Multi node storage, ZFS

[Michael Loftis]

--On Wednesday, March 24, 2010 5:12 PM +0000 Michal <michal at ionic.co.uk> 
wrote:

>
>> If you were to do something like this, I'd make sure to have a fast
>> local ZIL (log) device on the head node.  That would reduce latency
>> for writes, you might also do the same for reads.  Then your bulk
>> storage comes from the iSCSI boxes.
>>
>> Just a thought.
>
> I've not come across ZIL so I think I will have to do my research

ZFS Intent Log, basically ZFS's WAL (Write Ahead Log) -- write committed to 
the ZIL are considered durable, and the system then batches up ZIL writes 
to normal storage.  For reads there's the cache devices.  I honestly do not 
know the state of this in FreeBSD, I gave up using ZFS on FreeBSD for now 
due to poor performance.  Also the *linux* iSCSI kernel initiator is 
*really* buggy, can't say anything about FreeBSD iSCSI initiator nor 
target, nor anything about the Linux iSCSI target.

By using (fast) ZFS log and ZFS cache devices that are local to the 'san 
head end' you can *greatly* increase the array's perceived/usable speed.

>
>
>> At least in theory you could use geom_gate and zfs I suppose, never
>> tried it though.
>> ggatec(8), ggated(8) are your friends for that.
>>
>> Vince
>
> Just had a look at ggatec, I've not seen or heard of that so I will
> continue looking through that.
>
>
> Many thanks to all, if I get something solid working I will be sure to
> update the list with what will hopefully be a very cheap (other then
> HDD's) SAN
>
>
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"