Re: [HACKERS] performance tests, initial results

First, thanks for fixing runtests.pl, David (I'm not so cool perl
programmer as I would like to be -:)

> - Assertion checking seems to cost about 7% except for create index where
> it costs 90%

Wow! This should be discovered!

> - Something is very wrong with our buffer cache implementation. Going from
> 256 buffers to 2048 buffers costs an average of 25%. In the 8K txn case
> it costs about 70%. I see looking at the code and profiling that in the 8K
> txn case this is in BufferSync() which examines all the buffers at commit
^^^^^^^^^^^^^^^^^^^^^^^^
> time. I don't quite understand why it is so costly for the single 8K row
> txn (35%) though.

This one is in my plans for 1 - 1.5 year :)

Vadim