Recovery Assurance – Which Job is Right For You?

Let’s assume that ReliableDr is up and running in your environment and connected to all of the hosts and Unitrends appliances that you will be including in the DR plan.  We went over these steps in the first three entries in the Recovery Assurance series.  Please take a look at the series from the beginning if you are not familiar with ReliableDR.

Part 1: How to Design a DR Plan Using ReliableDR

Part 2: Where should I Deploy the ReliableDR Server?

Part 3: How to Configure ReliableDR

Now it is time to choose your ReliableDR job type for each of the DR automation tests you will be setting up.  We’ll go over the different job types and the benefits of each below.

What happens during a ReliableDR job test?  It’s very simple.  ReliableDR will spin up the VMs in the proper boot order, make sure they boot properly and get an IP address, run application tests, validate RTO and RPO SLA’s and then shut down all of the VMs.  If the test is successful, then it will create a Certified Recovery Point.

First, I’d like to go over a few things to keep in mind while creating your ReliableDR replication jobs.  Create jobs based on applications used in your environment like I listed in part one of the series, How to Design a DR Plan with ReliableDR.  Notice any common or dependent VMs and create a “Continuous Dependency” job.  A continuous dependency job will leave the VM up and running after a test has finished, so it is available for other tests.  For example, an Active Directory server would be a good candidate for a continuous dependency job, so other jobs can use the Active Directory services.

Choose your job type from the four listed below.

• ReliableDr Replication – Built into ReliableDR and creates replica VMs at the DR site for VMware
• Storage Replication – Utilizes existing SAN-to-SAN replication and spins up the VMs replicated by the arrays
• Unitrends Backup and Replication –
  • validates backups of Vmware and Hyper-V VMs taken with Unitrends Backup
  • Validates Windows Instant Recovery VMs created with Unitrends Bridge in Vmware and Hyper-V
• Hyper-V replica – spins up VMs replicated with Hyper-V replication

 ReliableDR Replication

• Creates replica VMs at the DR site, so you will need storage, RAM and CPU to match the source VMs
• Make sure you allow for extra storage at your DR site for snapshots of the VMs.  The Certified Recovery Points for this type of job are snapshots of each VM at the DR site
• If you need to throttle the bandwidth, then consider using the Unitrends Backup appliances to backup and replicate the VMs to take advantage of the bandwidth throttling feature
• This job type is Independent of Unitrends Backup and uses ReliableDR to replicate the VMs

Storage Replication

• Mounts a datastore in the DR site from the array, registers, starts and tests the VMs
• Does not affect the primary replication operations
• Lower RPOs can be achieved
• Certified Recovery Points are snapshots of the volumes in the DR array
• Netapp, Datacore, Nimble and 3Par storage  can use advanced failover/failback procedures
• You may want to restructure VM location and volume/LUN association in the production site for full failover scenarios.  Full failover will shut down the VMs in the primary site and unmount the datastores, so you want to make sure other production VMs aren’t affected if they are not included in the job

Unitrends Backup and Replication

• ReliableDR automates the instant recovery of the VMs performed by the Unitrends appliance
• Either VMware backups or Hyper-V backups can be tested in a job – no mixing hypervisors in the same job
• Windows Instant Recovery servers created with Unitrends Bridge included in a job is a feature that allows the testing of physical windows machines in the DR site
• Bandwidth throttling can be enabled using this job type if there are bandwidth considerations

Hyper-V Replica

• Performs automated testing and orchestrates disaster recovery for Hyper-V replicas
• Performs failback operations and planned migrations
• Creates a test VM from the replica’s latest snapshot and verifies the VM and its hosted applications are running properly
• Works with both standalone hosts and Hyper-V clusters

At this point you will have a good understanding of the different options ReliableDr offers for getting the production VMs replicated to the DR site for testing.  The next two editions of the series will involve setting up different job types showing the steps involved.

Recovery Assurance Series

Part 1: How to Design a DR Plan Using ReliableDR

Part 2: Where should I Deploy the ReliableDR Server?

Part 3: How to Configure ReliableDR

Part 4: What Job is Right For You?

If you have any feedback on the series, then we would love to hear from you.