B8C70-B6C7-456C-B938-F2C08532FB69@cisco.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/f.1a.0.160910 x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [171.71.130.238] Content-Type: text/plain; charset="utf-8" Content-ID: <41CC2A6E2B416F4681BB76010E726D09@emea.cisco.com> Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Fri, 26 May 2017 20:27:21 -0000 QWRkaXRpb25hbCBjb25zaWRlcmF0aW9uIGlzIHRoYXQgMS90MSB3b3VsZCBpbnZvbHZlIGRpdmlk ZSwgd2hpY2ggd2lsbCBiZSBoYXJkIHRvIGltcGxlbWVudCBpbiBoYXJkd2FyZeKApg0KDQpSZWdh cmRzLA0KDQpSb25nDQoNCk9uIDUvMjYvMTcsIDExOjI5IEFNLCAiYXFtIG9uIGJlaGFsZiBvZiBS b25nIFBhbiAocm9wYW4pIiA8YXFtLWJvdW5jZXNAaWV0Zi5vcmcgb24gYmVoYWxmIG9mIHJvcGFu QGNpc2NvLmNvbT4gd3JvdGU6DQoNCiAgICBNaWNoYWVsIGFuZCBCb2IsDQogICAgDQogICAgVGhl IGRlcGFydF9yYXRlIGlzIGludmVyc2VkIGluIGNhbGN1bGF0aW9uIGRlbGF54oCmLg0KICAgIERl bGF5ID0gcXVldWVfbGVuZ3RoL2RlcGFydF9yYXRlOyANCiAgICBIZW5jZSwgY3VycmVudF9xZGVs YXkgPSBxdWV1ZV8uYnl0ZV9sZW5ndGgoKSAqIFBJRS0gPmF2Z19kcV90aW1lXy9EUV9USFJFU0hP TEQ7DQogICAgDQogICAgQmFzaWNhbGx5IHRoZSBhdmVyYWdlIGRxX3RpbWUgZm9yIGRlcXVldWVp bmcgRFFfVEhSRVNIT0xEIGlzIFBJRS0+ZHFfdGltZTsgV2hhdCBpcyB0aGUgYXBwcm94aW1hdGUg dGltZSB0byBkZXF1ZSB0aGUgY3VycmVudF9xbGVuPw0KICAgIEN1cnJlbnRfcWxlbi9EUV9USFJF U0hPTEQod2hhdCBwb3J0aW9uIGlzIGN1cnJlbnQgcXVldWUgbGVuZ3RoIHJlbGF0aXZlIHRvIERR X1RIUkVTSE9MRCk/ICogYXZnX2RxX3RpbWUuDQogICAgDQogICAgVGhhdCBpcyB0aGUgcmF0aW9u YWxlIGJlaGluZCBpdC4NCiAgICANCiAgICBUaGFua3MsDQogICAgDQogICAgUm9uZw0KICAgIA0K ICAgICAgICA+PiAoRVdNQSkgb2YgdGhlIHJhdGUgc2hvdWxkIGJlOg0KICAgICAgICA+Pg0KICAg ICAgICA+PiAgICAgIGV3bWEoZGVwYXJ0X3JhdGUpID0gRFFfVEhSRVNIT0xEIOKIlyBld21hKDEv dDEsMS90MiwxL3QzLC4uLikNCiAgICAgICAgPj4gICAgICAgICAgICAgICAgICAgICAgICE9IERR X1RIUkVTSE9MRCAvIGV3bWEodDEsdDIsdDMsLi4uKQ0KICAgICAgICA+PiAiDQogICAgICAgID4+ IFBJRSB1c2VzIHRoZSBzZWNvbmQgKGluY29ycmVjdCkgZm9ybXVsYS4gSW4gdGhlIHJldmlldywg SSBkaXNjdXNzIGhvdw0KICAgICAgICA+PiB3cm9uZyB0aGlzIGNvdWxkIGJlLCB3aXRoIGFuIGV4 YW1wbGUuDQogICAgICAgID4gVGhhbmtzLCBCb2IsIGZvciBwb2ludGluZyB0aGlzIG91dCB0byBt ZS4NCiAgICAgICAgPg0KICAgICAgICA+IFJvbmcsIGlzIFBJRSBkb2luZyB0aGlzIGJ5IGludGVu dCAoaWYgc28sIHdoYXQncyB0aGUgcmVhc29uPykgb3IgaXMgdGhpcw0KICAgICAgICA+IGEgZmxh dz8NCiAgICAgICAgPg0KICAgIA0KICAgIA0KICAgIF9fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fDQogICAgYXFtIG1haWxpbmcgbGlzdA0KICAgIGFxbUBpZXRm Lm9yZw0KICAgIGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYXFtDQogICAg DQoNCg== From nobody Fri May 26 17:45:11 2017 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48874129B48; Fri, 26 May 2017 17:45:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.001 X-Spam-Level: X-Spam-Status: No, score=-2.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ofDoGhLjL0OC; Fri, 26 May 2017 17:45:08 -0700 (PDT) Received: from server.dnsblock1.com (server.dnsblock1.com [85.13.236.178]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5A26C129447; Fri, 26 May 2017 17:45:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=Content-Transfer-Encoding:Content-Type: In-Reply-To:MIME-Version:Date:Message-ID:From:References:Cc:To:Subject:Sender :Reply-To:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id:List-Help: List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=Ii9MO9NytsJbG7jwt7E7TF6PqhimYzPAvKrZ9HDsVVg=; b=blvUdB+EBEc7O/wH38OBUFqxp4 GXk8EFVOzL5qTwTrTxNwPkT4OVa+GJDkssKPfvpySbU/kiLFu2yqkFOWNFP+23qh4FnSvzeKrWqIX