Hi, I'm trying to measure the performance of the NFS client and server in Solaris 10. I have two machines, both running Solaris 10u2. One machine is a x86 V20z and the other one is a UltraSparc Fire V240. The systems are interconnected using gigabit with a HP ProCurve switch in between (gigabit the whole way).
Both have arrays connected to them. The V240 has a 6140 array with a zfs pool. I have measured the read and write performance locally to over 200MB/s. I've disabled the zil_update on this pool. The V20z has a generic sata array and I've neasured write performance above 100MB/s sustained. I've created a 8GB file on the 6140 and I'm now trying to read and write from the V20z using NFS. I've tried combinations of NFSv3 and NFSv4 but I get really poor performance. Reading the file using the following command: $ dd if=/mnt/bigfile of=/dev/null bs=128k gives me a reading speed of around 35-40MB/s which is disappointing. But now for the weird part $ dd if=/dev/zero of=/mnt/bigfile.1 bs=128k from the V20z to the V240 gives me a write speed of 80MB/s. So my problem is that I get reading speed of 40MB/s over NFS and with the same options I get writing speed of 80MB/s. I'm having a really hard time to understand this and I was hoping someone here could provide some light in this very dark tunnel. What could cause read speed to be so much slower than write? cheers, Nickus -- Have a look at my blog for sysadmins! http://aspiringsysadmin.com
