Because it is snapshot type data, Tivoli Data Warehouse (TDW) data works well for trending and analysis. TDW data is not necessarily equivalent to SMF data, which tends to be more interval or event driven in nature. TDW is not equivalent to CICS SMF 110's or DB2 Accountig trace data. But, if you want to be able to do things like trend resource usage (such as CPU, DASD, or memory) over time, TDW can work well in that context. Also, when doing trending analysis, the TDW summarization process can add some interesting information to the data.
Here's an example of what I'm talking about. In this example, I'm trending z/OS CPU over time using CPU TDW snapshot data summarized by hour. Note that in addition to average CPU usage you also get a Min and Max CPU usage for the same time interval, and here I show how easy it is to chart that information on a single graph.