cbench throughput performance went down for not redesigned ofp

the throughput cbench test had rates reaching 80k-90k per second, now it lowered to 20k, this happened on May 11; latency remained stable around 15k

this is the last build with good numbers
https://jenkins.opendaylight.org/releng/view/openflowplugin/job/openflowplugin-csit-1node-cds-cbench-performance-only-master/251/

some info from it's console

Nexus timestamp is 0.3.0-20150511.184955-1533
+ BUNDLEFOLDER=distribution-karaf-0.3.0-SNAPSHOT
+ BUNDLE=distribution-karaf-0.3.0-20150511.184955-1533.zip
+ BUNDLEURL=http://nexus01.dfw.opendaylight.org:8081/nexus/content/repositories/opendaylight.snapshot/org/opendaylight/integration/distribution-karaf/0.3.0-SNAPSHOT/distribution-karaf-0.3.0-20150511.184955-1533.zip
+ echo 'Distribution bundle URL is http://nexus01.dfw.opendaylight.org:8081/nexus/content/repositories/opendaylight.snapshot/org/opendaylight/integration/distribution-karaf/0.3.0-SNAPSHOT/distribution-karaf-0.3.0-20150511.184955-1533.zip'
Distribution bundle URL is http://nexus01.dfw.opendaylight.org:8081/nexus/content/repositories/opendaylight.snapshot/org/opendaylight/integration/distribution-karaf/0.3.0-SNAPSHOT/distribution-karaf-0.3.0-20150511.184955-1533.zip
+ echo 'Distribution bundle is distribution-karaf-0.3.0-20150511.184955-1533.zip'
Distribution bundle is distribution-karaf-0.3.0-20150511.184955-1533.zip
+ echo 'Distribution folder is distribution-karaf-0.3.0-SNAPSHOT'

cbench: controller benchmarking tool
running in mode 'throughput'
connecting to controller at 10.30.11.44:6633
faking 8 switches offset 1 :: 10 tests each; 12000 ms per test
with 10000 unique source MACs per switch
learning destination mac addresses before the test
starting test with 0 ms delay after features_reply
ignoring first 1 "warmup" and last 0 "cooldown" loops
connection delay of 0ms per 1 switch(es)
debugging info is off
18:59:38.183 8 switches: flows/sec: 122599 121682 118032 120732 117003 120001 119874 122151 total = 80.168771 per ms
18:59:50.287 8 switches: flows/sec: 126213 135245 126912 131326 126149 141285 128660 133043 total = 87.381786 per ms
19:00:02.388 8 switches: flows/sec: 125177 131055 125292 127680 129307 124622 130733 133346 total = 85.600008 per ms
19:00:14.492 8 switches: flows/sec: 118733 119226 115797 116684 109574 113204 118596 110046 total = 76.817141 per ms
19:00:26.594 8 switches: flows/sec: 109714 115962 107379 111203 104073 107253 113117 114710 total = 73.616737 per ms
19:00:38.695 8 switches: flows/sec: 112521 114269 117061 130326 116425 116462 122951 121267 total = 79.270488 per ms
19:00:50.796 8 switches: flows/sec: 125080 131979 125252 122190 119770 124159 129416 129707 total = 83.959434 per ms
19:01:02.898 8 switches: flows/sec: 95076 97745 91623 94823 95441 90515 100410 97453 total = 63.587162 per ms
19:01:15.025 8 switches: flows/sec: 126704 124309 120760 126724 129114 127585 117316 123665 total = 82.833889 per ms
19:01:27.126 8 switches: flows/sec: 131301 136270 136504 130175 137593 133015 133318 132569 total = 89.227798 per ms
RESULT: 8 switches 9 tests min/max/avg/stdev = 63587.16/89227.80/80254.94/7563.92 responses/s