Re: pg_dump is O(N) in DB table count N even if dumping only one table - Mailing list pgsql-bugs

From Gunnlaugur Thor Briem
Subject Re: pg_dump is O(N) in DB table count N even if dumping only one table
Date
Msg-id CAPs+M8L2qA1HjbQqrK7B9wGDdFbwrQPbCOjHKWNdwY5eioVkjQ@mail.gmail.com
Whole thread Raw
In response to Re: pg_dump is O(N) in DB table count N even if dumping only one table  (Andres Freund <andres@2ndquadrant.com>)
Responses Re: pg_dump is O(N) in DB table count N even if dumping only one table
List pgsql-bugs
On Mon, Jun 10, 2013 at 2:08 PM, Andres Freund <andres@2ndquadrant.com>wrote:

> On 2013-06-10 13:28:32 +0000, Gunnlaugur Thor Briem wrote:
> > The three queries taking O(N) time are listed below. AFAICT each of these
>
> Which of those queries take how long in your case?
>

They were as follows:

duration: 27770.917 ms  statement: SELECT c.tableoid, c.oid, c.relname, ...
duration: 28133.407 ms  statement: SELECT tableoid, oid, typname, ...
duration: 55751.996 ms  statement: SELECT classid, objid, refclassid, ...

However, that was running pg_dump remotely over a slow network, and that
may exaggerate durations since it's a bunch of data. So I ran this again
locally on the production server:

duration: 16012.518 ms  statement: SELECT c.tableoid, c.oid, c.relname, ...
duration: 16516.708 ms  statement: SELECT tableoid, oid, typname, ...
duration: 13400.694 ms  statement: SELECT classid, objid, refclassid, ...

Duration decrease is probably *not* significantly due to a warmer cache,
because I had two runs over the slow network and their durations were
similar (actually longer the second time).

Gulli

pgsql-bugs by date:

Previous
From: Gunnlaugur Thor Briem
Date:
Subject: Re: pg_dump is O(N) in DB table count N even if dumping only one table
Next
From: Andres Freund
Date:
Subject: Re: pg_dump is O(N) in DB table count N even if dumping only one table