[pgpool-general: 7821] Why running select count(*) from a huge table via pgpool is using only one worker?

[Luca Maranzano]

Hi all,
we noticed that if we run a select count(*) from a huge table connecting to
pgpool, postgres will start only 1 worker, while if we start the same query
with a direct connection to the DB postgres will start 5 workers, so
obviously execution time will be 5x time slower via pgpool.
PGpool should pass the statement to postgres so the execution on the
backend should be the same.

I'm missing the reason for this behaviour, can anyone explain?

Current versions of the components are PostgreSQL 12.8 and PGPool-II 4.2.3.

Thanks a lot in advance
Best regards
Luca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-general/attachments/20211023/a0c6c0eb/attachment.htm>