Why is virtual machine performance poor?

Virtual Private Cloud 4.17 and later Bare metal worker nodes only RHCOS only

Your virtual machine (VM) starts, but the workload performance is lower than expected.

Performance can be affected by node sizing, storage characteristics, VM placement, and resource contention on the worker node.

Use the following checks to investigate the environment.

1. Review the VM resource allocation.

   oc get vm <vm_name> -n <namespace> -o yaml | grep -A 5 resources
   

2. Verify that the VM is running on a bare metal worker node.

   oc get vmi <vm_name> -n <namespace> -o wide
   

3. Review node resource utilization.

   oc describe node <node_name>
   oc adm top node <node_name>
   

4. Review your storage configuration.

   oc get storageclass <storage_class_name> -o yaml
   

5. Compare the current deployment to the planning guidance in Planning your OpenShift Virtualization deployment and the storage guidance in Setting up storage for OpenShift Virtualization. If needed, adjust your worker flavor selection, VM sizing, or storage choice to better match the workload profile.

6. If you use ODF, review the storage cluster status.

   oc get storagecluster -n openshift-storage
   

7. If performance remains poor, collect node, VM, and storage details before you contact IBM Cloud support. Include the command output when you open a support case so that the cluster, node, and storage conditions can be reviewed.