MySQL Forums
Forum List  »  Partitioning

Re: Partitioning a 250+ million table
Posted by: Rick James
Date: May 16, 2009 05:04PM

You could use a hash partition with userid.

Consider InfoBright's engine -- it has the compression benefit of Archive, plus the ad hoc nature of indexing all columns automatically.

You really have a Data Warehouse application. This often leads to having only one index on the raw data, summarizing the data into other table(s), then doing queries on the summary tables. Show me your queries, I'll be more specific.

Options: ReplyQuote

Written By
May 14, 2009 11:44AM
Re: Partitioning a 250+ million table
May 16, 2009 05:04PM

Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.