Re: Dynamic Partitioning using Segment Visibility Maps

From: Simon Riggs <simon(at)2ndquadrant(dot)com>
To: Markus Schiltknecht <markus(at)bluegap(dot)ch>
Cc: Andrew Sullivan <ajs(at)crankycanuck(dot)ca>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Dynamic Partitioning using Segment Visibility Maps
Date: 2008-01-04 22:01:31
Message-ID: 1199484091.18598.188.camel@ebony.site
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, 2008-01-04 at 22:26 +0100, Markus Schiltknecht wrote:

> I'm still puzzled about how a DBA is expected to figure out which
> segments to mark. Simon, are you assuming we are going to pass on
> segment numbers to the DBA one day?

No Way!

That would stop Richard's idea to make the segment stride configurable,
apart from being a generally ugly thing.

> If not, a more user friendly command like "MARK READ ONLY WHERE date <=
> 2006" would have to move tuples around between segments, so as to be
> able to satisfy the split point exactly, right?

Yes, just a simple WHERE clause that we can translate into segments
under the covers. It would be an alter table, so we get an exclusive
lock.

ALTER TABLE foo SET READ ONLY WHERE ....

possibly with a few restrictions on the WHERE clause. Anyway this is
just futures and dreams, so far, so lets just say something like that is
possible in the future and work out more when we pass the first few
hurdles.

--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Brett Schwarz 2008-01-04 22:07:15 Re: Problem with PgTcl auditing function on trigger
Previous Message Andrew Sullivan 2008-01-04 21:40:45 Re: Dynamic Partitioning using Segment Visibility Maps