Tuesday, 25 March 2008

Collab08: The Aussie DBAs are coming! Tony Jambu and Penny Cookson

Australians have a reputation for being down to earth, hard working sorts who don't ponce about, and I'm happy to advertise two of Australia's best Oracle DBA zero-poncers will be presenting at Collaborate 08 this year, Penny Cookson and Tony Jambu. I recommend you check out their presentations:

Penny Cookson: New Tuning Features in Oracle 11g: How to make managing your database free of stress

Penny was awarded the Oracle Educator of the Year 2004 by Oracle Magazine, Oracle ACE hood in 2008, and has won numerous best paper awards.

Abstract: One of the key problems that has haunted Oracle sites since the introduction of the cost based optimiser is the ability to provide a stable level of performance over time. The very responsiveness of the CBO to factors such as changes in statistics and initialisation parameters can lead to sudden changes in performance levels. Oracle 11g introduces a number of features that will assist the DBA in providing a stable environment for mission critical applications. Excitement is for out of work time, not for managing production databases. This presentation will discuss the best use of these new tuning features to reduce the stress levels of DBAs.

Tony Jambu: Understanding Oracle's Histograms

Tony was awarded Oracle ACE-hood in 2007, Oracle Magazine Consultant of the Year in 2003, and is a well known and popular presenter on the Oracle scene in Australia.

Abstract: Oracle’s histogram is one of the most misunderstood and least used feature. On top of that, there are a number of myths regarding its use. This presentation will look under the cover of Oracle’s histogram.

It will cover the basics e.g. from data distribution, density and cardinality to how to generate accurate histogram and tuning using histograms.

For the experienced DBAs, the presentation will dig into the Oracle dictionary as well as analysing CBO’s trace files.

  • Understanding of what a histogram is and is not
  • How to gather appropriate histogram data
  • Tuning with proper appropriate histogram data

No comments: