| From: | James Robinson <jlrobins(at)socialserve(dot)com> |
|---|---|
| To: | sdv mailer <sdvmailer(at)yahoo(dot)com> |
| Cc: | PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: PostgreSQL pre-fork speedup |
| Date: | 2004-05-06 16:44:14 |
| Message-ID: | 9BC2443E-9F7C-11D8-9C07-000A9566A412@socialserve.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On May 6, 2004, at 12:19 PM, sdv mailer wrote:
>
> 15x Slower:
> -----------
> Client <--TCP--> PgPool <--UNIX--> PostgreSQL
> Client <--TCP--> PgPool <--TCP--> PostgreSQL
>
> 5x Faster:
> ----------
> Client <--UNIX--> PgPool <--UNIX--> PostgreSQL
> Client <--UNIX--> PgPool <--TCP--> PostgreSQL
>
If the problem were in the TCP stack utilization itself, one would
expect case #1 to be equivalent to case #4, since both use one UNIX
domain connection and one TCP connection. Likewise, one would expect
case #2 to be the worst.
Does PgPool markedly differ codewise when reading from TCP socket
instead of UNIX domain? Pulling down code ...
----
James Robinson
Socialserve.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | sdv mailer | 2004-05-06 16:48:25 | Re: PostgreSQL pre-fork speedup |
| Previous Message | Bruce Momjian | 2004-05-06 16:43:46 | Re: ALTER TABLE TODO items |