I've created a personal "storage lab" for testing out various filesystems and raid implementations. At the moment, I'm trying out ZFS. I've created a zpool at /zfs-raid and am attempting to test /zfs-raid/bench. I was attempting to perform a 128MB sequential write with Linux AIO, unbuffered and direct. However, Phoronix was giving a result of 'The test run did not produce a result'. So I performed a debug run, and this is what I think is the relevant snippet of output:
So apparently I can't test ZFS with those settings.
To make sure that the zpool is actually online and usable I performed the following rudimentary benchmark:
Can I even benchmark ZFS with Phoronix? If I can, what test parameters are sane for ZFS?
Code:
fio: native_fallocate call failed: Operation not supported fio: looks like your file system does not support direct=1/buffered=0 fio: destination does not support O_DIRECT test: No I/O performed by libaio, perhaps try --debug=io option for details?
To make sure that the zpool is actually online and usable I performed the following rudimentary benchmark:
Code:
# dd if=/dev/zero of=/zfs-raid/bench/testfile.img bs=128M count=40 40+0 records in 40+0 records out 5368709120 bytes (5.4 GB, 5.0 GiB) copied, 12.998 s, 413 MB/s