Areca vs. ZFS performance testing.

Danny Carroll fbsd at dannysplace.net
Wed Jan 21 05:15:20 PST 2009


Koen Smits wrote:
>     Areca Support:
>     Dear Sir,
>     the only difference is
>     in JBOD mode, controller configure all drives as passthrough disk.
>     in RAID mode, you have to configure passthrough disk by yourself in RAID
>     mode
> 
>     in other words, you can use raid with passthrough disks at saem time in
>     RAID mode but JBOD mode not.
> 
>     Me:
>     So does that mean if I use passthrough, I am not protected by the
>     cache/battery backup?  I ask because there is an option for cache mode
>     when creating a passthrough disk.  i.e. Write-Back or Write-Through
> 
> 
> So 'passthrough' means that the controller lets the OS see the physical
> disks just as they are, but with an invisible cache in between that
> buffers operations. This way there is no advantage of the onboard XOR
> engine, but you do profit from the intelligent cache, which is the most
> important anyway imho.

Not exactly.   In JBOD mode ALL disks are passed through to the OS.  You
cannot have RAID.  The cache is set to Write-Back.

In RAID mode, you can mix raid5, raid6 and Passthrough (which are like
JBOD but allow writethrough or writeback cache at your discretion).

> JBOD mode is at a disadvantage because in this mode the OS sees one
> large drive, and is not able to stripe the data to multiple disks, not
> taking advantage of the fact that you have multple spindles available.
> Makes sense to me :).

No, in JBOD, the OS sees all disks individually.  What you are talking
about is a concatenated disk set which I don't think has a raid level.

> I must admit, I do like these results. Very promising.

Me too, although I am not sure if I like the idea of turning off the
cache flushes in ZFS.  I'd be a lot happier if the Areca card would tell
me how 'full' the cache was.  I'd also love to know if there was a way
for the disk to tell me what the status if it's own cache is.
	

> Further tests would be using an SSD for the ZIL, testing linux and NT,
> etc. But let's not go there ;).

Nope :-)

-D


More information about the freebsd-fs mailing list