"Apparently (after reading the documentation link provided by Tom) there can
be value to indexes when accessed with leading columns missing".
That is a new one on me too - interesting.
I suppose it comes down to testing at the end of the day - if you "set
enable_seqscan to false" and "EXPLAIN ANA
From: pgsql-general-ow...@postgresql.org
[mailto:pgsql-general-ow...@postgresql.org] On Behalf Of Dann Corbit
Sent: Monday, November 15, 2010 12:21 PM
To: 'Dan Halbert'; pgsql-general@postgresql.org
Subject: Re: [GENERAL] Indexes on individual columns of composite primary key
From: pgs
From: pgsql-general-ow...@postgresql.org
[mailto:pgsql-general-ow...@postgresql.org] On Behalf Of Dan Halbert
Sent: Monday, November 15, 2010 12:01 PM
To: pgsql-general@postgresql.org
Subject: [GENERAL] Indexes on individual columns of composite primary key
I have a table with four columns
Dan,
It depends on your application. There is no point in creating an index with
the same 3 columns in the primary key (in the same order).
If you have an index on COL1, COL2 and COL3 (in that order) then if you have
a query such as SELECT COL1, COL2, COL3 from T1 then the index will be
consider
"Dan Halbert" writes:
> I have a table with four columns. Three of those columns are defined
> as the composite primary key. Does it make sense to create indexes on
> any or all of those three columns individually for performance
> reasons?
Please see
http://www.postgresql.org/docs/9.0/static/ind
I have a table with four columns. Three of those columns are defined as the
composite primary key. Does it make sense to create indexes on any or all of
those three columns individually for performance reasons? PG does let me create
the indexes. But perhaps it's redundant, since there's an impl
