jmeter-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From sebb <seb...@gmail.com>
Subject Re: JMeter Lag Over Multiple Loops
Date Tue, 06 Jun 2006 21:36:39 GMT
On 06/06/06, Morgan Monger <mongermd@yahoo.com> wrote:
> Essentially, when I run a scenario of 100 threads, 200
> sec. ramp-up, and 1 iteration, I get pretty fast and
> accurate response times. When I increase the
> iterations to 10, there is considerable lag. If this
> becomes the situation where too many threads are
> active, I may pursue a thread pool option. Any help is
> appreciated. Thanks

In the first case, it is likely that the early threads will have
finished by the time the later threads start, so the number of threads
currently active may well be far fewer than you think.

Running 10 loops means that you will more likely actually have 100
threads active at the same time.

So when you say there is a log, this could be because the system you
are testing is starting to be stressed.

Also, some to the listeners are resource intensive. Ensure that you
are using as few as possible.

As it happens, I'm just running a test on the current JMeter code:

2000 threads with a ramp-up of 60, iteration count 1000.
Single Java Request sampler, and a single Summariser Post-Processor.

The test was able to sustain 5700 - 5900 request per second, and I was
able to still just about use the PC to compose this e-mail...

The PC is not new - Athlon 1.6GHz 1.5GB Ram.

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


Mime
View raw message