Re: ASM Filter Driver Performance
- From: Mladen Gogala <gogala.mladen@xxxxxxxxx>
- To: Jared Still <jkstill@xxxxxxxxx>
- Date: Fri, 26 Jan 2018 06:36:51 -0500
On 01/25/2018 08:36 PM, Jared Still wrote:
Interesting that there would be an expectation for me (or anyone) to
be able to explain just what Oracle Developers did to improve their code.
Sorry, I cannot accept this. If such an improvement is made, I need to
know just the basic facts about how and why. It's not your fault, Oracle
simply doesn't produce the necessary documentation any more.
As I have already written, ASM Filter Driver is very new. Any bugs
in the driver mechanism can bring your instance, and therefore
your database, down. Such outlandish claims do not convince me on
the merit of using filter driver.
Outlandish claims?
No claims were made, just results of performance tests.
Claim is that there is 20% improvement. And I find that number very
suspicious without some background information. No disrespect to you,
but what would that say about the RAC as a product if 20% performance
improvement can be achieved by simply changing the disk drivers?
As for "more details on Hotsos and IOUG", I am awaiting for a MOS
document which would explain better performance and the basic principles
behind achieving it. ASMLib is a raw disk driver. The only way such a
large performance improvement can be made is to use block devices and OS
cache, which has a read-ahead and write back mechanisms. That, on the
other hand, has ramifications for the commit processing and ACID
compliance. This is not just a performance question, we are talking
about the basic principles here. As you said, filter drivers are very
new. And I will ignore them until some reliable documentation is
produced. The performance numbers you produced are alarming.
--
Mladen Gogala
Database Consultant
Tel: (347) 321-1217
Other related posts: