Re: [PATCH 004 of 11] md: Increase the delay before marking metadata clean, and make it configurable.

[Neil Brown]

On Tuesday May 2, bert.hubert@xxxxxxxxxxxxx wrote:
> On Mon, May 01, 2006 at 03:30:19PM +1000, NeilBrown wrote:
> > When a md array has been idle (no writes) for 20msecs it is marked as
> > 'clean'.  This delay turns out to be too short for some real
> > workloads.  So increase it to 200msec (the time to update the metadata
> > should be a tiny fraction of that) and make it sysfs-configurable.
> 
> What does this mean, 'too short'? What happens in that case, backing block
> devices are still busy writing? When making this configurable, the help text
> better explain what the trade offs are.

"too short" means that the update happens often enough to cause a
noticeable performance degradation.

In an application writes steadily very 21msecs (or maybe 30msecs) then
there will be 2 superblock writes and 1 application write every
21msecs, and this causes enough disk io to close the app down. - I
guess all the updates fill up the 21msec space.

With a larger delay - 200msec - you could still get bad situations
e.g. with the app writing every 210msecs.  However 2 superblock
updates plus one app write is a much smaller fraction of 200msecs, so
there shouldn't be as many problems.

Yes, a more detailed explanation should go in Documentation/md.txt

NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/