[OpenIndiana-discuss] ZIL write cache performance

[Jan Owoc]

On Thu, Jan 12, 2012 at 3:55 PM, Matt Connolly
<matt.connolly.au at gmail.com> wrote:
> Hi, I've installed an SSD drive in my OI machine and have it partitioned (sliced) with a main slice to boot from and a smaller slice to use as a write cache (ZIL) for our data pool.
>
> I've noticed that for many tasks, using the ZIL actually slows many tasks at hand (operation within a qemu-kvm virtual machine, mysql loading importing a dump file, etc). I know I bought a cheap SSD to play with so I wasn't expected the best performance, but I would have expected some improvement, not a slow down.

Adding a ZIL device means that synchronous writes (apparently your
workload has some) hit the ZIL first. Cheap SSDs have poor (in some
cases very poor) write speed.


> In one particular test, I have mysql running in a zone and loading a test data set takes about 40 seconds without the ZIL and about 60 seconds with ZIL. I certainly wasn't expecting a 50% slow down.
>
> Is this to be expected?
>
> Are there any best practices for testing an SSD to see if it will actually improve performance of a zfs pool?

The change (if any) depends on your workload. I would, however, look
for an SSD with a high write speed (you'll likely be writing to a ZIL
a lot more than reading). If you are able to quantify what is the
typical transfer size in your workload, you could look at SSDs with
good write speeds for 4k writes (or 512k writes).


Jan