Understanding performance tuning

Client options are used with commands to control Data Protection for Microsoft Hyper-V operations. Learn about options and parameters that can help you to tune the performance in your environment.

When you are planning for performance, you typically have goals for Recovery Point Objectives (RPO) and Recovery Time Objectives (RTO). You want to ensure that the performance is good enough to meet your goals, but also balance the resources is used by data protection operations on the Hyper-V hosts in your environment.

The amount of data that is transferred between Data Protection for Microsoft Hyper-V and the IBM Storage Protect server is different for different operations and scenarios:
Backup operations
The amount of data that is transferred for backup operations depends on the state of the backup.
Back up at initial ingestion
When you install Data Protection for Microsoft Hyper-V, no virtual machines (VMs) are backed up and full backup operations are required for each VM. During these initial backup operations, a large amount of data is transferred.
Back up after adding VMs to a Hyper-V host
When you add VMs to a host, the VMs require a full backup. Depending on the number and size of the VMs, a corresponding amount of data is transferred.

In a clustered environment, VMs might move between host nodes. While the total amount of data backed up for the cluster does not change, the amount of data transferred from a particular host changes.

Back up after initial ingestion (steady state)
After the initial full backup operations that occur at initial ingestion, incremental backup operations occur. Incremental backup operations in combination with default-enabled compression and deduplication features result in smaller amounts of data that is transferred. The majority of backup operations are incremental.
Restore operations
Restoring individual or multiple VMs causes a corresponding amount of data to be moved. Consider your RTO and the potential impact to the host and adjust the options.
Disaster recovery operations
In a disaster recovery scenario, you might be more interested in maximizing throughput rather than reducing the impact to the host and might want to adjust the parameters.