[Spread-users] spread performance testing

[Jayesh Maganlal]

Hi Ryan and all,

Need urgent help.

Referring to your (Ryan's) mail below, i conducted a test on my own with the
same scenario that you mentioned, With two spread daemons running on
different machines (on a 100 Mbit ethernet), a receive-only spflooder on one
machine, and a send/receive spflooder on the other, I sent 10000 packets of
1000 bytes each. The results that i got is no where near what you mentioned.
Look at the output below. I don't understand why is it slow. I can do the
same test using "spsend" and "sprecv" around 1sec. Could it be because of my
configuration settings that i have. Below is the configuration that i'm
using. Please advice. Looking forward for your reply.

Output:
===================================================
C:\spread-bin-3.17.1\win>java Flooder -m 10000 -b 1024 -wo
Starting multicast of 10000 messages, 1024 bytes each (self discarding).
Completed 1000 messages
Completed 2000 messages
Completed 3000 messages
Completed 4000 messages
Completed 5000 messages
Completed 6000 messages
Completed 7000 messages
Completed 8000 messages
Completed 9000 messages
Completed 10000 messages
Time: 6409ms (12.18 Mbps)   <<<< Time taken to send 6.4 seconds

C:\spread-bin-3.17.1\win>java Flooder -m 10000 -b 1024
Starting multicast of 10000 messages, 1024 bytes each.
Completed 1000 messages
Completed 2000 messages
Completed 3000 messages
Completed 4000 messages
Completed 5000 messages
Completed 6000 messages
Completed 7000 messages
Completed 8000 messages
Completed 9000 messages
Completed 10000 messages
Time: 7051ms (22.15 Mbps)  <<<< Time taken to send 7.051
===================================================

Spread Config file (comments are not in the config file):
=====================================
Spread_Segment  192.168.1.255:4803 {
 Test02 192.168.1.100       <<<<<<<<<<<<<<<<<<--------  receiver machine
 Test04 192.168.1.10         <<<<<<<<<<<<<<<<<---------sender machine
}


Thank you.
Regards,
Jayesh



Original Message dated Tue, 25 Feb 2003 23:33:55 -0500 :
Ryan Caudy mailto:caudy at jhu.edu
Tue, 25 Feb 2003 23:33:55 -0500

  a.. Previous message: [Spread-users] spread performance testing
  b.. Next message: [Fwd: Re: [Spread-users] Spread 'kill' my server]
  c.. Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

----------------------------------------------------------------------------
----

Check the archives for this list; I recall seeing such numbers in the
past.  However, in order to do some testing of your own, examine the
spflooder program included with Spread.  The source is in flooder.c
(which may be under the daemon directory, depending on what you
downloaded).  You can use this to push a large number of packets of some
specified size through spread using multiple senders/receivers.  Only
small modifications to the code should be necessary to add timing and
other data gathering.  Alternatively, you could write your own very
simple application to do approximately the same thing... one key thing
you might consider, if you do so, is making sure the set of
sending/receiving nodes waits until all other nodes have been started
and joined some specified group before they begin the experiment.

Just to give you something rough to know if you're on the right track
with your own testing, I ran a quick test of my own, using spflooder and
the unix time command.  With two spread daemons running on different
machines (on a 100 Mbit ethernet), a receive-only spflooder on one
machine, and a send/receive spflooder on the other, I sent 10000 packets
of 1000 bytes each in 0.928 seconds.  Running several such tests in an
automated fashion should give you a decent average, but the result of
the time command will always be a little higher than simply the
transmission and receipt time for the messages, since it measures total
process execution time.  So, I would suggest embedding gettimeofday()
calls in the code.

If you do generate such results, let us know what you found.  Also, I'd
be interested to hear what kind of application you're building.

Good luck,
Ryan Caudy

jayesh wrote:

>hello,
>
>i'm currently doing my masters thesis and a part of developing my project
(a
>collaborative application) is implementing SPREAD into the application to
>serve as a multicast communication. My question is has anyone done any kind
>of performance / network / bandwidth / packet , testing on spread and a
>comparison with other network (internal and external networks). If yes can
>anyone share those results with me. The reason i need those results is to
>show prove to my supervisor that spread is better so that i can proceed in
>developing my project. If no results are available can anyone please guide
>me on how to get one. Thanks.
>
>Regards,
>jayesh
>
>
>_______________________________________________
>Spread-users mailing list
>Spread-users at lists.spread.org
>http://lists.spread.org/mailman/listinfo/spread-users
>
>
>