Re: One query run twice in parallel results in huge performance decrease

From: Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>
To: Jan Michel <Jan(at)mueschelsoft(dot)de>
Cc: "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
Subject: Re: One query run twice in parallel results in huge performance decrease
Date: 2013-12-03 01:59:46
Message-ID: CAOR=d=3H5LZhVVx+2eC+DTTpCC8ksM5GCTaoA22eJ62eNwEqaA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Fri, Nov 29, 2013 at 3:07 PM, Jan Michel <Jan(at)mueschelsoft(dot)de> wrote:
> Dear all,
> I have a quite strange problem running an extensive query on geo data
> checking for crossing ways. I don't know if this is a postgres or postgis
> problem, but I hope you can help. Running one thread is no problem, it
> finishes within 10-15 minutes. Run two of those queries in parallel and
> they will not finish within 24 hours. It is definitely not a caching or I/O
> problem.

What does your IO subsystem look like when you're running the query
both once and twice?

iostat, vmstat, iotop etc area ll useful here.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Metin Doslu 2013-12-03 13:41:43 Parallel Select query performance and shared buffers
Previous Message Jeff Janes 2013-12-02 23:17:16 Re: One query run twice in parallel results in huge performance decrease