ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » IBM MQ Performance Monitoring » Very poor performance on connect to MQ v7

Post new topic  This topic is locked: you cannot edit posts or make replies.
 Very poor performance on connect to MQ v7 « View previous topic :: View next topic » 
Author Message
trystan2k
PostPosted: Tue Oct 23, 2012 11:32 am    Post subject: Very poor performance on connect to MQ v7 Reply with quote

Novice

Joined: 22 Oct 2012
Posts: 22

Hi,

We had 4 machines (all of them linux, RHat 5) where we installed on two of them MQ v7.0.1.6 (I will call them A and B) and other two MQ v7.1.0.0 (call them C and D).

I´ve wrote a small Java client, using the MQSample.java code provided in MQ installation, and changed it to be able to execute 32 threads at same time, putting messages on MQ queue.

But I notice that on machine A and on machine C, to publish 500 msgs, it takes less than 10 seconds, while in machine B it take 22 secs and in D it takes 1 minute !

I then notice that it is not the put action or even the accessQueue call that takes time, but the line:

MQQueueManager qMgr = new MQQueueManager("MYQUEUEMANAGER");

I have delete all accessQueue and put lines, leaving only MQQueueManager, and got the exactly same times (well 1 or 2 secs slower, but excpected).

I executed all tests from my local machine, but even when I run it on the same machine where MQ is installed, got same time.

I also tried with amqsput and got exaclty same times:
In my DEV server, it took 6 seconds, while in my PROD box it took 55 seconds. Both are in same MQ version (7.1).

I used this shell script to put 500 using amqsput:

Code:

#!/bin/bash
#
echo " "
echo "----------------------------------------------------------------"
echo " "

PATH=$PATH:/opt/mqm/samp/bin;

export MQSERVER=EAI.CLIENT.SVRCONN/TCP/'localhost(1414)'

date1=$(date +"%s")


LIMITE=500
for ((a=1; a <= $LIMITE ; a++))
do
echo "Test for MQ Performance"|amqsput TESTQ QMANAGER
done;

date2=$(date +"%s")

diff=$(($date2-$date1))

echo "$(($diff / 60)) minutes and $(($diff % 60)) seconds elapsed."

echo "END";


Does anyone know which setting or what could be changed on C and D, so I can get same performance on A and B ?

I really appreciate any help !

THank you !

Regards,
-Thiago
Back to top
View user's profile Send private message
mqjeff
PostPosted: Tue Oct 23, 2012 11:34 am    Post subject: Reply with quote

Grand Master

Joined: 25 Jun 2008
Posts: 17447

What is the difference if you use the ping command from the OS to address the relevant hostnames (localhost, ip address, etc.)?
Back to top
View user's profile Send private message
trystan2k
PostPosted: Tue Oct 23, 2012 11:55 am    Post subject: Reply with quote

Novice

Joined: 22 Oct 2012
Posts: 22

mqjeff wrote:
What is the difference if you use the ping command from the OS to address the relevant hostnames (localhost, ip address, etc.)?


These are for the PROD server (bad performance):
[thiago]$ ping localhost
PING localhost.localdomain (127.0.0.1) 56(84) bytes of data.
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=1 ttl=64 time=0.057 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=2 ttl=64 time=0.049 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=3 ttl=64 time=0.053 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=4 ttl=64 time=0.059 ms

--- localhost.localdomain ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3001ms
rtt min/avg/max/mdev = 0.049/0.054/0.059/0.008 ms
[thiago]$ ping hostname.net
PING hostname.net (222.44.22.23) 56(84) bytes of data.
64 bytes from hostname.net (222.44.22.23): icmp_seq=1 ttl=64 time=0.081 ms
64 bytes from hostname.net (222.44.22.23): icmp_seq=2 ttl=64 time=0.053 ms
64 bytes from hostname.net (222.44.22.23): icmp_seq=3 ttl=64 time=0.052 ms
64 bytes from hostname.net (222.44.22.23): icmp_seq=4 ttl=64 time=0.046 ms

--- hostname.net ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3000ms
rtt min/avg/max/mdev = 0.046/0.058/0.081/0.013 ms
[thiago]$ ping 222.44.22.23
PING 222.44.22.23 (222.44.22.23) 56(84) bytes of data.
64 bytes from 222.44.22.23: icmp_seq=1 ttl=64 time=0.077 ms
64 bytes from 222.44.22.23: icmp_seq=2 ttl=64 time=0.044 ms
64 bytes from 222.44.22.23: icmp_seq=3 ttl=64 time=0.059 ms
64 bytes from 222.44.22.23: icmp_seq=4 ttl=64 time=0.050 ms

