amazon web services - Counting number of requests per second generated by JMeter client -
this how application setup goes -
- 2 c4.8xlarge instances
- 10 m4.4xlarge jmeter clients generating load. each client used 70 threads
while conducting load test on simple request (685 bytes size page). came across issue of reduced throughput after time of test run. throughput of 18000 requests/sec reached 700 threads, remains @ level 40 minutes , drops. thread count remains 700 throughout test. have executed tests different load patterns results have been same.
the application response time considerably low throughout test -
according elb monitor, there reduction in number of requests (and suppose hence lower throughput ) -
there no errors encountered during test run. set connect timeout http request yet no errors.
i discussed issue aws support @ length , according them not blocked network limit during test execution.
given number of threads remain constant during test run, these threads doing? there metrics can check on find out number of requests generated (not hits/sec) jmeter client instance?
testplan - http://justpaste.it/qyb0
try adding following test elements:
- http cache manager
- and dns cache manager might situation threads hitting 1 c4.8xlarge instance while remaining 1 idle. see the dns cache manager: right way test load balanced apps article explanation , details.
Comments
Post a Comment