Skip site navigation (1) Skip section navigation (2)

Re: Scalability in postgres

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Greg Smith <gsmith(at)gregsmith(dot)com>
Cc: Mark Mielke <mark(at)mark(dot)mielke(dot)cc>, david(at)lang(dot)hm, Scott Carey <scott(at)richrelevance(dot)com>, Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, Dimitri <dimitrik(dot)fr(at)gmail(dot)com>, Flavio Henrique Araque Gurgel <flavio(at)4linux(dot)com(dot)br>, Fabrix <fabrixio1(at)gmail(dot)com>, James Mansion <james(at)mansionfamily(dot)plus(dot)com>, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Scalability in postgres
Date: 2009-06-05 19:48:17
Message-ID: 603c8f070906051248s43b8c8f2q62c350e2fd652c1d@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-performance
On Fri, Jun 5, 2009 at 1:02 PM, Greg Smith<gsmith(at)gregsmith(dot)com> wrote:
> On Fri, 5 Jun 2009, Mark Mielke wrote:
>> I disagree that profiling trumps theory every time.
> That's an interesting theory.  Unfortunately, profiling shows it doesn't
> work that way.

I had a laugh when I read this, but I can see someone being offended
by it.  Hopefully no one took it that way.

> Let's see if I can summarize the state of things a bit better here:
>
> 1) PostgreSQL stops working as efficiently with >1000 active connections
>
> 2) Profiling suggests the first barrier that needs to be resolved to fix
> that is how the snapshots needed to support MVCC are derived
>
> 3) There are multiple patches around that aim to improve that specific
> situation, but only being tested aggressively by one contributor so far
> (that I'm aware of)

I am actually aware of only two forays into this area that have been
reduced to code.  I am pretty much convinced that Jignesh's
wake-all-waiters patch is fundamentally - dare I say theoretically -
unsound, however much it may improve performance for his particular
workload.  The other is Simon's patch which AIUI is a fast-path for
the case where nothing has changed.  Are you aware of any others?

Thanks for the summary.

...Robert

In response to

pgsql-performance by date

Next:From: David BlewettDate: 2009-06-05 21:53:26
Subject: Re: Bad Plan for Questionnaire-Type Query
Previous:From: Robert HaasDate: 2009-06-05 19:42:01
Subject: Re: GiST index performance

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group