| From: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Cc: | Metin Doslu <metin(at)citusdata(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Parallel Select query performance and shared buffers |
| Date: | 2013-12-04 05:10:20 |
| Message-ID: | CAGTBQpZCtDHhifv6m3EJrB5CZu3bnyqYHweM8-0G=cVm-n9V0A@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers pgsql-performance |
On Wed, Dec 4, 2013 at 12:57 AM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>> As a quick side, we also repeated the same experiment on an EC2 instance
>> with 16 CPU cores, and found that the scale out behavior became worse there.
>> (We also tried increasing the shared_buffers to 30 GB. This change
>> completely solved the scaling out problem on this instance type, but hurt
>> our performance on the hi1.4xlarge instances.)
>
> Instead of 30GB, you can try with lesser value, but it should be close
> to your data size.
The OS cache should have provided a similar function.
In fact, larger shared buffers shouldn't have made a difference if the
main I/O pattern are sequential scans, because they use a ring buffer.
Can we have the explain analyze of those queries, postgres
configuration, perhaps vmstat output during execution?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Álvaro Hernández Tortosa | 2013-12-04 06:42:14 | Re: RFC: programmable file format for postgresql.conf |
| Previous Message | Peter Eisentraut | 2013-12-04 04:40:08 | Re: Why we are going to have to go DirectIO |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Metin Doslu | 2013-12-04 12:19:03 | Re: Parallel Select query performance and shared buffers |
| Previous Message | Amit Kapila | 2013-12-04 03:57:18 | Re: Parallel Select query performance and shared buffers |