Strategies to Minimize Data Loss with Hyper-V Replica

Last Update: Sep 04, 2024 | Published: Sep 18, 2017

SHARE ARTICLE

Azure Hero Server scaled

Hyper-V Replica is a VM disaster recovery and data protection feature that’s especially well-suited to enable small and medium sized organizations improve their disaster recovery capabilities. Microsoft first introduced Hyper-V Replica as a part of Windows Server 2012 R2 and has continued to enhance it in the Windows Server 2016 release. Hyper-V Replica essentially works by asynchronously sending the changes in a local VM’s virtual hard disk from a primary site to a replica site. The Hyper-V server that hosts the replica doesn’t need to be identical to your source Hyper-V server but it needs have adequate processing power and memory to run the replica VMs if the primary VM fails. You can setup Hyper-V Replica using the Hyper-V Manager. No additional software is required. When you create a new Hyper-V Replica there is a normally an initial synchronization process which typically takes place across the network. However, for large VMs you can also perform the initial synchronization using media like a USB drive or external hard drive.

Setting the replication interval

Once the initial replication takes place then Hyper-V Replica will use periodic asynchronous replication to send updates from the primary VM to the secondary replica. One of the biggest factors that governs the amount of possible data loss that can occur with Hyper-V Replica is the replication interval. The data on the secondary replica is synchronized according to the replication frequency you configure. You can select a replication interval of 30 seconds, 5 minutes, or 15 minutes. Business critical VMs need to be synchronized at more frequent intervals to minimize possible data loss.

Adding Recovery Points

Another important factor to consider when you are configuring Hyper-V Replica are the number of recovery points that you want to keep. Recovery points are basically a snapshot in time from which you can recover a VM. By default Hyper-V Replica only uses the latest recovery point but you can optionally add more recovery points that allow you to recover from earlier points in time. The more recovery points that you specify the more disk storage will be required on the secondary replica. You can use up to 24 recovery points.

Enhanced DR with Hyper-V Extended Replica

Extended Replication allows you to create a third replica. This feature provides an additional option for disaster recovery. If a VM failure occurs then you can choose to recover from either the secondary replica or from the extended replica. The most common use for this feature is where you have one replica within the local datacenter and the extended replica is located in a remote location or in the cloud.

Managing the Failover Process

Finally, when you’re planning to minimize data loss it’s important to understand the different types of failover types. Hyper-V Replica supports three major failover types:

  • Test failover — A test failover can be performed at any time. It’s important to realize that a test failover option doesn’t actually result in a failover. Instead, the process creates a new test VM. This test VM lacks network connectivity and can be powered on and tested without interfering in the replication process or your production workloads.
  • Planned Failover – You can also perform a planned failover at any time. However, before performing a planned failover the primary replica VM must be powered off. After the VM is failed over, Hyper-V can begin replicating the changes back to the primary server. For this to work the primary server needs to be configured to receive replication from the secondary server.
  • Unplanned Failover — An unplanned failover should only be performed in the event that your primary VM has an unexpected outage and becomes unavailable. You run an unplanned failover from the secondary server. An unplanned failover does not perform any synchronization as part of the failover process so any data that had not already synchronized will be lost. The actual amount of data lost depends on the length of your replication interval and the amount of data that was changed in the primary VM before the outage.

Hyper-V Replica does not provide automatic failover nor does it attempt to update DNS entries like some enterprise-level replication products might do.

SHARE ARTICLE