Re: 9.2 and index only scans - Mailing list pgsql-general

From Chris Travers
Subject Re: 9.2 and index only scans
Date
Msg-id CAKt_Zfv+Y5T2eW6t1UV2zoZRd42yKbiWCiiBBdfE2hBFN7Y1ig@mail.gmail.com
Whole thread Raw
In response to Re: 9.2 and index only scans  (Craig Ringer <ringerc@ringerc.id.au>)
List pgsql-general


On Tue, Aug 28, 2012 at 6:04 AM, Craig Ringer <ringerc@ringerc.id.au> wrote:
On 08/28/2012 05:51 PM, Thomas Kellerer wrote:
Martijn van Oosterhout, 28.08.2012 10:02:
I'm not sure how oracle avoids the same issues:
- The index has no visibility information, so you can't tell if an
   index entry refers to a row you can actually see in your session.
   The visibility map might help here in the future.

In Oracle an index (entry) has the information about transactional
visibility.

Wow. Doesn't that mean that indexes are insanely expensive to update, since each index (and possibly also the table its self) needs updating?

I was thinking of read performance.  But it might be a case where the global optimization might be worth the local cost.  I don't know.

Best Wishes,
Chris Travers

pgsql-general by date:

Previous
From: "Sahagian, David"
Date:
Subject: order of checking the unique constraints
Next
From: "Marc Mamin"
Date:
Subject: Re: order of checking the unique constraints