The following will work for either scenario, if you want to step-up in
increments of 500, or gradually ramp up to 500, run at 500 for a specified
amount of time, then gradually add 500 and so-on.

   - Use 10 thread groups, each with 500 threads, and set the startup delay
   equal, or greater than the ramp-up period of the previous thread.


If you want to use one thread group (simpler and more elegant), the
following may work:

   - Using beanshell read thread start times from a csv file, and set the
   start time of each thread.  Use the read line from CSV function, not the CSV
   dataset config.

This way, your ramp up, and start times are explicit, and data-driven.  You
could script the creation of the csv file with bash, perl, vbscript,
whatever.

Just out of curiousity, what hardware/OS are your 17 machines that you're
driving this with?


On Mon, Jun 15, 2009 at 9:41 AM, StudentJM <meister-kons...@gmx.de> wrote:

>
> Of course.
>
>
> I want to simulate 5000 virtual Users (Threads) on my application.
> The way i want to do this, is not to start all 5000 of them within a
> specific time.
> My imagination to make this possible with distributed testing.
>
> I've got 17 computers. 16 slaves, 1 master.
> The master gets the instruction to start 5000 threads in a specifiv ramp up
> period time.
> At this point, is there any way to say them to start the first 500 Threads,
> then wait for instance one minute, start the next 500 threads (in 500
> seconds maybe), then wait another minute, start the next 500 Threads and so
> on, until it reaches the full amount of 5000 Threads. I think its called
> scaling test.
>
>
> Thanks
>
> --
> View this message in context:
> http://www.nabble.com/Help-with-Constant-Timer-and-Listener%2C-gradually-Threads-tp24034570p24034866.html
> Sent from the JMeter - User mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscr...@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-h...@jakarta.apache.org
>
>

Reply via email to