Graph Result Throughput

[kr...@mmm.com]

I can't find a way to search the mailing list archive to see if this
question has been asked before, so I apologize if it has. I'm thinking
there must be a way to do this, I just don't know what it is...

In any case, I'm a little confused over what the throughput average
actually means. I ran my first test with just 1 user and noticed that the
throughput started at about 20 trxn / min and climbed all the way to 217
trxn / min within 14000 samples. There is no think time. All other results
remained reasonably constant. I would expect that throughput should also
remain fairly constant. I want to run tests with more users and compare
throughput; and response but I don't trust the numbers I'm getting for
throughput.

Is there a problem here or can this be explained?

Thanks in advance...


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

Re: Graph Result Throughput

[Michael Stover <ms...@apache.org>]

Well, given that you only simulated 1 user, you should be able to
approximately calculate what the expected throughput should be based on
the average response time.  That should give you an idea if the 217
number is trustworthy.

Throughput calculations typically get more accurate as time goes by.  At
the very start, it's going to be wildly inaccurate.  It's almost certain
to start too low and rise to the correct value, though it should start
approximating the correct value fairly quickly.  You didn't mention how
long it took to go from 20 to close to 217.  Was it a steep climb?  Slow
and steady?  Slow and steady would indicate a problem to me.  It depends
on how much variance the average response time showed.

-Mike

On Sat, 2005-03-12 at 21:01, kroberts@mmm.com wrote:
> 
> 
> I can't find a way to search the mailing list archive to see if this
> question has been asked before, so I apologize if it has. I'm thinking
> there must be a way to do this, I just don't know what it is...
> 
> In any case, I'm a little confused over what the throughput average
> actually means. I ran my first test with just 1 user and noticed that the
> throughput started at about 20 trxn / min and climbed all the way to 217
> trxn / min within 14000 samples. There is no think time. All other results
> remained reasonably constant. I would expect that throughput should also
> remain fairly constant. I want to run tests with more users and compare
> throughput; and response but I don't trust the numbers I'm getting for
> throughput.
> 
> Is there a problem here or can this be explained?
> 
> Thanks in advance...
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
-- 
Michael Stover <ms...@apache.org>
Apache Software Foundation


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