One of the selling points of DB2 10.5 and the Columnar options is that existing row-based tables can be converted to column-based tables and will be accessed by DB2 seamlessly and with no extra effort from the DBA. All well and good, and the performance should be much better, but what if it’s not? What if you wanted to find what is happening behind the scenes? A new CTQ plan operator represents the transition between column-organized data processing and row-organized data processing. We will look at the access path of a query that includes columnar data, and try and find out what the optimizer is doing “under the covers” .
Check out this IDUG Tech Talk full length video on YouTube here