From: | "Albe Laurenz" <laurenz(dot)albe(at)wien(dot)gv(dot)at> |
---|---|
To: | "svcntk *EXTERN*" <svcntk(at)HotPOP(dot)com> |
Cc: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: esql vs libpq |
Date: | 2009-12-30 11:03:37 |
Message-ID: | D960CB61B694CF459DCFB4B0128514C2039380F9@exadv11.host.magwien.gv.at |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
svcntk wrote:
> ESQL = ECPG (Embedded SQL in C)
>
>>> I'm doing a comparison between ESQL interfaces and libpq. For libp I
>>> use pgbench, based on TCP-C, while for ESQL have a program that also
>>> follows the transactions carried out on TCP-C.
>>>
>>> However, the result with libpq is much better, with about 700
>>> transactions per second, whereas with ESQL does not reach the figure
>>> of 400. Does anyone know if libpq has superior performance to ESQL, or
>>> is there something I am not taking into account?
>>
>> What is ESQL?
I am to lazy to undo your top posting...
For one thing, ecpg uses libpq, so it can hardly be more performant.
For a more detailed analysis, you should compare what the test
cases are actually doing. You could set log_statements='all' to
see what arrives at the server, or you can use PQtrace
(http://www.postgresql.org/docs/current/static/libpq-control.html)
in your client code.
Yours,
Laurenz Albe
From | Date | Subject | |
---|---|---|---|
Next Message | Joshua Tolley | 2009-12-30 11:42:55 | Re: DataBase Problem |
Previous Message | Nicola Farina | 2009-12-30 10:35:55 | Out of memory problem |