Hi all. I'm new to ZFS, and I have just installed my first ZFS pools and file systems. My Oracle DBA tells me that he's seeing poor performance and would like to go back to VxFS.
Here's my hardware: Sun E4500 with Solaris 10, 08/07 release. SAN attached through a Brocade switch to EMC CX700. There is one LUN per file system. Here are the file systems: # df -h Filesystem size used avail capacity Mounted on /dev/md/dsk/d0 2.0G 142M 1.8G 8% / /devices 0K 0K 0K 0% /devices ctfs 0K 0K 0K 0% /system/contract proc 0K 0K 0K 0% /proc mnttab 0K 0K 0K 0% /etc/mnttab swap 194M 1.2M 192M 1% /etc/svc/volatile objfs 0K 0K 0K 0% /system/object /dev/md/dsk/d9 3.9G 3.1G 827M 80% /usr fd 0K 0K 0K 0% /dev/fd /dev/md/dsk/d12 2.0G 403M 1.5G 21% /var swap 194M 1.6M 192M 1% /tmp swap 192M 32K 192M 1% /var/run /dev/md/dsk/d6 2.0G 1.2G 728M 63% /opt r12_oApps 78G 55G 23G 71% /opt/a01/oakwcr12 r12_data/d01 44G 42G 2.0G 96% /opt/d01/oakwcr12 r12_data/d02 42G 40G 2.1G 95% /opt/d02/oakwcr12 r12_data/d03 42G 40G 2.1G 95% /opt/d03/oakwcr12 r12_data/d04 44G 40G 3.7G 92% /opt/d04/oakwcr12 r12_data/d05 42G 33G 9.1G 79% /opt/d05/oakwcr12 r12_data/d06 42G 40G 2.4G 95% /opt/d06/oakwcr12 r12_data/d07 42G 31G 11G 75% /opt/d07/oakwcr12 r12_data/d08 42G 40G 1.7G 96% /opt/d08/oakwcr12 r12_data/d09 42G 40G 1.8G 96% /opt/d09/oakwcr12 r12_data/d10 44G 42G 2.5G 95% /opt/d10/oakwcr12 r12_data/d11 42G 39G 3.3G 93% /opt/d11/oakwcr12 r12_data/d12 42G 13G 29G 33% /opt/d12/oakwcr12 r12_data/d21 42G 37G 5.4G 88% /opt/d21/oakwcr12 r12_data/d22 42G 40G 1.9G 96% /opt/d22/oakwcr12 r12_data/d23 42G 40G 2.2G 95% /opt/d23/oakwcr12 r12_data/d24 42G 40G 2.1G 95% /opt/d24/oakwcr12 r12_logz 14G 24K 14G 1% /opt/l01/oakwrc12 r12_product 19G 4.7G 14G 26% /opt/p01/oakwcr12 r12_oWork 39G 28K 39G 1% /opt/w01/oakwcr12 # Here are the record sizes: # zfs get recordsize NAME PROPERTY VALUE SOURCE r12_data recordsize 8K local r12_data/d01 recordsize 8K local r12_data/d02 recordsize 8K local r12_data/d03 recordsize 8K local r12_data/d04 recordsize 8K local r12_data/d05 recordsize 8K local r12_data/d06 recordsize 8K local r12_data/d07 recordsize 8K local r12_data/d08 recordsize 8K local r12_data/d09 recordsize 8K local r12_data/d10 recordsize 8K local r12_data/d11 recordsize 8K local r12_data/d12 recordsize 8K local r12_data/d21 recordsize 8K local r12_data/d22 recordsize 8K local r12_data/d23 recordsize 8K local r12_data/d24 recordsize 8K local r12_logz recordsize 128K default r12_oApps recordsize 128K default r12_oWork recordsize 128K default r12_product recordsize 128K default # Doing some performance tests I see this: # time dd if=/dev/zero of=test.dbf bs=8k count=1048576 1048576+0 records in 1048576+0 records out real 6:29.9 user 6.2 sys 3:26.1 # time dd if=test.dbf of=/dev/null bs=8k 1048576+0 records in 1048576+0 records out real 3:06.4 user 5.5 sys 1:26.9 # I thought that I would see better performance than this. I've read a lot of the blogs, tried tuning this, and still no performance gains. Are these speeds normal? Did I miss something (or somethings)? Thanks for any help! _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
