FAQ
Thanks Jasen for your response to my previous question on table
partitioning on query performance.



A related question is how I should actually setup the table partition to
best optimize performance.

Unfortunately, my table of 20 columns will be about 1-1.5TB, much larger
than 20GB.



Specifically, should I be using a range partition or list partition or maybe
both (ie further partitioning a child table?) and how many partitions should
I actually use.



The table contains data for two years on a list of stocks and their stock
attributes (eg prices etc) at a particular date and time.

The majority of the queries run on this table will include the stock code &
date in the where clause.



For eg:



select * from tablename

where stock_code = 'IBM'

and date = '17-Oct-2009'



If I were to partition by month (24 tables over 2 years), I would still have
tables around 45GB. However, I assume there is a balance between performance
and administrative overhead.



Thanks again.

Search Discussions

  • Jasen Betts at Apr 15, 2010 at 10:08 am

    On 2010-04-14, Danny Lo wrote:
    This is a multi-part message in MIME format.

    ------=_NextPart_000_001F_01CADBC3.17BD5470
    Content-Type: text/plain;
    charset="US-ASCII"
    Content-Transfer-Encoding: 7bit

    Thanks Jasen for your response to my previous question on table
    partitioning on query performance.



    A related question is how I should actually setup the table partition to
    best optimize performance.

    Unfortunately, my table of 20 columns will be about 1-1.5TB, much larger
    than 20GB.



    Specifically, should I be using a range partition or list partition or maybe
    both (ie further partitioning a child table?) and how many partitions should
    I actually use.
    that depends on what sort of queries you will use.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouppgsql-novice @
categoriespostgresql
postedApr 14, '10 at 1:10a
activeApr 15, '10 at 10:08a
posts2
users2
websitepostgresql.org
irc#postgresql

2 users in discussion

Jasen Betts: 1 post Danny Lo: 1 post

People

Translate

site design / logo © 2022 Grokbase