Continuing this month's theme of Temporal Data (Time Travel Query), IDUG Content Committee chair Dan Luksetich goes behind the scenes to look at how DB2 processes temporal queries. Check out his article here -
The key message is that while temporal SQL constructs makes writing complex time series queries easier, you really need to understand and consider the performance implications, especially when it comes to queries which need to access both the base and history tables of a system temporal implementation. As someone said to me recently "magic doesn't just happen". It doesn't mean you shouldn't use these constructs. You just need to allow for the processing involved and tune accordingly. In some performance critical situations it may be better to code standard SQL statements for optimum performance.
Once again Dan offers us real insights into DB2 performance.