On 01/20/2017 03:11 PM, Henrique Fingler wrote:
I'm trying to run some fio benchmarks and it doesn't seem to
me. The example in GitHub
doesn't work (says "size=" is required) and it's very unclear if
there needs to be any configuration to point to the device.
It seems that when I run it everything goes straight to the root
filesystem (drive gets saturated with writes, all terminals freeze)
or memory (ungodly throughputs). Here's one input file:
And I get:
Run status group 0 (all jobs): READ: io=10227MB, aggrb=10227MB/s,
minb=10227MB/s, maxb=10227MB/s, mint=1000msec, maxt=1000msec
Am I using it correctly? Something seems off. Also, my root
filesystem gets saturated with writes. I can easily notice it using
dstat. Is this normal?
Could anyone give me a pointer or an example which works? My end goal
is to do some random writes benchmarks.
When using the SPDK NVMe fio_plugin, you need to specify a filename
in a special format. See examples/nvme/fio_plugin/README.md for the format.
fio yourconf.fio --filename=0000.04.00.0/1
Assuming yourconf.fio is the configuration above, this would run FIO
against the NVMe device at PCI domain 0, bus 4, device 0, function 0, NVMe
namespace ID 1.
There should not be any I/O activity to your root filesystems - all I/O
should bypass the kernel stack entirely if it is being processed by the
SPDK FIO plugin.