Re: performance problems with bulk inserts/updates on tsrange with gist-based exclude constrains

From: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
To: Chris Withers <chris(at)simplistix(dot)co(dot)uk>
Cc: pgsql-general <pgsql-general(at)postgresql(dot)org>
Subject: Re: performance problems with bulk inserts/updates on tsrange with gist-based exclude constrains
Date: 2016-09-19 16:41:33
Message-ID: CAMkU=1zgVjQZMUikVZg9=0M7E9k-s=aWN8EHtqkAbPdo4cukcw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Fri, Sep 16, 2016 at 2:01 AM, Chris Withers <chris(at)simplistix(dot)co(dot)uk>
wrote:

> Hi All,
>
> I have quite a few tables that follow a pattern like this:
>
> Table "public.my_model"
> Column | Type | Modifiers
> --------+-------------------+-----------
> period | tsrange | not null
> key | character varying | not null
> value | integer |
> Indexes:
> "my_model_pkey" PRIMARY KEY, btree (period, key)
> "my_model_period_key_excl" EXCLUDE USING gist (period WITH &&, key
> WITH =)
> Check constraints:
> "my_model_period_check" CHECK (period <> 'empty'::tsrange)
>

Try swapping the order of the columns in the exclude constraint. You want
the more selective criterion to appear first in the index/constraint.
Presumably "key with =" is the most selective, especially if many of your
periods are unbounded.

Cheers,

Jeff

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Jeff Janes 2016-09-19 16:55:15 Re: Index scan is not working
Previous Message Paul Jungwirth 2016-09-19 16:26:36 Re: journaling / time travel