Loading and managing large data sets in your Data Warehouse is not always a trivial task. Especially if the source systems can re-deliver data and you want to replace an old subset of your data. The quick answer you always hear is “just use partitioning”, but nobody really tells you how you should design or implement it. This talk shows one way to load and manage your very large data sets, by using layered partitioning. We will walk through partitioned tables and partitioned views, before moving on to the concept of layered partitioning.
The talk was presented at SQLBits XI.
Video is available at the SQLBits website.