On Tue, 3 Jun 2008, Niclas Sodergard wrote: > Hi everyone, > > This is not purely OpenSolaris but still Solaris + STK2540 related. > > This seems to be the performance I'm getting. 12MB/s per pair, asvc_t > of 350ms (??) and 100% busy.
Sorry to hear about your woes. > Clearly this is not the performance I expect from such a simple test > (yes, dd is not a good test but I would expect slightly better > sequential write speed especially with zfs) but after googling a lot I > can see people are happy with their arrays. My question is what kind > of performance do people get from SATA disks in a STK2540? What can I > expect? Is there anything to tune? See the writeup for my own experiences which may be found at "http://www.simplesystems.org/users/bfriesen/zfs-discuss/2540-zfs-performance.pdf";. Note that this is with fast SAS drives. While the available IOPS would be less with SATA drives, it seems that sequential write speed should not be terribly less since from what I have seen, the bottleneck for sequential writes is in the 2540's RAID controller and not the drives themselves. SATA drive latency is a lot higher than SAS so that is surely a factor and may put additional load on the RAID controller. There are a couple of tunables mentioned in my writeup which do substantially improve sequential write speeds. Make sure that you assign the full space available from the drive for the per-drive JBOD pools. Check to see if there is a formal Sun benchmark posted for your configuration. I have posted (see below) a somewhat modified version of a script from Jeff Bonwick to test the responsiveness of your drives. With my SAS-based 2540, this script returns values in the range of 179 to 221 MB/sec. Bob ====================================== Bob Friesenhahn [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ #!/bin/ksh # Date: Mon, 14 Apr 2008 15:49:41 -0700 # From: Jeff Bonwick <[EMAIL PROTECTED]> # To: Henrik Hjort <[EMAIL PROTECTED]> # Cc: [EMAIL PROTECTED] # Subject: Re: [zfs-discuss] Performance of one single 'cp' # # No, that is definitely not expected. # # One thing that can hose you is having a single disk that performs # really badly. I've seen disks as slow as 5 MB/sec due to vibration, # bad sectors, etc. To see if you have such a disk, try my diskqual.sh # script (below). On my desktop system, which has 8 drives, I get: # # # ./diskqual.sh # c1t0d0 65 MB/sec # c1t1d0 63 MB/sec # c2t0d0 59 MB/sec # c2t1d0 63 MB/sec # c3t0d0 60 MB/sec # c3t1d0 57 MB/sec # c4t0d0 61 MB/sec # c4t1d0 61 MB/sec # # The diskqual test is non-destructive (it only does reads), but to # get valid numbers you should run it on an otherwise idle system. disks=`format </dev/null | grep ' c.t' | nawk '{print $2}'` getspeed1() { ptime dd if=/dev/rdsk/${1}s0 of=/dev/null bs=64k count=1024 2>&1 | nawk '$1 == "real" { printf("%.0f\n", 67.108864 / $2) }' } getspeed() { for iter in 1 2 3 do getspeed1 $1 done | sort -n | tail -2 | head -1 } for disk in $disks do echo $disk `getspeed $disk` MB/sec done _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
