There’s never enough time to read everything that’s worth reading, so even though Guy Harrison’s blog is one of the ones worth reading I find that it’s often months since I last read it. Visiting it late last night, I found an interesting batch of articles spread over the last year about the performance of SSD – the conclusions may not be what you expect, but make sure you read all the articles or you might end up with a completely misleading impression:
- Using flash disk for redo on Exadata
- Using SSD for redo on Exadata – pt. 2
- Exadata smart flash logging
- Exadata smart flash logging – outliers
Don’t forget to read the comments as well. For other notes Guy has written about SSD, here’s a URL for his SSD tag.
I’m not too surprised about Guy’s conclusions about redo on SSD.
This is not because I am expert on SSD (I can spell it) but because I attended a presentation a couple of years ago by a couple of engineers from LSI.
Their job was to do performance testing of the new LSI SSD product for Oracle Flashcache.
They were surprised to consistently find that redo performed better on spinning rust than on SSD.
After discussing it with some other engineers they had a better understanding of the limitations of SSD when used with redo.
The Powerpoint for that presentation is here:
http://www.oswoug.org/Slides/LSI/SolidStateStorageinOracleEnvironmentsv4.pptx
The most interesting redo bits aren’t available in the presention, you had to be there.
There is mention however that due to sequential writes redo performs better on HDD.
Comment by Jared — October 5, 2012 @ 1:47 pm UTC Oct 5,2012 |