From: | Lee Azzarello <lee(at)dropio(dot)com> |
---|---|
To: | pgsql-admin(at)postgresql(dot)org |
Subject: | Re: postgresql with storage |
Date: | 2009-02-26 16:43:06 |
Message-ID: | 16b031c0902260843o21187a53k4ce476748153adf1@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
On Thu, Feb 26, 2009 at 7:02 AM, Nagalingam, Karthikeyan
<KARTHIKEYAN(dot)Nagalingam(at)netapp(dot)com> wrote:
> Hi,
>
> I have some basic query in postgresql with storage, Please help me for the
> following
>
> 1. What is the best practice to use postgresql with storage.
Depends on your workload. My workload has a high number of DELETE
statements with a relatively small total set size of recurring SELECT
statements, therefore the buffer cache gets filled quickly and reads
rarely touch disk, though the disks stays busy from the DELETES. In
general, throw as much as you can at the disks unless the budget
prevents your from doing so.
Also, watch out for cheap three disk RAID-5 arrays. There's a crop of
low end servers that have these. Serious trouble.
> 2. Which are the files and folders we can keep it in storage.
The postgresql installation manual documents the location of a cluster on disk.
-lee
From | Date | Subject | |
---|---|---|---|
Next Message | Scott Marlowe | 2009-02-26 17:18:00 | Re: Question to transaction ID wraparound |
Previous Message | Lee Azzarello | 2009-02-26 16:36:36 | Re: index for inet(ip address or ip range) |