Re: Performance large tables. - Mailing list pgsql-general

From Franz.Rasper@izb.de
Subject Re: Performance large tables.
Date
Msg-id D30121FCD4ADD51181D10002A587391608A04A24@M0000S0E
Whole thread Raw
In response to Performance large tables.  (Benjamin Arai <barai@cs.ucr.edu>)
Responses Re: Performance large tables.
Re: Performance large tables.
List pgsql-general
Hello,

may I ask you some questions.

What is the performance difference between U320 15kRPM and U320 10kRPM ?
Does your RAID crontoller has some memory (e.g. 128 MB or 256 MB )
and something like memory backup write cache (like HP DL 380 server) ?
Do you use Intel or Opteron cpus ?

regards,

-Franz

-----Ursprüngliche Nachricht-----
Von: Vivek Khera [mailto:vivek@khera.org]
Gesendet: Montag, 12. Dezember 2005 23:15
An: PG-General General
Betreff: Re: [GENERAL] Performance large tables.



On Dec 10, 2005, at 6:37 PM, Benjamin Arai wrote:

> For the most part the updates are simple one liners.  I currently
> commit in large batch to increase performance but it still takes a
> while as stated above.  From evaluating the computers performance
> during an update,  the system is thrashing both memory and disk.  I
> am currently using Postgresql 8.0.3.

Then buy faster disks.  My current favorite is to use U320 15kRPM
disks using a dual-chanel RAID controller with 1/2 the disks on one
channel and 1/2 on the other and mirroring them across channels, then
striping down the mirrors (ie, RAID10).

I use no fewer than 6 disks (RAID 10) for data and 2 for pg_log in a
RAID1.


---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?

               http://www.postgresql.org/docs/faq

pgsql-general by date:

Previous
From: Tom Lane
Date:
Subject: Re: Memory Leakage Problem
Next
From: John Sidney-Woollett
Date:
Subject: Re: Memory Leakage Problem