[Pgpool-general] Combining pgpool and pgbouncer

[Tatsuo Ishii]

> > Interesting, but how would it reduce the number of connections pgpool needs to deal with? Unless you can't get the pooling behaviour you want from pgpool? Is it not pooling the connections in the way you want?
> > 
> > In your previous message you stated you needed up to 600 concurrent connections, so if you also want up to 600 concurrent connections coming from pgbouncer you'd still need 600 pgpool backends no matter which way around you chain them wouldn't you?
> 
> Per my post, most of those connections are idle most of the time.  So if
> I use pgbouncer in "transaction" mode, I end up with only around 35
> connections.
> 
> We need pgpool because of the load balancing, which pgbouncer does NOT do.

Interesting. So once clients connects to pgbouncer, it keeps the
connection to clients. When a client starts a transaction, it connects
to PostgreSQL(or pgpool in your case). When a client finishes the
transaction, pgbouncer disconnects to PostgreSQL. Question is, what
does pgbouncer deal with session span data, such as temporary tables
or datestyle set by SET command?
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp