Re: RAID arrays and performance

On Thu, 18 Sep 2008, Matthew Wakeling wrote:

> On Tue, 29 Jan 2008, Gregory Stark wrote:

>> I have a patch which implements it for the low hanging fruit of bitmap
>> index scans. it does it using an extra trip through the buffer manager
>> which is the least invasive approach but not necessarily the best.
>
> Gregory - what's the status of that patch at the moment? Will it be making it
> into a new version of Postgres, or are we waiting for it to be implemented
> fully?

It and a related fadvise patch have been floating around the project queue
for a while now.  I just sorted through all the various patches and
mesasges related to this area and updated the list at
http://wiki.postgresql.org/wiki/CommitFestInProgress#Pending_patches
recently, I'm trying to kick back a broader reviewed version of this
concept right now.

> It's just that our system is doing a lot of bitmap index scans at the moment,
> and it'd help to be able to spread them across the 16 discs in the RAID
> array. It's the bottleneck in our system at the moment.

If you have some specific bitmap index scan test case suggestions you can
pass along (either publicly or in private to me, I can probably help
anonymize them), that's one of the things that has been holding this up.
Alternately, if you'd like to join in on testing this all out more help
would certainly be welcome.

--
* Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD