Re: [tcmtf] [tsvwg] New version (v7) of TCM-TF draft

"Jose Saldana" <jsaldana@unizar.es> Tue, 10 June 2014 14:23 UTC

Return-Path: <jsaldana@unizar.es>
X-Original-To: tcmtf@ietfa.amsl.com
Delivered-To: tcmtf@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8AD161A0162; Tue, 10 Jun 2014 07:23:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.851
X-Spam-Level:
X-Spam-Status: No, score=-1.851 tagged_above=-999 required=5 tests=[BAYES_50=0.8, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.651, SPF_PASS=-0.001] autolearn=ham
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 EPK2vhXNV7h5; Tue, 10 Jun 2014 07:23:49 -0700 (PDT)
Received: from isuela.unizar.es (isuela.unizar.es [155.210.1.53]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 879071A0641; Tue, 10 Jun 2014 07:23:48 -0700 (PDT)
Received: from usuarioPC (gtc1pc12.cps.unizar.es [155.210.158.17]) by isuela.unizar.es (8.13.8/8.13.8/Debian-3) with ESMTP id s5AENZWQ031529; Tue, 10 Jun 2014 16:23:35 +0200
From: Jose Saldana <jsaldana@unizar.es>
To: 'Julián Fernández-Navajas' <navajas@unizar.es>, "'Eggert, Lars'" <lars@netapp.com>
References: <014d01cf8487$62adea50$2809bef0$@unizar.es> <2066ADBD-A397-4A34-AC40-4413347DB6B3@netapp.com> <5396E5A4.6090508@unizar.es>
In-Reply-To: <5396E5A4.6090508@unizar.es>
Date: Tue, 10 Jun 2014 16:23:43 +0200
Message-ID: <017a01cf84b7$9599cd00$c0cd6700$@unizar.es>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_017B_01CF84C8.592471C0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQE96092ijokgBxLOjv2xa9rtUjGDQHIKuVXAhOfxE2cbrNtoA==
Content-Language: es
X-Mail-Scanned: Criba 2.0 + Clamd & Bogofilter
Archived-At: http://mailarchive.ietf.org/arch/msg/tcmtf/fmSbal3wrFYtpBZ0LaFzcjCgh_8
Cc: tcmtf@ietf.org, 'Martin Stiemerling' <mls.ietf@gmail.com>, tsv-area@ietf.org, tsvwg@ietf.org
Subject: Re: [tcmtf] [tsvwg] New version (v7) of TCM-TF draft
X-BeenThere: tcmtf@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "Tunneling Compressed Multiplexed Traffic Flows \(TCMTF\) discussion list" <tcmtf.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcmtf>, <mailto:tcmtf-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/tcmtf/>
List-Post: <mailto:tcmtf@ietf.org>
List-Help: <mailto:tcmtf-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcmtf>, <mailto:tcmtf-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Jun 2014 14:23:53 -0000

Thanks, Julián.
 
Just a brief summary:
 
a) VoIP
- The bandwidth savings for VoIP could be up to 57%, if RTP and G729a with 2
samples per packet are used. ECRTP was used for header compression.
- pps can be reduced in the same factor as the number of flows. Up to 40
flows can be grouped together.
- in this case, the additional latency would be 20ms (inter-packet time).
This was translated into a 1% reduction of R-Factor (ranging from 1 to 100).

- ITU’s R-factor [1] is calculated from delay and packet loss
- If background traffic appears, optimized traffic can even increase
subjective quality, since the overall traffic gets reduced and so does
packet loss rate. So optimization increases delay but may reduce packet
loss.
 
Another paper about multiplexing VoIP is [3]
 
 
b) UDP-based online games
- The bandwidth savings for 8 games were obtained. For IPv4, 30 or 35% of
the bandwidth was saved. For IPv6 the reduction was between 50 and 55%.
- pps were reduced by a factor of 5 or 10
- the additional latency tested was between 5 and 50 ms. The jitter has to
be taken into account, since multiplexing adds jitter (the first packets are
delayed more than the last ones)
- A subjective quality estimator for “Quake IV” was used [2]. It is based on
delay and jitter. Packet loss does not matter, because these games implement
very good methods for concealing it.
- The MOS could be above 3.5 when multiplexing was on.
 
Jose
 
[1] ITU-T Recommendation G.107 (2011) The E-model, a computational model for
use in transmission planning.
 
[2] Wattimena A.F, Kooij R.E, van Vugt J.M, Ahmed O.K (2006) Predicting the
perceived quality of a first person shooter: the Quake IV G-model. In
Proceedings 5th SIGCOMM workshop on Network and system support for games
(NetGames '06), ACM, New York, NY, USA
 
[3] Pereira R.M, Tarouco L.M (2009) Adaptive Multiplexing Based on E-model
for Reducing Network Overhead in Voice over IP Security Ensuring
Conversation Quality. In Proceedings Fourth international Conference on
Digital Telecommunications, Washington DC, pp. 53-58
 
De: Julián Fernández-Navajas [mailto:navajas@unizar.es] 
Enviado el: martes, 10 de junio de 2014 13:02
Para: Eggert, Lars; Jose Saldana
CC: tcmtf@ietf.org; Martin Stiemerling; tsv-area@ietf.org; tsvwg@ietf.org
Asunto: Re: [tsvwg] New version (v7) of TCM-TF draft
 
Hello Lars,

I send you some publications that quantify the benefits (in terms of
bandwidth) and drawbacks (in terms of latency and loss).
VoIP: this paper evaluates the bandwidth savings and also the influence of
multiplexing on VoIP subjective quality (using ITU’s E-Model):
"
<http://diec.unizar.es/%7Ejsaldana/personal/computer_networks_2012_in_proc.p
df> Evaluating the Influence of Multiplexing Schemes and Buffer
Implementation on Perceived VoIP Conversation Quality," Computer Networks
<http://www.journals.elsevier.com/computer-networks/>  (Elsevier), Volume
56, Issue 7, Pages 1893-1919, May 2012.
<http://dx.doi.org/10.1016/j.comnet.2012.02.004> doi
10.1016/j.comnet.2012.02.004
 <http://dx.doi.org/10.1016/j.comnet.2012.02.004>  
Online games: This paper also calculates bandwidth savings:
<http://dx.doi.org/10.1016/j.comnet.2012.02.004> 
 <http://dx.doi.org/10.1016/j.comnet.2012.02.004> First Person Shooters: Can
a Smarter Network Save Bandwidth without Annoying the Players?," IEEE
Communications Magazine, vol. 49, no.11, pp. 190-198, November 2011. doi:
10.1109/MCOM.2011.6069728
 <http://dx.doi.org/10.1109/MCOM.2011.6069728>  
And this one is focused on the effect on subjective quality (using a
subjective quality estimator for a FPS game):
<http://dx.doi.org/10.1109/MCOM.2011.6069728> 
 <http://dx.doi.org/10.1109/MCOM.2011.6069728> "Online FPS Games: Effect of
Router Buffer and Multiplexing Techniques on Subjective Quality Estimators,"
Multimedia Tools and Applications, Springer. doi 10.1007/s11042-012-1309-4
 <http://dx.doi.org/10.1007/s11042-012-1309-4>  
Regards
Julián
 
 El 10/06/2014 10:59, Eggert, Lars escribió:
Hi,
 
are there any measurements or simulations that would quantify the benefits
(in terms of bandwidth) and drawbacks (in terms of latency)?
 
Lars
 
  _____