Latency test for an openflow controller
hassan hatim <[email protected]>
| Newsgroups | gmane.ietf.bmwg |
|---|---|
| Message-ID | <[email protected]> |
Dears I did a latency test for an OpenFlow controller by cbench tool, the results is bellow: -K56CB:~$ taskset -c 0-3 cbench -c localhost -p 6633 -m 10000 -l 14 -w 2 -M 100000 -i 50 -I 5 -s 1 cbench: controller benchmarking tool running in mode 'latency' connecting to controller at localhost:6633 faking 1 switches offset 1 :: 14 tests each; 10000 ms per test with 100000 unique source MACs per switch learning destination mac addresses before the test starting test with 0 ms delay after features_reply ignoring first 2 "warmup" and last 0 "cooldown" loops connection delay of 50ms per 5 switch(es) debugging info is off 13:56:05.830 1 switches: flows/sec: 635968 total = 63.596794 per ms 13:56:15.931 1 switches: flows/sec: 765060 total = 76.505992 per ms 13:56:26.031 1 switches: flows/sec: 763596 total = 76.359592 per ms 13:56:36.131 1 switches: flows/sec: 765249 total = 76.524892 per ms 13:56:46.232 1 switches: flows/sec: 757221 total = 75.722062 per ms 13:56:56.332 1 switches: flows/sec: 771945 total = 77.194492 per ms 13:57:06.432 1 switches: flows/sec: 755700 total = 75.569992 per ms 13:57:16.532 1 switches: flows/sec: 755304 total = 75.530392 per ms 13:57:26.632 1 switches: flows/sec: 754737 total = 75.473692 per ms 13:57:36.733 1 switches: flows/sec: 755922 total = 75.592192 per ms 13:57:46.833 1 switches: flows/sec: 756075 total = 75.607492 per ms 13:57:56.933 1 switches: flows/sec: 757404 total = 75.740392 per ms 13:58:07.033 1 switches: flows/sec: 761826 total = 76.182592 per ms 13:58:17.134 1 switches: flows/sec: 772500 total = 77.249992 per ms RESULT: 1 switches 12 tests min/max/avg/stdev = 75473.69/77249.99/76062.31/614.30 responses/s Does it right that the calculation for the latency is done by dividing 100000 to the resulted number 76062.31 to get the average delay for all the tests in Microseconds. Thanks and regards. -Hassan _______________________________________________ bmwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/bmwg