--- 222.44.22.23 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 2999ms
rtt min/avg/max/mdev = 0.044/0.057/0.077/0.014 ms


And these are from DEV server (good performance)

[Thiago]$ ping localhost
PING localhost.localdomain (127.0.0.1) 56(84) bytes of data.
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=1 ttl=64 time=0.067 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=2 ttl=64 time=0.097 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=3 ttl=64 time=0.093 ms
64 bytes from localhost.localdomain (127.0.0.1): icmp_seq=4 ttl=64 time=0.074 ms

--- localhost.localdomain ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3000ms
rtt min/avg/max/mdev = 0.067/0.082/0.097/0.016 ms
[Thiago]$ ping hostname.net
PING hostname.net (222.44.22.19) 56(84) bytes of data.
64 bytes from hostname.net (222.44.22.19): icmp_seq=1 ttl=64 time=0.080 ms
64 bytes from hostname.net (222.44.22.19): icmp_seq=2 ttl=64 time=0.078 ms
64 bytes from hostname.net (222.44.22.19): icmp_seq=3 ttl=64 time=0.091 ms
64 bytes from hostname.net (222.44.22.19): icmp_seq=4 ttl=64 time=0.093 ms

--- hostname.net ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3000ms
rtt min/avg/max/mdev = 0.078/0.085/0.093/0.011 ms
[Thiago]$ ping 222.44.22.19
PING 222.44.22.19 (222.44.22.19) 56(84) bytes of data.
64 bytes from 222.44.22.19: icmp_seq=1 ttl=64 time=0.069 ms
64 bytes from 222.44.22.19: icmp_seq=2 ttl=64 time=0.075 ms
64 bytes from 222.44.22.19: icmp_seq=3 ttl=64 time=0.080 ms
64 bytes from 222.44.22.19: icmp_seq=4 ttl=64 time=0.077 ms

--- 222.44.22.19 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3000ms
rtt min/avg/max/mdev = 0.069/0.075/0.080/0.007 ms

Don´t see to have much difference right ?

Thanks,
-Thiago
Back to top
View user's profile Send private message
mqjeff
PostPosted: Tue Oct 23, 2012 11:58 am    Post subject: Reply with quote

Grand Master

Joined: 25 Jun 2008
Posts: 17447

Back to top
View user's profile Send private message
trystan2k
PostPosted: Tue Oct 23, 2012 12:03 pm    Post subject: Reply with quote

Novice

Joined: 22 Oct 2012
Posts: 22

mqjeff wrote:
So why are you blaming MQ for bad network performance?


Sorry, didn't get what you mean.

I can see easily in my tests that when I connect to my DEV MQ server (just connect, do not even access any queue) wit my Java client it took about 5x less time when I connect to PROD.

I did a lot of tests, reviewed the configurations of both MQ servers and still cannot figure it out why this happens. I was hoping someone already faced it or have any idea why this is happening, where more I can look into...

Thanks,
-Thiago
Back to top
View user's profile Send private message
Vitor
PostPosted: Tue Oct 23, 2012 1:18 pm    Post subject: Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

Don't double post!

If you think you're posted in the wrong section ask for the thread to be moved.

If you think you're not getting enough help in your original thread then ask somewhere else.

Thread locked as a duplicate
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
bruce2359
PostPosted: Tue Oct 23, 2012 2:55 pm    Post subject: Reply with quote

Poobah

Joined: 05 Jan 2008
Posts: 9399
Location: US: west coast, almost. Otherwise, enroute.

trystan2k wrote:

I can see easily in my tests that when I connect to my DEV MQ server (just connect, do not even access any queue) wit my Java client it took about 5x less time when I connect to PROD.

I'm not surprised.

Are you saying that all four hardware platforms are configured (with CPU, memory, disk) exactly the same?

Most shops don't provision TEST/DEV boxes with anywhere near the resources given to PROD.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  This topic is locked: you cannot edit posts or make replies. Page 1 of 1

MQSeries.net Forum Index » IBM MQ Performance Monitoring » Very poor performance on connect to MQ v7
Jump to:  



You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
Protected by Anti-Spam ACP
 
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.