Re: Force specific index disuse - Mailing list pgsql-general

From Steve Crawford
Subject Re: Force specific index disuse
Date
Msg-id 537BA26D.20106@pinpointresearch.com
Whole thread Raw
In response to Re: Force specific index disuse  (Seamus Abshere <seamus@abshere.net>)
List pgsql-general
On 05/20/2014 09:44 AM, Seamus Abshere wrote:
> On 5/20/14, 1:38 PM, Steve Crawford wrote:
>> Is there a way to force a specific index to be removed from
>> consideration in planning a single query?
>
> hi Steve,
>
> What is the query? Or at least a sanitized but complete version?

I've now resolved the issue with the one update query I was
investigating this morning. But this involved building a test where I
removed the index then ran explain and timed the query.

The question is actually general as I anticipate reviewing the benefit
of dropping more indexes and it would be much more quick and convenient
to do something akin to:

begin;
disable index foodex;
explain update bar set baz ....;
commit;

I'm not sure what would be involved in adding this. It seems that simply
hiding an index from the planner would be all that is necessary but I'm
sure there are, as always, subtleties.

Cheers,
Steve



pgsql-general by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: Force specific index disuse
Next
From: Steve Crawford
Date:
Subject: Re: Force specific index disuse