Mike Gerdts wrote: [...] >> Can you try 1024000, this is the default in snv_119. >> Did you verify the new setting take effect on NFS connections? > >I'm *so* glad to see that this is finally getting some attention. >Are there any good reasons not to increase them to that value on S10 - >snv_118? > >> >> -Dai > >On Solaris 10 (vintage October 2007) I noticed that the following >tuning was helpful on gigabit networks. The part that is probably >interesting to you is nfs3_bsize, which had a clearly inadequate >default on for 1 GigE. I'm not sure that the transfers really ever >got up to the size I tuned it to (suspect it was choked elsewhere) but >I was happy with the improvement I saw. > >== Begin quote from my notes == >Details: Simple single-stream TCP tests on a T2000 show that >increasing the xmit & recv hiwat values increases throughput by nearly >2x (500 Mb/s -> 904 Mb/s). On top of that, tuning the nfs3 block size >to 1 MB increases large NFS read performance by 3.6x (243 Mb/s -> 872 >Mb/s). > >Details: Simple single-stream TCP tests on a T2000 show that >increasing the xmit & recv hiwat values increases throughput by nearly >2x (500 Mb/s -> 904 Mb/s). On top of that, tuning the nfs3 block size >to 1 MB increases large NFS read performance by 3.6x (243 Mb/s -> 872 >Mb/s). > >ndd -set /dev/tcp tcp_xmit_hiwat 131072 >ndd -set /dev/tcp tcp_recv_hiwat 131072 > >Add to /etc/system: > >* NFS Client tuning - do 1 MByte transfers by default >* T2000 NFS read performance jumps by 3.6x >set nfs:nfs3_bsize=0x100000 >* END NFS client tuning >== End quote from my notes == > >I don't know if there is similar tuning for NFSv4.
There is. I have previously played with the nfsv3_bsize / nfsv4_bsize but never went as high as 1 MB. I am bit concerned about the NFS client issuing 1 MB reads / writes when the servers filesystem is using a lower blocksize since this kind of misalignment could multiply the IOPS on the serverside. I will have to test this to be sure though. It does however show improvement : henrik at opensolaris:/tmpfs# rm /nfs/testfile_5g; ptime cp testfile_5g /nfs/ real 31.321109059 user 0.017099704 sys 11.804887174 Which gives me about ~165 mb/s - the fastest so far for a cp operation. I'll have a closer look at the Solaris Tunable Parameters Reference Manual and check if I have missed something ... >-- >Mike Gerdts >http://mgerdts.blogspot.com/ -- Med venlig hilsen / Best Regards Henrik Johansen
