From: | Lincoln Yeoh <lyeoh(at)pop(dot)jaring(dot)my> |
---|---|
To: | Richard Huxton <dev(at)archonet(dot)com>, Poul Møller Hansen <freebsd(at)pbnet(dot)dk> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Setting up a database for 10000 concurrent users |
Date: | 2005-09-06 16:56:12 |
Message-ID: | 5.2.1.1.1.20050907004102.04c718a8@localhost |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
At 09:45 PM 9/5/2005 +0100, Richard Huxton wrote:
>Poul Møller Hansen wrote:
>>I'm trying to setup a database for 10000 concurrent users for a test.
>>I have a system with 1GB of RAM where I will use 512MB for PostgreSQL.
>>It is running SuSE 9.3
>
>I think you're being horribly optimistic if you actually want 10000
>concurrent connections, with users all doing things. Even if you only
>allow 1MB for each connection that's 10GB of RAM you'd want. Plus a big
>chunk more to actually cache your database files and do work in. Then, if
>you had 10,000 concurrent queries you'd probably want a mainframe to
>handle all the concurrency, or perhaps a 64-CPU box would suffice...
10GB of RAM isn't that farfetched nowadays.
However I/O might be a problem. A single drive can typically write/read
about 10MB a second (64KB chunks random access - not sure if you'd want to
bet on getting sequential throughput ;) ).
Anyway, it'll be something interesting to see ;).
Link.
From | Date | Subject | |
---|---|---|---|
Next Message | Vlad | 2005-09-06 17:45:47 | "select ..... for update of ..." doesn't support full qualified table name? |
Previous Message | Michal Hlavac | 2005-09-06 16:06:25 | tsearch2 & unicode |