JMeter doesn't have a 'problem' as such, it just takes longer to write to a file that is on a different machine as compared to doing the same thing for a local file. Any computer program will experience the same issue. And ot's logical that if this takes longer then the processing of a request (and I mean the logical time including Server response, JMeter processing and network response) will take longer. If it takes longer then you will get less throughput. It's simple math.
The way to get around this is to configure pacing, with something like a Constant Throughput Timer you will be able to define a buffer such that even if JMeter has to wait longer to write the results this will not impact the rate it makes requests at. If you are looking for a repeatable test to use as a baseline then this is pretty much essential. You need to create a defined load of X requests per second and run controlled tests. Simply spinning up a test where the threads try to run as fast as possible causes inaccurate results as things like faster machines/networks/responses all act together to change the load profile (as per your experience) which means the test is not repeatable. ----- http://www.http503.com/ -- View this message in context: http://jmeter.512774.n5.nabble.com/Jmeter-Performance-using-jmeter-server-VS-running-in-the-local-instance-tp4631144p4826253.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
