You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by afmustac <af...@yahoo.com> on 2008/05/01 23:04:04 UTC

scalability test

Dear all,

I am trying to measure the scalability of my web service implementation by testing one of its methods. I am trying to send increasing # of messages per second in each test to measure round trip times coming from my service. 
When I use JMeter for this test, I have created a test plan with a Thread Group and a WebService (SOAP) Request (Beta Code), and a Graph Result. In thread group, I set the "users" field to the # of messages that I want to send per second(change this in each run) and set ramp-up period to 0, and loop count to 1. I am expecting to get like 300-400 messages/sec but I am getting like 3600/min.. So, I am wondering what I am doing wrong? 

I really do appreciate any help or suggestions. 

thank you in advance.

afm


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

Re: scalability test

Posted by sebb <se...@gmail.com>.
2008/5/1 afmustac <af...@yahoo.com>:
> Dear all,
>
>  I am trying to measure the scalability of my web service implementation by testing one of its methods. I am trying to send increasing # of messages per second in each test to measure round trip times coming from my service.
>  When I use JMeter for this test, I have created a test plan with a Thread Group and a WebService (SOAP) Request (Beta Code), and a Graph Result. In thread group, I set the "users" field to the # of messages that I want to send per second(change this in each run) and set ramp-up period to 0, and loop count to 1. I am expecting to get like 300-400 messages/sec but I am getting like 3600/min.. So, I am wondering what I am doing wrong?
>

Graph Result is expensive to run; use Summary Report instead whilst
conducting the actual test, and reload the data for analysis later.

A loop count of 1 does not really give JMeter time to get going; there
is some startup overhead the first time elements are used.

Much better would be to use a larger loop count and use the Constant
Throughput Timer to control the rate.

Try creating a dummy test using the Java Request Sampler so you can
test things out without stressing the server or the network.

Then replace it with the SOAP sampler.  BTW, the XML-RPC sampler uses
fewer resources, so use that it it works for you.


>  I really do appreciate any help or suggestions.
>
>  thank you in advance.
>
>  afm
>
>
>
>       ____________________________________________________________________________________
>  Be a better friend, newshound, and
>  know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org