Re: [bmwg] Feedback for Benchmarking Methodology for OpenFlow SDN Controller Performance

"Bhuvan \(Veryx Technologies\)" <bhuvaneswaran.vengainathan@veryxtech.com> Wed, 04 June 2014 14:57 UTC

Return-Path: <bhuvaneswaran.vengainathan@veryxtech.com>
X-Original-To: bmwg@ietfa.amsl.com
Delivered-To: bmwg@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0CCD81A0290 for <bmwg@ietfa.amsl.com>; Wed, 4 Jun 2014 07:57:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.443
X-Spam-Level:
X-Spam-Status: No, score=0.443 tagged_above=-999 required=5 tests=[BAYES_05=-0.5, HTML_MESSAGE=0.001, J_CHICKENPOX_32=0.6, RELAY_IS_203=0.994, RP_MATCHES_RCVD=-0.651, SPF_PASS=-0.001] autolearn=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 jAn2GLcWZSuf for <bmwg@ietfa.amsl.com>; Wed, 4 Jun 2014 07:57:02 -0700 (PDT)
Received: from mail.veryxtech.com (mail.veryxtech.com [203.196.171.45]) by ietfa.amsl.com (Postfix) with ESMTP id EA1901A0080 for <bmwg@ietf.org>; Wed, 4 Jun 2014 07:57:00 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mail.veryxtech.com (Postfix) with ESMTP id 5263F3740DE; Wed, 4 Jun 2014 20:26:53 +0530 (IST)
X-Virus-Scanned: amavisd-new at veryxtech.com
Received: from mail.veryxtech.com ([127.0.0.1]) by localhost (mail.veryxtech.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id q6UO5Fwg15NC; Wed, 4 Jun 2014 20:26:41 +0530 (IST)
Received: from LT015PC (unknown [192.168.12.87]) by mail.veryxtech.com (Postfix) with ESMTPSA id 9F73E3740D1; Wed, 4 Jun 2014 20:26:41 +0530 (IST)
From: "Bhuvan (Veryx Technologies)" <bhuvaneswaran.vengainathan@veryxtech.com>
To: Brian.Castelli@spirent.com, bmwg@ietf.org
References: <CFB2B4B8.52D9%brian.castelli@spirent.com>
In-Reply-To: <CFB2B4B8.52D9%brian.castelli@spirent.com>
Date: Wed, 04 Jun 2014 20:26:41 +0530
Message-ID: <003c01cf8005$31d099e0$9571cda0$@veryxtech.com>
MIME-Version: 1.0
Content-Type: multipart/related; boundary="----=_NextPart_000_003D_01CF8033.4B924BC0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQIj3j5OTwbcoRHRvVEERF4r2Aso7Jq4AHHA
Content-Language: en-us
Archived-At: http://mailarchive.ietf.org/arch/msg/bmwg/HOTxu7hPOg6olAqmb2oMoUzuX_E
Cc: 'Anton Basil' <anton.basil@veryxtech.com>, 'Vishwas Manral' <vishwas.manral@gmail.com>
Subject: Re: [bmwg] Feedback for Benchmarking Methodology for OpenFlow SDN Controller Performance
X-BeenThere: bmwg@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: Benchmarking Methodology Working Group <bmwg.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bmwg>, <mailto:bmwg-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bmwg/>
List-Post: <mailto:bmwg@ietf.org>
List-Help: <mailto:bmwg-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bmwg>, <mailto:bmwg-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Jun 2014 14:57:06 -0000

Dear Brain,

 

Thank you for your valuable comments. Please see my responses inline to your
comments.

 

Trust my responses address your review comments. 

Please revert back if you need any clarifications on my responses.

 

Look forward your additional comments and suggestions on this draft.

 

Best Regards,

 

Description: Description: Description: Description: Description:
Description: Description: Description: Description: Veryx-Logo for Webinar



Bhuvan | Cloud and SDN Solutions

Phone: +91 44 4567 2222 x242

Mobile: +91 988 428 6693 

 <http://www.veryxtech.com/> www.veryxtech.com

 

From: Castelli, Brian [mailto:Brian.Castelli@spirent.com] 
Sent: Wednesday, June 04, 2014 3:30 AM
To: anton.basil@veryxtech.com; bhuvaneswaran.vengainathan@veryxtech.com;
vishwas.manral@gmail.com
Subject: Feedback for Benchmarking Methodology for OpenFlow SDN Controller
Performance

 

Dear Authors,

 

I have just joined the mailing list today. I don't know how to respond back
to the mail thread after the fact, so I am sending this feedback on the SDN
Controller benchmark document directly to you.

[Bhuvan] I'm including IETF mail id in the CC of my response. In future you
can share your comments to the email id - bmwg@ietf.org.

 

I read the document with great interest because I am working on SDN
Controller benchmarks myself. I have many comments and suggestions for
improving the document. Because I am new to the IETF process, I am sending
you comments through page 10 to validate that these are the kinds of
comments you are soliciting. I don't want to waste your time with comments
that are not constructive.

[Bhuvan] We will be happy to receive your comments and suggestions for
improving the draft.

 

Here are my comments on the document pages 1-10:

 

5.3. Southbound Connection Setup Requirements: Seems awkward to refer to
"channel setup ways". I prefer "channel configuration". The first sentence
would then read as follows: "The methodologies defined in this document are
independent of OpenFlow version, channel configuration, and connection type
(e.g. TCP or TLS/SSL)." Note the suggested punctuation changes as well.

[Bhuvan]We accept your suggestion. We are planning to rephrase the sentence
as "The methodologies defined in this document are independent of OpenFlow
version, channel setup configuration (Single/Multiple/Auxillary
connections), and connection type (e.g. TCP or TLS/SSL)."

