Solaris

Hello,
we have a machine with Solaris Express 11, 2 LSI 9211 8i SAS 2 controllers (multipath to disks), multiport backplane, 16 Seagate Cheetah 15K RPM disks.

Each disk has a sequential performance of 220/230 MB/s and in fact if I do a
dd if=/dev/zero of=/dev/rdsk/<diskID_1> bs=1024k count=16384

writing 16 GB of sequential 0 goes at that speed (iostat confirms that, with load in round robin on the 2 controllers and 100% disk utilization)

If I now create a simple zpool made up of 1 single disk (the same disk as above) with zpool create tank <diskID_1>

and then issue a

dd if=/dev/zero of=/Tank/test bs=1024k count=16384
iostat shows a disk performance of less than 20 MB/s (1/10 !!!!!!!) with disk utilization around 96% (also zpool iostat shows similar figures)


I tried to set also checksum=off but it does not change anything.
Zpool is created with default values (compress=off, dedup=off, stripe 128k).

Of course if I turn compression on and dedup on everthing changes, but this is a "trick" and works because file are all zeros.

I tried to change the max_vdev_pending to 1, to 64 (default is 10) and nothing changes.

Anyone has a clue of what is going on ?

how much memory do you have in your server?
have you tried to use zpool iostat 1 as output-tool?