Configuring and controlling NetBackup for VMware
Successful VMware snapshots depend on the amount of I/O that occurs on the virtual machine datastore during the snapshot. When a snapshot is taken, a delta .vmdk file is created. The delta disk represents the difference between the running state of the virtual machine and the state that existed at the time of the snapshot. A virtual machine with a large amount of I/Os during snapshot increases the size of the delta disks.
To delete such a snapshot, a large amount of information must be read and written to a disk. This process can reduce the virtual machine performance until the consolidation is complete. The time to delete snapshots and consolidate the snapshot files depends on the amount of data that the guest operating system writes to the virtual disks after you take the last snapshot. The general recommendation is to strive to saturate backup throughput, so the backup duration is as small as possible. Backups should be scheduled when relatively little I/O activity is expected.
VMware intelligent policies (VIPs) - For policies that select virtual machines automatically using the query filter:
The Limit jobs per policy option control the number of parent (discovery) jobs that run simultaneously for the policy. This option does not limit the number of snapshot jobs and backup (bpbkar) jobs that the parent job launches.
VMware policies using manual selection of virtual machines:
Limit jobs per policy controls the number of virtual machines that the policy can back up simultaneously. Because no discovery job is needed, each virtual machine backup begins with a snapshot job. Each snapshot counts against the Limit jobs per policy setting. If this option is set to 1: the backup of the next virtual machine that is specified in the policy cannot begin until the first snapshot job and its backup are complete.
You can use the NetBackup Resource Limit dialog to control the number of simultaneous backups that can be performed on a VMware resource type. The settings apply to all VMware policies. This powerful feature gives you great control over how to maximize performance without causing trashing of virtual infrastructure.
Following are some key resource limits that are available in NetBackup and how to use them to tune the load of backup operations on virtual infrastructure.
Resource limit | Description |
|---|---|
vCenter | Maximum jobs that can run per vCenter. Useful to protect a small vCenter server. |
Snapshot | Maximum snapshot creates/deletes that can run at the same time. This resource can be useful for limiting the effect that multiple snapshot operations have on the vCenter server. This is a global per-vCenter limit; consider your most resource-constrained vCenter to set the appropriate value. This should be used if NetBackup is reporting failures during snapshot creation or deletion. Snapshot only limits, but will not affect the number of backup jobs. |
VMXDatastoret | Controls the maximum number of simultaneous backups per datastore where VMX files are stored. During a backup of a virtual machine snapshot, vmdk writes are cached on this datastore. This resource type is useful for VMs that have VMX and VMDK files distributed across multiple datastores. |
Datastore | Maximum number of backup jobs per datastore. Useful to avoid overloading datastore with limited spindles. |
ESXserver | Limits jobs running per host. |
DatastoreNFSHos | Limits backups per datastore at an NFS Host level. |