Re: Composite keys - Mailing list pgsql-performance

From Tom Lane
Subject Re: Composite keys
Date
Msg-id 5180.1320087559@sss.pgh.pa.us
Whole thread Raw
In response to Re: Composite keys  (Claudio Freire <klaussfreire@gmail.com>)
List pgsql-performance
Claudio Freire <klaussfreire@gmail.com> writes:
> On Mon, Oct 31, 2011 at 2:08 PM, Robert Haas <robertmhaas@gmail.com> wrote:
>>> Multicolumn indices on (c1, c2, ..., cn) can only be used on where
>>> clauses involving c1..ck with k<n.

>> I don't think that's true. �I believe it can be used for a query that
>> only touches, say, c2. �It's just extremely inefficient.

> Does postgres generate those kinds of plans?

Sure it does.  It doesn't usually think they're efficient enough,
because they require full-index scans.  But sometimes that's the
best you can do.

            regards, tom lane

pgsql-performance by date:

Previous
From: Merlin Moncure
Date:
Subject: Re: SSL encryption makes bytea transfer slow
Next
From: Tom Lane
Date:
Subject: Re: does update of column with no relation imply a relation check of other column?