KFHxGsWzJABcSCnhJCODv+9Ssn6ar3DvszFnBCtlxbVf2yn+uMhCwHn8icF5AINlXIIiEpkg/gNsY Jxj002fl5p0clds5HeZTlWjNzZm83fOy1E+9EBNcc6j+QZuOmKZVuZ2DSRXV68eUOQsHOV6uqin5L MJYA90ONxd3w5PQPGw5JQtwedZbRXFjeJ3pLh1BvmLZIs17fW73m51M52osgNH8ecyf7qPmsNaHpI QpvNceuA==; Received: from 197.74.9.51.dyn.plus.net ([51.9.74.197]:38778 helo=[192.168.0.2]) by server.dnsblock1.com with esmtpsa (TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128) (Exim 4.89) (envelope-from ) id 1dEPqk-0005k7-J4; Sat, 27 May 2017 01:45:00 +0100 To: "Rong Pan (ropan)" , Michael Menth Cc: tsvwg IETF list , AQM IETF list References: <9ddba389-e368-9050-3b14-aa235c99fcb8@bobbriscoe.net> <77D4FC66-C99F-49D0-BB73-27A0CEF70F31@gmail.com> <99a7b737-fc3c-efd0-b6c8-d71a089b7de8@bobbriscoe.net> <471e91b1-c469-3d36-9af1-0411e5661286@uni-tuebingen.de>

<404874be-32a9-4425-b2ff-5fbff804b5ee@bobbriscoe.net> <9E1B8C70-B6C7-456C-B938-F2C08532FB69@cisco.com> From: Bob Briscoe Message-ID: Date: Sat, 27 May 2017 01:44:56 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.1 MIME-Version: 1.0 In-Reply-To: <9E1B8C70-B6C7-456C-B938-F2C08532FB69@cisco.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Content-Language: en-GB X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server.dnsblock1.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - bobbriscoe.net X-Get-Message-Sender-Via: server.dnsblock1.com: authenticated_id: in@bobbriscoe.net X-Authenticated-Sender: server.dnsblock1.com: in@bobbriscoe.net Archived-At: Subject: Re: [aqm] Questioning each PIE heuristic - moving averages and rate measurement X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.22 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: Sat, 27 May 2017 00:45:10 -0000 Rong, That's a good rationale. I withdraw my criticism of PIE on this point. The code is OK, it's just the explanation that is misleading. it shouldn't say it is measuring the average dequeue rate, and there's no need to. It should describe the calculation as a moving average of the time to dequeue a set amount of bytes, scaled by the queue size relative to that number of bytes. Cheers Bob On 26/05/17 19:29, Rong Pan (ropan) wrote: > Michael and Bob, > > The depart_rate is inversed in calculation delay…. > Delay = queue_length/depart_rate; > Hence, current_qdelay = queue_.byte_length() * PIE- >avg_dq_time_/DQ_THRESHOLD; > > Basically the average dq_time for dequeueing DQ_THRESHOLD is PIE->dq_time; What is the approximate time to deque the current_qlen? > Current_qlen/DQ_THRESHOLD(what portion is current queue length relative to DQ_THRESHOLD)? * avg_dq_time. > > That is the rationale behind it. > > Thanks, > > Rong > > >> (EWMA) of the rate should be: > >> > >> ewma(depart_rate) = DQ_THRESHOLD ∗ ewma(1/t1,1/t2,1/t3,...) > >> != DQ_THRESHOLD / ewma(t1,t2,t3,...) > >> " > >> PIE uses the second (incorrect) formula. In the review, I discuss how > >> wrong this could be, with an example. > > Thanks, Bob, for pointing this out to me. > > > > Rong, is PIE doing this by intent (if so, what's the reason?) or is this > > a flaw? > > > > > _______________________________________________ > aqm mailing list > aqm@ietf.org > https://www.ietf.org/mailman/listinfo/aqm -- ________________________________________________________________ Bob Briscoe http://bobbriscoe.net/