numam-spdk/examples/nvme/fio_plugin

Compiling

First, clone the fio source repository from https://github.com/axboe/fio

git clone https://github.com/axboe/fio

Then check out the fio 2.18 tag

cd fio && git checkout fio-2.18

Finally, compile the code with

./configure && make

Next, edit the CONFIG file located in the root of the SPDK repository and set CONFIG_FIO_PLUGIN to y and FIO_SOURCE_DIR to the location of the fio repository that was just created.

Further, you'll need to build DPDK with -fPIC set. You can do this by modifying your DPDK config file (i.e. config/defconfig_x86_64-native-linuxapp-gcc) to include the line

EXTRA_CFLAGS=-fPIC

At this point, build SPDK as per normal. The fio plugin will be placed in the same directory as this README.

Usage

To use the SPDK fio plugin with fio, simply set the following in the fio configuration file (see example_config.fio in the same directory as this README).

ioengine=<path to fio_plugin binary>

To select NVMe devices, you simply pass an identifier as the filename in the format

'key=value [key=value] ... ns=value'

Do not have any ':' in filename, otherwise it will be spilt into several file names. Also the NVMe namespaces start at 1, not 0! And it should be put on the end. For example,

1. For local PCIe NVMe device - 'trtype=PCIe traddr=0000.04.00.0 ns=1'. traddr for local NVMe device should use this format: domain.bus.slot.func
2. For devices exported by NVMe-oF target, 'trtype=RDMA adrfam=IPv4 traddr=192.168.100.8 trsvcid=4420 ns=1'

Currently the SPDK fio plugin is limited to thread usage model, so fio jobs must also specify thread=1 when using the SPDK fio plugin.

When testing random workloads, it is recommended to set norandommap=1. fio's random map processing consumes extra CPU cycles which will degrade performance over time with the fio_plugin since all I/O are submitted and completed on a single CPU core.