Call: +44 (0)1904 557620 Call

Pete Finnigan's Oracle Security Weblog

This is the weblog for Pete Finnigan. Pete works in the area of Oracle security and he specialises in auditing Oracle databases for security issues. This weblog is aimed squarely at those interested in the security of their Oracle databases.

[Previous entry: "Tension between security vendors, bug hunters continues to simmer"] [Next entry: "The Best of Oracle Security 2006 (in German)"]

Pete Finnigan's UKOUG presentation on FGA, VPD and audit performance

It has been a while since the UKOUG finished but if you were there and spoke to me you would remember that I had a very bad cold and cough, well the bad news is that I have still got it nearly 5 weeks after getting it. I have taken all manner of cold products, cough mixture, tablets, powders, anti-biotics last week and now codine phosphate. This is the worst cold i have ever had so its laid me up somewhat. I have a big backlog of links and posts I want to make and also quite a few updates and changes to my site. I am also involved in the development of GreyMatter Blog software and participate in the forum that is used for this sites blogs so my ill health has stopped me on my current project to build some comment moderation features into Greymatter, which I would like to get into the next planned release so that i can turn on comments here finally.

Anyway back to the main subject of the post. At the UKOUG in Birmingham I talked about the issues of performance degredation when Audit, VPD and FGA are implemented in a database. This is a common issue and an important one as a lot of sites don't use audit and a lot of people think that these technologies simply kill the database. This is true in some cases but what I wanted to concentrate on was the important task of designing and planning so that you "tune the algorithm" rather than the technology. I have read books in the past by Michael Abrash famous as a games developer in C and assembler and he gives some goood lessons. I remember one chapter or article where he showed a program, did every kind of tuning you could think to it and it went faster but not blindingly so. Then he turned to the algorithm, tuned the algorithm and the increase in performance was astronomic. This is the sense I wanted to cover in my presentation titled "Does VPD, FGA or audit really cause performance issues?"