[pgpool-general: 6512] performance issue?

[Chen HS]

Hi All,

   I used 'repmge' built 1(master)+2(standby) postgresql cluster, and set
pgpool for this cluster. But I do pgbench test, can not got any improvement.
that's my test result:
10.193.10.208 is pgpool server, 10.193.10.205 is master server. I saw
access directly  is better than pgpool server, why?
----
test at ubuntu:~$ pgbench -c 10 -T 10 -S -h 10.193.10.208 -U pea pea
Password:
starting vacuum...end.
transaction type: <builtin: select only>
scaling factor: 1
query mode: simple
number of clients: 10
number of threads: 1
duration: 10 s
number of transactions actually processed: 13882
latency average = 7.248 ms
tps = 1379.733298 (including connections establishing)
tps = 1380.020165 (excluding connections establishing)

test at ubuntu:~$ pgbench -c 10 -T 10 -S -h 10.193.10.205 -U pea pea
Password:
starting vacuum...end.
transaction type: <builtin: select only>
scaling factor: 1
query mode: simple
number of clients: 10
number of threads: 1
duration: 10 s
number of transactions actually processed: 83917
latency average = 1.192 ms
tps = 8390.654113 (including connections establishing)
tps = 8396.359804 (excluding connections establishing)
------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20190418/fa343f59/attachment.html>