Benchmark Kubernetes persistent disk volumes with fio: Read/write IOPS, bandwidth MB/s and latency
Benchmark Kubernetes persistent disk volumes with fio
: Read/write IOPS, bandwidth MB/s and latency.
storageClassName
to match your Kubernetes provider's Storage Class kubectl get storageclasses
kubectl apply -f dbench.yaml
1000Gi
(default) using storageClassName: ssd
(default)fio
tests on the newly provisioned diskkubectl logs -f job/dbench
(empty output means the Job not yet created, or storageClassName
is invalid, see Troubleshooting below)==================
= Dbench Summary =
==================
Random Read/Write IOPS: 75.7k/59.7k. BW: 523MiB/s / 500MiB/s
Average Latency (usec) Read/Write: 183.07/76.91
Sequential Read/Write: 536MiB/s / 512MiB/s
Mixed Random Read/Write IOPS: 43.1k/14.4k
kubectl delete -f dbench.yaml
and that should deprovision the persistent disk and delete it to minimize storage billing.kubectl get storageclasses
. Also check that the volume size of 1000Gi
(default) is available for provisioning.4000Gi
volume will perform better than a 1000Gi
volume. Just edit the yaml, kubectl delete -f dbench.yaml
and run kubectl apply -f dbench.yaml
again after deprovision/delete completes.fio
tests are in docker-entrypoint.sh.