Hello all,
I''m building a file server (or just a storage that I intend to access
by Workgroup from primarily Windows machines) using zfs raidz2 and openindiana
148. I will be using this to stream blu-ray movies and other media, so I will be
happy if I get just 20MB/s reads, which seems like a pretty low standard
considering some people are getting 100+. This is my first time with OI, and
raid, for that matter, so I hope you guys have a little patience for a noob. :)
I figured out how to setup the vdevs and smbshare after some trial and error,
and got my Windows box to see the share. Transferring a 40GB file to the share
yields 55-80MB/s, not earth-shattering, but satisfactory IMO. The problem is
when I transfer the same file back to the Windows box, it went to less than
5MB/s. I then copied a 1GB file, and then moved that 1GB file from the raidz2
drives to the root drive (SSD), in attempt to isolate the problem. That was less
than 5MB/s. The same file, once again, copied from the root drive to the raidz2
was fast, maybe 70-100MB/s.
The problem here as far as I can tell is either some setting within zfs or the
HBA controller. Or maybe... even the timing issue with WD Green drives
shouldn''t create that much disparity.
I''ve attached the iostat of when activity is idle, when copying from
raidz2 to root ("read"), and for comparison, copying to raidz2 from
root ("write"). Please note the intermittent idling in all disks
(except 1?) when the file is copied from the raidz2 volume to anywhere else. I
have no idea what that''s about, but the drives will drop to 0 every
couple of seconds, and repeat.
My system is as follows:
10 WD20EARS (bad idea? I only found out after I bought them.) in raidz2 config
32GB SSD for root drive for OS install
Supermicro USAS-L8i HBA card (1068E chipset I believe?)
6GB RAM
500 watt power supply
AMD Athlon II X2 260 CPU
Here''s my zpool:
pool: rpool
state: ONLINE
scan: none requested
config:
NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
c2t0d0s0 ONLINE 0 0 0
errors: No known data errors
pool: solaris
state: ONLINE
scan: none requested
config:
NAME STATE READ WRITE CKSUM
solaris ONLINE 0 0 0
raidz2-0 ONLINE 0 0 0
c2t4d0 ONLINE 0 0 0
c2t5d0 ONLINE 0 0 0
c1t0d0 ONLINE 0 0 0
c1t1d0 ONLINE 0 0 0
c1t2d0 ONLINE 0 0 0
c1t3d0 ONLINE 0 0 0
c1t4d0 ONLINE 0 0 0
c1t5d0 ONLINE 0 0 0
c1t6d0 ONLINE 0 0 0
c1t7d0 ONLINE 0 0 0
errors: No known data errors
I''d greatly appreciate it if someone could give me some leads on where
the problem might be. I''ve spent the past 2 days on this, and
it''s very frustrating since I would actually be very happy getting even
10MB/s read.
Regards.
--
This message posted from opensolaris.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Idle-iostat
Type: application/octet-stream
Size: 14091 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20110720/214fe0e9/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Read-iostat
Type: application/octet-stream
Size: 17381 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20110720/214fe0e9/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Write-iostat
Type: application/octet-stream
Size: 3270 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20110720/214fe0e9/attachment-0005.obj>