Re: _mdfd_getseg can be expensive - Mailing list pgsql-hackers

From Andres Freund
Subject Re: _mdfd_getseg can be expensive
Date
Msg-id 20160701020851.4ln3bdjwjhfmivje@alap3.anarazel.de
Whole thread Raw
In response to Re: _mdfd_getseg can be expensive  (Peter Geoghegan <pg@heroku.com>)
Responses Re: _mdfd_getseg can be expensive  (Peter Geoghegan <pg@heroku.com>)
List pgsql-hackers
On 2016-06-30 18:34:20 -0700, Peter Geoghegan wrote:
> On Thu, Jun 30, 2016 at 6:23 PM, Andres Freund <andres@anarazel.de> wrote:
> > I plan to, once the tree opens again. Likely needs some considerable
> > updates for recent changes.
> 
> Offhand, do you think that CREATE INDEX calls to smgrextend() could be
> appreciably affected by this bottleneck? If that's a very involved or
> difficult question, then no need to answer now.

If you have a big enough index (maybe ~150GB+), sure. Before that,
probably not.

It's usually pretty easy to see in cpu profiles whether this issue
exists.

Andres



pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: Is a UDF binary portable across different minor releases and PostgreSQL distributions?
Next
From: Michael Paquier
Date:
Subject: Re: OpenSSL 1.1 breaks configure and more