Re: Performance issues with large amounts of time-series data

From: Greg Smith <gsmith(at)gregsmith(dot)com>
To: Hrishikesh (हृषीकेश मेहेंदळे) <hashinclude(at)gmail(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Performance issues with large amounts of time-series data
Date: 2009-08-27 22:57:47
Message-ID: alpine.GSO.2.01.0908271853580.19917@westnet.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Wed, 26 Aug 2009, Hrishikesh (??????? ????????) wrote:

> key = {device_id (uint64), identifier (uint32), sub_identifier (uint32), unix_time} (these four taken together are unique)

You should probably tag these fields as NOT NULL to eliminate needing to
consider that possibility during query planning. As of V8.3 this isn't as
critical anymore, but it's still good practice.

--
* Greg Smith gsmith(at)gregsmith(dot)com http://www.gregsmith.com Baltimore, MD

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message bricklen 2009-08-27 23:02:15 Vacuum duration + hint bits?
Previous Message Jean-Michel Pouré 2009-08-27 22:31:51 Re: PostgreSQL does CAST implicitely between int and a domain derived from int