Re: [PATCHES] updated hash functions for postgresql v1

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Kenneth Marshall <ktm(at)rice(dot)edu>
Cc: Hannu Krosing <hannu(at)2ndQuadrant(dot)com>, Asko Oja <ascoja(at)gmail(dot)com>, Jeff Davis <pgsql(at)j-davis(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: [PATCHES] updated hash functions for postgresql v1
Date: 2009-10-28 19:57:12
Message-ID: 4236.1256759832@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Kenneth Marshall <ktm(at)rice(dot)edu> writes:
> On Wed, Oct 28, 2009 at 03:31:17PM -0400, Tom Lane wrote:
>> Hash indexes are so far from being production-grade that this argument
>> is not significant.

> In addition that change from 8.3 -> 8.4 to store only the hash and not
> the value in the index means that a reindex would be required in any event.

Indeed, and I fully expect there will be some more on-disk format
changes required before we get to the point where hash indexes are
actually interesting for production. If we start insisting that they
be in-place-upgradable now, we will pretty much guarantee that they
never become useful enough to justify the restriction :-(

(As examples, the hash bucket size probably needs revisiting,
and we ought to think very hard about whether we shouldn't switch
to 64-bit hash values. And that's not even considering some of the
more advanced suggestions that have been made.)

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Greg Smith 2009-10-28 20:08:19 Re: Parsing config files in a directory
Previous Message Jeff Davis 2009-10-28 19:51:49 Re: [PATCHES] updated hash functions for postgresql v1