Hello Tatsuo-san,
> $ pgbench -C -n -p 11002 -c 10 -T 30 -f test.sql test
> tps = 95.799114 (including connections establishing)
> tps = 124.487149 (excluding connections establishing)
>
> Interesting thing is, the number from "excluding connections
> establishing". 124.487149 tps means 0.008032 seconds per
> transaction. Since the query executes pg_sleep(0.1), I think the
> number should be equal to or greater than 0.1. Maybe a tolerance, but
> 20% of error seems to be too high for me.
Indeed.
> Note that if "-C" does not present, the TPS number seems to be sane.
Hmmm... I never use -C. The formula seems ok:
tps_exclude = normal_xacts / (time_include - (INSTR_TIME_GET_DOUBLE(conn_total_time) /
nthreads));
conn_total_time is the cumulated time spent by all threads.
A quick look at the logic reveals some minor issues (conn_time is passed
as an argument to doCustom, although it is already available as a "thread"
field, stange). I spotted "threads" used instead of "thread" in an
accumulation, but it is not related to this computation.
Hmmm. I'll have a slower look...
--
Fabien.
