To roll out table partitioning in your deployment, you need to identify the tables to partition, and then identify the column or columns to partition on. You then need to determine the partition type to use, list or range, and the values for the columns defining what data will be directed to the different partitions. Additionally, you need to identify the location to store the data for each of the defined partitions.
There are several things to consider when identifying a table as being a good candidate for partitioning. Partitioning a table should not be based solely on the size of the table. Additionally, your partitioning scheme will serve you better if you spend a little time to investigate partitioning deeper than simply employing the all too common “just partition on the date field” strategy. The sections that follow ask general questions about your data to help you identify tables that could be good candidates for table partitioning. The more aspects in the sections below that pertain to your existing tables, the better candidate the table is for partitioning.
