Re: Shared buffers, db transactions commited, and write IO on Solaris

[Erik Jones <erik@xxxxxxxxxx>]

On Mar 29, 2007, at 7:23 PM, Josh Berkus wrote:

Erik,

Wow, thanks for the post.

We've just started testing the option of sizing shared_buffers bigger than 

the database, and using forcedirectio in benchmarks at Sun.  So far, our 

experience has been *equal* performance in that configuration, so it's 

*very* interesting to see you're getting a gain.

-- 

--Josh

Josh,

You'er welcome!  However, I believe our situation is very different from what you're testing if I understand you correctly.  Are you saying that you're entire database will fit in memory?  If so, then these are very different situations as there is no way ours could ever do that.  In fact, I'm not sure that forcedirectio would really net you any gain in that situation as the IO service time will be basically nil if the filesystem cache doesn't have to page which I would think is why your seeing what you are.

erik jones <erik@xxxxxxxxxx>

software developer

615-296-0838

emma(r)