 

6.1.1  Flow setup rate: For easier reading in subsequent paragraphs, I
suggest that you add the acronym to the title, as in "Flow Setup Rate
(FSR)". 

[Bhuvan] We will add the acronym as per your suggestion.

 

6.1.1  Flow setup rate: I think the paragraph after the title should not
specify "maximum". Testing could be for max, average, minimum, etc. So the
text ought to read as, "The FSR is the number of flows transmitted by a
controller, expressed in flows per second."

[Bhuvan] We will incorporate this feedback in the draft.

 

6.1.1.1  Flow setup rate under constant network load: Use the acronym
proposed above. Should say, "FSR under constant network load"

[Bhuvan] As suggested, we will change the title.

 

6.1.1.1.1 Objective: The test does not measure the total number. It measures
the rate. Plus the test only measures responses in the reactive mode case,
not in the proactive mode case. It would be more correct to write, "To
measure the rate of OpenFlow flow setup messages transmitted by a controller
under constant network load."

[Bhuvan] This is a valid feedback. We will rephrase as "To measure the total
number of OpenFlow flow responses (Packet-Out/Flow-Mod) received per second
from the controller under constant network load.

 

6.1.1.1.2 Setup Parameters: 

*	Number of flows is an OpenFlow setup parameter, not a network setup
parameter

[Bhuvan] Number of flows here defines the number of network flows that the
controller can setup on OpenFlow connection. We will therefore rephrase as
"Defines unique number of network flows setup on each OpenFlow connections"

 

*	Test Iterations: "need" should be "needs"
*	Test Duration: "duration of test iteration" should be "duration of
each test iteration"

[Bhuvan] We will change as per your suggestion.

 

Section 6.1.1.1.3 Procedure: 

*	In a few places we have written, "all the specified number of
switches," when, "the specified number of switches" would be more correct.

[Bhuvan] We will change as per your suggestion.

 

*	In the reactive mode setup, the text reads, "Load the controller
with Packet-In messages send from all switches for a configured number of
flows and for a specified test duration." "send" should be "sent". 

[Bhuvan] We will make the suggested correction.

 

And what does it mean to send packet-in messages "for a configured number of
flows"? Do you mean to say to send packet-ins that correspond to a number of
desired flows? 

[Bhuvan] Yes

 

Do we need to configure the controller to send down a new flow for every new
destination IP address received or something?

[Bhuvan] No need to configure the controller. Controllers by default respond
to packet-in messages with packet-out/flow-mod messages.

 

How is it that we are going to assist the person running the test at
creating the proper packet-in messages? What are the characteristics of the
packet-ins that will make this test successful?

[Bhuvan] To make the test successful, we need to simulate packet-in messages
for each flow. Typically flows are identified using dst.mac and dst.ip. 

 

*	In the reactive mode setup, there is no mention of the rate at which
packet-ins are sent, but the test result can be skewed by the rate. If I
send only one packet-in per second, for example, the FSR can't be faster
than one flow per second. I think we should consider specifying the rate at
which the packet-ins are sent to the controller, iterating over packet-in
rates, iterating over numbers of switches.

[Bhuvan] We agree with your comment. We are planning to define the minimum
number of packet-ins that need to be sent on each OpenFlow connection
(typically greater than 1). But the assumption is that we need to send as
many packet-in messages as possible on an established OpenFlow connection to
measure the FSR.

 

*	For the proactive mode, why is there a test duration? We ought to
configure the controller to push down a number of flows, connect the proper
number of switches, and measure how long it takes to send all the flows
without error.

[Bhuvan] In proactive mode, we typically measure how much flows a controller
can provision per second. Hence we need the test duration.

 

*	Without the iteration I have suggested, above, I don't understand
why the test should be repeated multiple times? Do we not expect the test
results to be consistent between test runs?

[Bhuvan] We expect slight variations across test runs. Hence we are
repeating the test multiple times to provide tester a reliable metric.

 

Section 6.1.1.1.4 Measurement:

*	FSRi equation is incorrect. FSR = number of flows/time, not
OFR/time.

[Bhuvan] I would like to provide some more clarification on this equation.
Here, the number of flows = number of packet-in messages sent to the
controller. We have conducted this test with various controllers and
observed that the number of OF responses received from the controller is
always lesser than the number of packet-in messages sent to the controller
due to the TCP window restrictions. Hence we recommend, FSR should be
measured based on the OF responses received from the controller. The FSR
equation will therefore be OF responses/time.

 

Section 6.1.1.1.5 Reporting:

*	Without iteration over the variables I have suggested, above,
running the same exact test multiple times would yield a flat line. I don't
understand why we would graph that kind of result. We should use a graph
when there are variables.

[Bhuvan] As explained above, we recommend iterating the test for multiple
times as we see variations in the results. 

 

I will withhold additional comments until I hear back from you.

 

Thank you,

Brian

 

E-mail confidentiality.
--------------------------------
This e-mail contains confidential and / or privileged information belonging
to Spirent Communications plc, its affiliates and / or subsidiaries. If you
are not the intended recipient, you are hereby notified that any disclosure,
copying, distribution and / or the taking of any action based upon reliance
on the contents of this transmission is strictly forbidden. If you have
received this message in error please notify the sender by return e-mail and
delete it from your system.

Spirent Communications plc
Northwood Park, Gatwick Road, Crawley, West Sussex, RH10 9XN, United
Kingdom.
Tel No. +44 (0) 1293 767676
Fax No. +44 (0) 1293 767677

Registered in England Number 470893
Registered at Northwood Park, Gatwick Road, Crawley, West Sussex, RH10 9XN,
United Kingdom.

Or if within the US, 

Spirent Communications,
26750 Agoura Road, Calabasas, CA, 91302, USA.
Tel No. 1-818-676- 2300