Re: IOzone testing

[Jeff Layton <laytonjb-fOdFMYwuEsI@public.gmane.org>]

Jérôme Poulin wrote:
> On Sat, Sep 26, 2009 at 8:13 AM, Jeff Layton <laytonjb-fOdFMYwuEsI@public.gmane.org> wrote:
>   
>> Good morning,
>>
>> I'm running IOzone on a NILFS2 partition and it's running alot
>> slower that I had expected. I'm pretty sure it's because of the
>> checkpoints. I can change the protection_period to something
>> very small (or zero). Would that improve the performance?
>>
>>     
> The protection period only affects how the cleaner daemon will clean
> the old checkpoints, if 0 I guess it will just clean them up as soon
> as it can but it won't stop creating them. You could, just for
> benchmark purpose, kill nilfs_cleanerd which will stop checkpoint
> related activities.
>   
It will only stop the clean up of checkpoints outside of policy - correct?
So it won't stop them from being created but it will stop the cleaning 
process.

>> In addition, is there a way to turn off checkpointing?
>>
>>     
> The whole filesystem is based on those checkpoint, the fact that it is
> a log filesystem makes it so all writes are made at the end of the
> device, and when a write is complete, a checkpoint is made.
>   

Understood. But for benchmarks like IOzone where you are creating
files and most likely syncing, this will trigger a checkpoint. I haven't
looked too much at IOzone but it might even be causing a sync after
each record right, again forcing a checkpoint.

So I was looking for some way to control how often checkpoints are
created - something like the protection_period option for the GC. But
I understand that the checkpoint is a part of the design of the file system.

Thanks!

Jeff