checkpoints to often

checkpoints to often

[krzf83@gmail.com]

There should be a way to make automatic checkpoints less frequent. On
the busy ssd I have about 7 cp every second. If it were for example
once every 5 minutes then one could set garbage removal every few
days.
Also garbage removal should have option to clean only if like 90% of
drive is used. there is no need to remove cp-s when there is still
many GB of space left.

Re: checkpoints to often

[Chris Mason]

Excerpts from krzf83@gmail.com's message of 2011-03-31 10:30:51 -0400:
> There should be a way to make automatic checkpoints less frequent. On
> the busy ssd I have about 7 cp every second. If it were for example
> once every 5 minutes then one could set garbage removal every few
> days.
> Also garbage removal should have option to clean only if like 90% of
> drive is used. there is no need to remove cp-s when there is still
> many GB of space left.

I'm wondering how you're measuring check points?  Btrfs does a
transaction commit ever 30 seconds, so finding 7 commits per second
seems strange.

We do have a lighter weight commit that is done for fsyncs, but this is
not a full check point.

There's no garbage removal in btrfs, we use reference counting instead
of a traditional log structured design.

-chris