From: | Mark Harrison <mh(at)pixar(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | how can I change a btree index into a hash index? |
Date: | 2003-11-12 21:46:42 |
Message-ID: | 3FB2AA42.6090600@pixar.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
I create a table like so:
create table types (
typeid integer unique not null,
typename varchar(255) unique not null
);
and I get the expected messages:
NOTICE: CREATE TABLE / UNIQUE will create implicit index 'types_typeid_key' for table 'types'
NOTICE: CREATE TABLE / UNIQUE will create implicit index 'types_typename_key' for table 'types
Since these are all unique things, and will only be tested for
equality, I am guessing that making a hash index will be better
than making a btree index.
1. Exactly how do I do this? I'm getting some hints that I should
drop a constraint rather than drop an index, but I'm still
not sure exactly what to type.
2. Is this in general a good idea? I will benchmark both ways
but I'm interested in hearing any discussions regarding
hashed indexing.
Many TIA!
Mark
From | Date | Subject | |
---|---|---|---|
Next Message | Joshua D. Drake | 2003-11-12 21:58:20 | Re: serial type vs. sequences |
Previous Message | Andrew Sullivan | 2003-11-12 21:43:03 | Re: [HACKERS] Proposal for a cascaded master-slave replication system |