Re: Checkpoint question

From: u15074 <u15074(at)hs-harz(dot)de>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Checkpoint question
Date: 2003-07-21 14:24:36
Message-ID: 1058797476.3f1bf7a4018bc@webmail.hs-harz.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Zitat von Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
> Probably insufficient disk bandwidth. If you have two drives available,
> try putting the WAL files (pg_xlog directory) on a different drive from
> the data files. Assuming you have adequate RAM, updates will be mainly
> limited by writes to WAL, while checkpoint doesn't touch WAL and is all
> about pushing data from RAM to the data files. So with a proper drive
> split, checkpoint really shouldn't affect update rate at all. (It could
> affect the time for SELECT queries, if they need to fetch data that
> isn't in RAM, but that didn't seem to be your complaint.)
>
> regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: the planner will ignore your desire to choose an index scan if your
> joining column's datatypes do not match
>
>
Can you specify more exactly what you mean with update rate? I moslty perform
inserts on the database (is that what you mean?).
Also I do not understand, why checkpoint does not touch WAL, but RAM. I thought
that a checkpoint reads the information from the WAL-Files and pushes these
information to the data files.

Thanks Andreas

-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Shridhar Daithankar 2003-07-21 14:32:27 A doubt w.r.t WAL
Previous Message nolan 2003-07-21 14:23:04 psql and toggles