> Um, that's not too surprising, because they're exactly the same patch?
Wrong diff. Here is correct one.
Everything else is right. I just re-checked :)
step2a:
number of transactions actually processed: 16325
latency average: 49.008 ms
latency stddev: 8.780 ms
tps = 163.182594 (including connections establishing)
tps = 163.189818 (excluding connections establishing)
step2b-fixed:
number of transactions actually processed: 16374
latency average: 48.867 ms
latency stddev: 8.223 ms
tps = 163.658269 (including connections establishing)
tps = 163.666318 (excluding connections establishing)