Horton

Re: PostgreSQL pre-fork speedup

Bruce Momjian <pgman@candle.pha.pa.us>

View thread View on postgr.es
From: Bruce Momjian <pgman@candle.pha.pa.us>
To: sdv mailer <sdvmailer@yahoo.com>
Cc: Rod Taylor <pg@rbt.ca>, Tatsuo Ishii <t-ishii@sra.co.jp>, Scott Marlowe <scott.marlowe@ihs.com>, steve@blighty.com, PostgreSQL Development <pgsql-hackers@postgresql.org>
Date: 2004-05-06T16:37:41Z
Lists: pgsql-hackers 
sdv mailer wrote:
> Hi Bruce,
> 
> Sorry for the confusion because Rod asked a question
> and I answered too quickly. This is what I mean.
> 
> 15x Slower:
> -----------
> Client <--TCP--> PgPool <--UNIX--> PostgreSQL
> Client <--TCP--> PgPool <--TCP--> PostgreSQL
> 
> 5x Faster:
> ----------
> Client <--UNIX--> PgPool <--UNIX--> PostgreSQL
> Client <--UNIX--> PgPool <--TCP--> PostgreSQL
> 
> 
> Hope this helps! Pgpool speeds up connection time by
> 5x with UNIX socket due to pre-fork and connection
> pooling. However, pgpool slows down by 15x under TCP
> socket for some unknown reason.

How does this compare to using tcpip without pgpool?  Is it the tcp
startup/shutdown time that is hurting performance?

pgpool is using persistent connections so I don't think a difference
would show up there, but the client/pgpool connections are being
created/destroyed frequently.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073