Re: [ippm] :draft-sun-ippm-flowbased-pm-00.txt

McGrady Sun <lishunsun@gmail.com> Fri, 26 October 2012 05:54 UTC

MIME-Version: 1.0
In-Reply-To: <CAHJgTXVMDjNVioBSgSGmPNH9S5_vFMe=CcB4x0D7xjFJkg86kA@mail.gmail.com>
References: <CAHJgTXVMDjNVioBSgSGmPNH9S5_vFMe=CcB4x0D7xjFJkg86kA@mail.gmail.com>
Date: Fri, 26 Oct 2012 13:54:04 +0800
Message-ID: <CAHJgTXVu69Ui2wuuZKb9jr7h0xmePGyr=WYNO+mtrYLs5R-ppQ@mail.gmail.com>
From: McGrady Sun <lishunsun@gmail.com>
To: ippm@ietf.org
Content-Type: multipart/alternative; boundary="047d7bb04f227bb1b604cceff27b"
Subject: Re: [ippm] :draft-sun-ippm-flowbased-pm-00.txt
Precedence: list

Hi,

 While TWAMP works well in most situations, it has some problems in
application flow measurement. For such considerations FPM is proposed.

 It is necessary to real-time measure application flows in some cases:
network needs to real-time monitor application flows, and take actions such
as resource allocation or congestion control in time according to
measurement results.

 TWAMP has to copy a flow similar to data flow in network in order to get
more accurate results if it wants to support real-time measurement of
application flows. However, it will increase network load, and has bad
effects on the performance of online applications.

 FPM measurement proposed here can address the problem above well: it
injects small number of test packets in application flows, and carries
flow-related information in test packets to track and record states of each
flow. In this way it can work well to real-time measure application flows
with little effect on them.

regards,

Lishun Sun

2012/10/17 McGrady Sun <lishunsun@gmail.com>

> Dear all,
>
>     Last week we posted a new draft based on
> http://datatracker.ietf.org/doc/draft-sun-tsvwg-flowbased-pm/?include_text=1
> . The following modifications were made.
>
>     1. A detailed description of the problem statement is contained,
> including both the motivation and scenarios of the proposed method. Also, a
> use case is attached.
>
>     2. Describes how the IPPM metrics are supported by the proposed
> flow-based performance measurement.
>
>     3. The description of the measurement procedure is also revised.
>
>     We would like to thank Al Morton and Andreas Johnsson for their
> comments and suggestions on the original version.
>
>
> Best Regards,
>
> Lishun Sun
>
> ---------- Forwarded message ----------
> From: <internet-drafts@ietf.org>
> Date: 2012/10/12
> Subject: New Version Notification for draft-sun-ippm-flowbased-pm-00.txt
> To: lishunsun@gmail.com
> Cc: wdwang@bupt.edu.cn, grace.yufang@huawei.com
>
>
>
> A new version of I-D, draft-sun-ippm-flowbased-pm-00.txt
> has been successfully submitted by Lishun Sun and posted to the
> IETF repository.
>
> Filename:        draft-sun-ippm-flowbased-pm
> Revision:        00
> Title:           Flow-based Performance Measurement
> Creation date:   2012-10-12
> WG ID:           Individual Submission
> Number of pages: 21
> URL:
> http://www.ietf.org/internet-drafts/draft-sun-ippm-flowbased-pm-00.txt
> Status:
> http://datatracker.ietf.org/doc/draft-sun-ippm-flowbased-pm
> Htmlized:        http://tools.ietf.org/html/draft-sun-ippm-flowbased-pm-00
>
>
> Abstract:
>    The performance measurements of service flow are becoming significant
>    important for administrators monitoring the fitness of the network.
>    This memo defines an end-to-end flow-based performance measurement
>    method, which is achieved by generating synthetic measurement
>    packets, injecting them to the network and analyzing the statistics
>    carried in the measurement packets.This measurement method can
>    measure flow characteristics such as delay, ipdv (IP Packet Delay
>    Variation) and packet loss.
>
>
>
>
> The IETF Secretariat
>
>
>

[ippm] :draft-sun-ippm-flowbased-pm-00.txt McGrady Sun
Re: [ippm] :draft-sun-ippm-flowbased-pm-00.txt McGrady Sun