Probably like many other's I've wondered why no SSD manufacturer puts a
small BBU on a SSD drive. Triggered by Greg Smith's mail
here, and also anandtech's review at
http://www.anandtech.com/show/2899/1 (see page 6 for pictures of the
capacitor) I ordered a SandForce drive and this week it finally arrived.
And now I have to test it and was wondering about some things like
* How to test for power failure? I thought by running on the same
machine a parallel pgbench setup on two clusters where one runs with
data and wal on a rotating disk, the other on the SSD, both without BBU
controller. Then turn off power. Do that a few times. The problem in
this scenario is that even when the SSD would show not data loss and the
rotating disk would for a few times, a dozen tests without failure isn't
actually proof that the drive can write it's complete buffer to disk
after power failure.
* How long should the power be turned off? A minute? 15 minutes?
* What filesystem to use on the SSD? To minimize writes and maximize
chance for seeing errors I'd choose ext2 here. For the sake of not
comparing apples with pears I'd have to go with ext2 on the rotating
data disk as well.
Do you guys have any more ideas to properly 'feel this disk at its teeth' ?
pgsql-performance by date
|Next:||From: David Boreham||Date: 2010-07-24 07:37:01|
|Subject: Re: Testing Sandforce SSD|
|Previous:||From: Craig Ringer||Date: 2010-07-24 06:36:19|
|Subject: Re: Pooling in Core WAS: Need help in performance tuning.|