How not to Build a(n Autonomous) Data Warehouse
David Kurtz, Accenture Enkitec Group
There are features in the Oracle Database that are specifically designed to optimize performance of data warehouse queries. If your data model is correct, and therefore you built your data warehouse according the generally accepted principles of good data warehouse design, then you should naturally take advantage of them. If you do not, then the database is likely to have to do a lot more work to execute your queries. In this session, we will look at star schemas, star transformation and bloom filters. We will look at how they work, and by modelling examples taken from real life, what sort of thing is often done that stop the database using these optimisations. It will look at how the approach changes on Engineered Systems, and we will also look at what happens inside Oracle's Autonomous Data Warehouse Cloud (ADWC) that is built on Engineered Systems, but is slightly different again.
David Kurtz has worked with the Oracle database since 1989 and with PeopleSoft ERP applications since 1996, specialising in system performance tuning. He now works for the Accenture Enkitec Group as a performance tuning consultant. David is a regular presenter at Oracle and PeopleSoft conferences. He is the author of PeopleSoft for the Oracle DBA (www.psftdba.com), and blogs about PeopleSoft (blog.psftdba.com) and Oracle (blog.go-faster.co.uk). David is an Oracle ACE Director and a proud member of the Oak Table.