Here is a link to my updated Exadata PowerPoint presentation.
I’ve been trying to improve my Exadata talk for the Collaborate 13 conference. I’ve done this talk at work and at the ECO conference in October but I’m not completely happy with it. The criticism I’ve received boils down to these things:
- The slides don’t stand on their own
- I jump into the middle of the detail without enough introduction
So, I’ve updated the slides to have comments on many of the slides. This will be good as a reminder of what I want to say and to make the slides more meaningful to someone who just has the PowerPoint.
I’ve also added several slides to show the execution plan of a sample query to do a better job of setting up the slides I already have that discuss how data flows through an Exadata server as compared with a normal server. The point is that when the table is accessed blocks are read and certain rows are selected and certain columns are projected.
Lastly I added a slide on direct path read which shows how the buffer cache can be bypassed even on a normal Oracle database server in some cases. This is also part of the introduction in that it provides background needed to understand how the Exadata Smart Scan bypasses the buffer cache.
If anyone has time to read through the slides and give me their feedback I’d be happy to hear it. Hopefully the net result will be a presentation that is both useful to me and my company and to those attending the conference.
P.S. Edited again 3/22/2013