ZFS performance weirdness on a a Netra X1

I have a Netra X1 with a pair of 120GB (non-Sun) disks, and 1152MB of memory (so enough, I hope). It's running 10 6/06, and the disks are split between SVM and: /, /var and swap are UFS filesystems on SVM mirrors, and the rest is taken up by zfs:
# zpool status
pool: export
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM
export ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t0d0s7 ONLINE 0 0 0
c0t2d0s7 ONLINE 0 0 0

errors: No known data errors
# zfs list
NAME USED AVAIL REFER MOUNTPOINT
export 520M 91.5G 27.5K /export
export/home 364M 91.5G 364M /export/home
export/zones 156M 91.5G 28K /export/zones
export/zones/ts 156M 91.5G 77.9M /export/zones/ts

I've been making zones with their roots in this space (for instance the ts zone has it's root in export/zones/ts). I'm aware this isn't supported, but it works on other machines and it's great for development stuff as you can snapshot zones.

It's very, very slow. For instance installing ts, which is a completely sparse zone, took over an hour. It's a slow machine, of course, but still this is a really long time.

I did some rudimentary experimentation with something like:

dd if=/dev/zero of=file bs=1024 count=102400

with resulting write rates of about 3MB/s through zfs, and about 14 through UFS and SVM, onto the same pair of disks, obviously. 3MB/s is pretty crippling. Obviously these aren't realistic FS benchmarks, but they are indicative of something.

What am I doing wrong?

Thanks

--tim

.