The basis of query optimization is Estimation and Statistics. Without a decent estimation of the amount of rows to be processed, then the entire query plan could well be ‘wrong’. These estimations are built via the statistics and the interactions between them is vital piece of understanding.
With this in mind its always a great pleasure to present my “Estimation and Statistics” session, to spread the knowledge and watch as a few pennies start to drop in the audience whilst revealing the internals. Last week I was able to do this twice at the Cardiff and London legs of the SQL Relay chain of events that tracked it way around the country, a big hat tip to all the organisers and committee members and helpers for putting on a great series of events.
As promised that slides and scripts from my session are now available here :
Hope that you find them useful and if it is still all a bit of a mystery, feel free to mail me : firstname.lastname@example.org