Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Luke Lonergan
Yay - looking forward to your results! - Luke On 8/16/07 3:14 PM, "Merlin Moncure" <[EMAIL PROTECTED]> wrote: > On 8/16/07, Luke Lonergan <[EMAIL PROTECTED]> wrote: >> >> Hi Michael, >> >> There is a problem with some Dell "perc 5" RAID cards, specifically we've >> had this problem with the

Re: [PERFORM] Indexscan is only used if we use "limit n"

2007-08-16 Thread Josh Berkus
Sebastian, > Whe are running PostgreSQL 8.2.0 on amd64-portbld-freebsd6.2, compiled > by GCC cc (GCC) 3.4.6 [FreeBSD] 20060305. The query only uses the index > if we have a "limit n": Um, why are you running an unpatched version of 8.2? You should be runing 8.2.4. -- --Josh Josh Berkus Postg

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Decibel!
On Thu, Aug 16, 2007 at 01:30:11PM -0400, [EMAIL PROTECTED] wrote: > Hi Miki, > by 40%. I have not tried RAID 10 so I cannot help you there. My > suggestion is test both RAID 5 and RAID 10, and report back to us what > you find. Unless you're running something like a data warehouse, I'd put a re

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Scott Marlowe
On 8/16/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > > > Hi Miki, > > I am using a Dell 2950, and I recently switched from using RAID 5 of all six > disks to three RAID 1 pairs with the OS on the first pair, postgres on the > second except for pg_xlog, which I moved to the third pair. This >

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Merlin Moncure
On 8/16/07, Luke Lonergan <[EMAIL PROTECTED]> wrote: > > Hi Michael, > > There is a problem with some Dell "perc 5" RAID cards, specifically we've > had this problem with the 2950 as of 6 months ago – they do not support > RAID10. They have a setting that sounds like RAID10, but it actually > im

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread tyrrill_ed
Hi Miki, I am using a Dell 2950, and I recently switched from using RAID 5 of all six disks to three RAID 1 pairs with the OS on the first pair, postgres on the second except for pg_xlog, which I moved to the third pair. This configuration change increased the insert performance of my applicatio

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Decibel!
On Thu, Aug 16, 2007 at 07:59:15PM +0200, Steinar H. Gunderson wrote: > On Thu, Aug 16, 2007 at 10:53:00AM -0700, Luke Lonergan wrote: > > They have a setting that sounds like RAID10, but it actually > > implements spanning of mirrors. > > That's interesting. I'm pretty sure it actually says "RAID

Re: [PERFORM] Bad planner decision - bitmap scan instead of index

2007-08-16 Thread Frank Schoep
On Aug 16, 2007, at 7:01 PM, Tom Lane wrote: … Why is the sort step so slow? Sorting a mere 13k rows shouldn't take very long. Maybe you are overrunning work_mem and it's falling back to a disk sort ... what is work_mem set to? By default work_mem is set to "1024". Increasing the value to "819

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Steinar H. Gunderson
On Thu, Aug 16, 2007 at 10:53:00AM -0700, Luke Lonergan wrote: > They have a setting that sounds like RAID10, but it actually > implements spanning of mirrors. That's interesting. I'm pretty sure it actually says "RAID10" in the BIOS, but is this a lie? /* Steinar */ -- Homepage: http://www.sess

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Luke Lonergan
Hi Michael, There is a problem with some Dell ³perc 5² RAID cards, specifically we¹ve had this problem with the 2950 as of 6 months ago ­ they do not support RAID10. They have a setting that sounds like RAID10, but it actually implements spanning of mirrors. This means that you will not get more

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Michael Ben-Nes
Thanks for all the answers. It seems its a capable card. Did any one changed the default stripe of 128kb ? On 8/16/07, Steinar H. Gunderson <[EMAIL PROTECTED]> wrote: > > On Thu, Aug 16, 2007 at 11:26:52AM +0300, Michael Ben-Nes wrote: > > Does any one have any experience with RAID 10 & perc 5/i

Re: [PERFORM] Bad planner decision - bitmap scan instead of index

2007-08-16 Thread Tom Lane
Frank Schoep <[EMAIL PROTECTED]> writes: > Limit (cost=4002.04..4002.29 rows=100 width=48) (actual > time=1469.565..1470.097 rows=100 loops=1) > -> Sort (cost=3997.29..4031.18 rows=13556 width=48) (actual > time=1460.958..1467.993 rows=2000 loops=1) > Sort Key: name >

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Steinar H. Gunderson
On Thu, Aug 16, 2007 at 11:26:52AM +0300, Michael Ben-Nes wrote: > Does any one have any experience with RAID 10 & perc 5/i ? Without having done PostgreSQL benchmarking, we have a 2950 with four SATA disks in RAID 10 (and two SAS disks in RAID 1), and have not seen any performance issues. /* Ste

[PERFORM] Bad planner decision - bitmap scan instead of index

2007-08-16 Thread Frank Schoep
Hello everyone, This being my first e-mail to the mailing list, I hope my question is relevant and on-topic. I'm seeing poor performance on a few queries where the planner decides to use a bitmap scan instead of using indices. I'm using a stock PostgreSQL 8.1.9 on Debian 4.0r0 (x86). The

Re: [PERFORM] Integrated perc 5/i

2007-08-16 Thread Merlin Moncure
On 8/16/07, Michael Ben-Nes <[EMAIL PROTECTED]> wrote: > Hi > > I wanted to know if the integrated perc 5/i which come with Dell 2950 will > yield maximum performance from RAID 10 ( 15K SAS ). > Or should I ask for different card ? > > I read an old post that shows that RAID 10 does not work eficen

[PERFORM] Integrated perc 5/i

2007-08-16 Thread Michael Ben-Nes
Hi I wanted to know if the integrated perc 5/i which come with Dell 2950 will yield maximum performance from RAID 10 ( 15K SAS ). Or should I ask for different card ? I read an old post that shows that RAID 10 does not work eficently under perc 5/i http://groups.google.com/group/pgsql.performance