Hello All,
I am beginning to partition some database tables and need some help thinking things through. After evaluating my data, it appears that the best partition would be based on year and month. I was planning to create my tables in the format of tablename_yyyy_mm.
I planned to create a parent table and child tables (similar to the example at http://www.postgresql.org/docs/8.1/static/ddl-partitioning.html). I would create the child tables for the range of my data and maybe even a few months in advance. I have all of the insert, update, and delete rules worked out to manage the data. The data will be routed to tables based on a date that is passed to the table (e.g. start_time). The date can be in the future or in the past.
The problem is that the databases are running on remote servers and I will not have access to them. I need to find a way to automatically create the child tables, indexes, and rules. Any suggestions would be helpful. I am stuck.
Thanks,
Steve