I've already chosen granularity for samples. 5 minute samples for example can expire after two days.
Aggregated/summarized rows of this data (30 minute sample for example, which is the aggregation of the past 30 minutes worth of 5 minute samples in a given window) expire after a week, etc. I'm more concerned as to why inserts begin to slow down so much due to the large table size. -----Original Message----- From: Wallace Reis [mailto:[EMAIL PROTECTED] Sent: Friday, 27 July 2007 1:02 AM To: Andrew Armstrong Cc: mysql@lists.mysql.com Subject: Re: Data Warehousing and MySQL vs PostgreSQL On 7/26/07, Andrew Armstrong <[EMAIL PROTECTED]> wrote: > Do you have a suggestion to how this should be implemented? > Data is aggregated over time and summary rows are created. I think that you didnt design correctly your DW. It should have just one very larger table (the fact table). Data should never be deleted. If your client want to query data about 'five minutes sample' when they are already expired? You should decide the data's granularity. And if you want to agregate them, do roll up. Or you can create materialized views for these aggregates. -- wallace reis/wreis Núcleo de Biologia Computacional e Gestão de Informações Biotecnológicas/LABBI -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
