Skip navigation.

John Watson's blog

ADVANCED table compression: an improvement

Following previous blog re BASIC compression, here are a couple of simple tests with Advanced Compression - which is supposed to survive conventional DML.

BASIC table compression: use with care

BASIC table compression is bundled with all database editions, and can achieve respectable compression ratios. But beware! Subsequent DML may be disastrous.

How and why to quiesce the database

The ability to quiesce the database was introduced in release 9, but to this day I find that many people are not aware of it. It can be really useful - so let's describe it. I'll begin by positioning the quiesce capability: when is it useful. Then detail how to do it, then reverse engineer the mechanism.

Book review: Brian Peasland, Oracle RAC Performance Tuning

Excellent book - I've just posted this review on Amazon:

Tuning SQL with dynamic sampling

Another example of what I think of as "the self-tuning database". Setting optimizer_dynamic_sampling=11 can fix many performance problems, without the DBA needing to use his brain at all.

Book review: Kevin Meade, Oracle SQL Performance Tuning and Optimization

Excellent exposition of a very effective SQL tuning metodology

Redo and undo volumes - take two

Following a question on OTN https://community.oracle.com/message/12801175 I did another test on redo and undo, just to prove that frequent COMMIT can be bad for performance. The results surprised me. I expected that row-by-row commit would be worse then a single commit at the end of a multi-row transaction, but I hadn't expected it to be this bad. As well as being much slower, both undo and redo volumes are vastly greater.

Redo and undo volumes

How much undo and redo does Oracle generate for different operations? More than you might think.

Using reverse key indexes to solve buffer busy wait problems

Buffer busy wait and related events can cripple performance of concurrent inserts. Bad in a single instance database, far worse in a RAC (think "gc buffer busy"). Often the problem is because of a primary key populated from a sequence. Reversing the index can fix this problem.

Using hash clusters to solve buffer busy wait problems

Concurrent inserts into a table will often result in crippling buffer busy wait problems as sessions serialize on access to the last block(s) of the table segment. Using hash clusters can remove the issue.