FreeNAS Performance: Part 3

I recently storage-VMotion’d all of the lab home lab VMs over to FreeNAS based storage. It is an NFS share based on 4 10K 146G spindles configured for RAIDZ – deduplication is turned off, instead opting for lzjb compression. The physical specs of the FreeNAS host are the same as in this post.

Keep in mind that there are several other VMs banging away at these 4 disks. That said, 75% write\25% read (75% random) is considerably lower for this configuration. Again – this is not the final configuration – no SSDs for ZIL, RAIDZ (RAID5) instead of a mirror (RAID1), and more RAM.

200 IOPS isn’t terrible, but it’s not great. It’s plenty for my home lab (for now).

Just to compare and show that the L1ARC (RAM) cache is doing its job, I configured a 100% read specification (75% random).

2355 read IOPS is pretty good for these disks – but you can clearly see that it is pretty much all coming out of the RAM cache:

[root@filer] /usr/local/www/freenasUI/tools# ./arcstat.py -f read,hits,miss,hit%,arcsz 1

More testing to continue…

4 thoughts on “FreeNAS Performance: Part 3”

  1. [quote]It is an NFS share based on 4 10K 146G spindles configured for RAIDZ[/quote]

    As an FYI, a 4 disk array is not optimal for ZFS, please see:

    http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#RAIDZ_Configuration_Requirements_and_Recommendations

    A RAIDZ configuration with N disks of size X with P parity disks can hold approximately (N-P)*X bytes and can withstand P device(s) failing before data integrity is compromised.

    Start a single-parity RAIDZ (raidz) configuration at 3 disks (2+1)
    Start a double-parity RAIDZ (raidz2) configuration at 6 disks (4+2)
    Start a triple-parity RAIDZ (raidz3) configuration at 9 disks (6+3)
    (N+P) with P = 1 (raidz), 2 (raidz2), or 3 (raidz3) and N equals 2, 4, or 6
    The recommended number of disks per group is between 3 and 9. If you have more disks, use multiple groups.

    Reply
  2. sbenji:

    I have read the RAIDZ best practices and agree that the current setup is not optimal – nor is it permanent. The final configuration will be several mirrored pairs of 10K disks.

    Reply
  3. Seems a bit odd to benchmark against a sub-optimal setup, i'd be much more interested on the numbers from a single parity RAIDZ by dropping one of the disks

    Reply
  4. I was looking for FreeNAS LZ4 benchmarks and came across this. I think that benchmarking an expected sub-optimal configuration provides a good base line for determining that your next 'optimal' configuration is actually better. I did this with a 4-disk RZ when I started with ZFS too, and it was interesting to compare a 4-disk RZ vs. 2×2-way mirror. Depending on your hardware and workload, it isn't very different, nor would I expect it to be until the spindle count is higher.

    Reply

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.