From: | "Oleg Serov" <serovov(at)gmail(dot)com> |
---|---|
To: | "Andrew Dunstan" <andrew(at)dunslane(dot)net> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Pl/Perl function: Speed of the First time executing pl/perl function in connection; |
Date: | 2008-11-16 09:59:49 |
Message-ID: | cec7c6df0811160159l5b942d53jb208e947bcf650bd@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Why pl/pgsql doesn't have this effect ?
2008/11/16 Andrew Dunstan <andrew(at)dunslane(dot)net>:
>
>
> Oleg Serov wrote:
>>
>> When perl function executes first time, it is too slowly, but if
>> execute perl function(not function which executed first time) again it
>> runs in 1000 times faster. Why ? how can i optimize it ?
>> Configure shared_preload_libraries = '$libdir/plperl' or
>> local_preload_libraries = '$libdir/plugins/plperl' does not help;
>>
>>
>
> The function is recompiled on the first call in each backend. (The same is
> true of most PLs, BTW, it's not a perl-only problem.) There is no immediate
> cure, unfortunately. Using pooled connections might help to mitigate the
> effect.
>
> One might imagine providing for certain functions to be loaded on backend
> startup, or even (where we aren't using BACKEND_EXEC) on postmaster start.
> But that would be a new feature, and we are past feature freeze for the
> upcoming release, so it's not going to happen any time soon.
>
> cheers
>
> andrrew
>
From | Date | Subject | |
---|---|---|---|
Next Message | Fujii Masao | 2008-11-16 10:02:26 | Re: Synchronous replication patch v2 |
Previous Message | Andrew Chernow | 2008-11-16 06:14:49 | Re: Compiling on HP-UX 10.20 fails |