- Generar un Namespace para las pruebas
oc new-project test-fio
- Generar un POD en ese namespace (Seleccionamos un master y agregamos las toleration para que se pueda ejecutar ahi)
apiVersion: v1
kind: Pod
metadata:
name: fio-tester-kkdp2
spec:
nodeSelector:
kubernetes.io/hostname: ocpnp-zdpgh-master-0
tolerations:
- key: node-role.kubernetes.io/master
operator: Exists
effect: NoSchedule
containers:
- name: fio-tester-kkdp2
image: quay.io/openshift-scale/scale-ci-fio
Conectarse al POD generado vía rsh
oc rsh {POD NAME}
Cambiarse al directorio /var/tmp y generar un dir de pruebas
cd /var/tmp
mkdir test-data
fio --rw=write --ioengine=sync --fdatasync=1 --directory=test-data --size=22m --bs=2300 --name=mytest
- ioengine=sync - Basic read(2) or write(2) I/O. lseek(2) is used to position the I/O location. See fsync and fdatasync for syncing write I/Os.
- fdatasync=int - Like fsync but uses fdatasync(2) to only sync data and not metadata blocks. In Windows, DragonFlyBSD or OSX there is no fdatasync(2) so this falls back to using fsync(2). Defaults to 0, which means fio does not periodically issue and wait for a data-only sync to complete.
fio --ioengine=libaio --direct=1 --gtod_reduce=1 --bs=8k --name=wtest --invalidate=0 --readwrite=randwrite --iodepth=64 --numjobs=1 --filename=/var/tmp/fiotest --size=10G --output=test_write.log --rwmixread=0 --rwmixwrite=100 --time_based --runtime=300s
fio --randrepeat=0 --verify=0 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=write_iops --filename=/var/tmp/fiotest --bs=4K --iodepth=64 --size=10G --readwrite=randwrite --time_based --ramp_time=2s --runtime=300s
fio --randrepeat=0 --verify=0 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=write_bw --filename=/var/tmp/fiotest --bs=128K --iodepth=64 --size=10G --readwrite=randwrite --time_based --ramp_time=2s --runtime=300s
fio --randrepeat=0 --verify=0 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=write_seq --filename=/var/tmp/fiotest --bs=1M --iodepth=16 --size=10G --readwrite=write --time_based --ramp_time=2s --runtime=300s --thread --numjobs=4 --offset_increment=500M