From: | "James Im" <im-james(at)hotmail(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Controlling memory of session |
Date: | 2007-01-18 07:41:27 |
Message-ID: | BAY7-F15094FDC150559204F437A96AA0@phx.gbl |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Ok so the solution is to limit the number of connections. But it seems
that there is no good way to choose the ideal number of connections as I
don't know how much RAM will a connection use.
If a connection takes 3MB (on windows I see the process in the Process
Monitor, in Linux the RSS is more like 5MB) just after creation can I
limit its growth potential so that I know that it will not 6,10, 20MB
under some circumstances.
Not knowing how much RAM can take the server is annoying. You have to be
extra careful and scale the server down as you don't know what will happen.
Richard Huxton wrote:
>James Im wrote:
>>Richard Huxton wrote:
>> > Is there a particular problem you're trying to solve?
>>
>>yes I'm trying to know how many connections can open to the database
>>without running out of memory. Ideally I would like to optimize stuff
>>so that I can open the maximum number of connection/session.
>
>What - just as many idle connections as you can? You're not interesting
>in running queries with them?
>
>>In total I can give 250Mb of RAM to Postgresql. It should not eat more
>>memory. This is what Im' trying to do.
>
>It hasn't got that sort of hard limit facility. The best you can do is
>to limit the maximum number of connections and then restrict the various
>memory settings per-backend and shared. That should let you keep it in
>the range you want.
>
_________________________________________________________________
Vlg selv hvordan du vil kommunikere - skrift, tale, video eller billeder
med MSN Messenger: http://messenger.msn.dk/ - her kan du det hele
From | Date | Subject | |
---|---|---|---|
Next Message | Thuy Nguyen | 2007-01-18 08:03:25 | Ask about large database? |
Previous Message | Tom Lane | 2007-01-18 07:08:25 | Re: Index bloat of 4x |