Re: HOT patch, missing things - Mailing list pgsql-hackers

From Pavan Deolasee
Subject Re: HOT patch, missing things
Date
Msg-id 2e78013d0708090048w4142e912seba3db54a74ebb35@mail.gmail.com
Whole thread Raw
In response to HOT patch, missing things  (Heikki Linnakangas <heikki@enterprisedb.com>)
Responses Re: HOT patch, missing things  (Gregory Stark <stark@enterprisedb.com>)
Re: HOT patch, missing things  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers


On 8/7/07, Heikki Linnakangas <heikki@enterprisedb.com> wrote:
There's three things clearly missing in the patch:

1. HOT updates on tables with expression or partial indexes. Hasn't been
done yet because it should be pretty straightforward and we've had more
important things to do. Though not critical, should be finished before
release in my opinion.



I started with this. ISTM to support partial indexes, we must check
the old and new tuple against partiality match.

HOT update is feasible iff

- old and new tuples, both match the partiality condition OR
- old and new tuples, both don't match the condition

In either case, we either had an index entry which can serve for
the new tuple OR we did not have an index entry for the old tuple,
but neither the new tuple needs it.

Of course, we still need to apply all other criteria to finally decide
whether to do HOT or COLD update.

For functional index, we should apply the function to the old and new
tuple and compare the outcome. If the results are same, HOT update
is feasible.

We still need to think about the best way to do this without any
modularity invasion and least possible overhead, but can anybody
see any issue with the broader approach ?

Thanks,
Pavan

--
Pavan Deolasee
EnterpriseDB     http://www.enterprisedb.com

pgsql-hackers by date:

Previous
From: "Adrian Maier"
Date:
Subject: Compilation of pg 7.4.17 fails on HP-UX
Next
From: tomas@tuxteam.de
Date:
Subject: Re: default_text_search_config and expression indexes