Re: Planner question - "bit" data types - Mailing list pgsql-performance

From Karl Denninger
Subject Re: Planner question - "bit" data types
Date
Msg-id 4AA2DAA1.1030006@denninger.net
Whole thread Raw
In response to Re: Planner question - "bit" data types  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Planner question - "bit" data types  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-performance
Tom Lane wrote:
Karl Denninger <karl@denninger.net> writes: 
That doesn't help in this case as the returned set will typically be
quite large, with the condition typically being valid on anywhere from
10-80% of the returned tuples.   
In that case you'd be wasting your time to get it to use an index
for the condition anyway.  Maybe you need to take a step back and
look at the query as a whole rather than focus on this particular
condition.
		regards, tom lane
 
The query, sans this condition, is extremely fast and contains a LOT of other conditions (none of which cause trouble.)

It is only attempting to filter the returned tuples on the permission bit(s) involved that cause trouble.

-- Karl
Attachment

pgsql-performance by date:

Previous
From: Tom Lane
Date:
Subject: Re: Planner question - "bit" data types
Next
From: Tom Lane
Date:
Subject: Re: Planner question - "bit" data types