Are you like one of the millions of people pondering the answer to ‘what is hypervisor-based replication and how will it change my disaster recovery approach’? I know I was.
So, let me help you with that!
Our technology experts here at Eze Castle Integration spent some time in the lab testing and evaluating hypervisor-based replication and recently incorporated it into our Eze Disaster Recovery 2.0 offering. We think it delivers excellent benefits, but let’s start with the basics.
What is hypervisor-based replication?
TechTarget defines hypervisor-based replication as “a technology that automatically creates and maintains replicas of virtual hard disks or entire virtual machines (depending on the platform that is being used).” Analyst firm IDC goes on to say that this replication approach “protects virtual machines (VMs) at the virtual machine disk format file level rather than at the LUN or storage volume level, thus replication can be done without the management and TCO challenges associated with array-based replication.”
Here is a diagram of how it works in our Eze DR solution, which uses Zerto hypervisor-based replication technology.
How does it compare with other software-based replication?
Before answering that question let’s quickly review the other replication category segments. Following are IDC’s definitions:
Host replication software typically resides at the file system or logical volume level within the operating system and makes a point-in-time copy or snapshot of a data set to disk used for disaster recovery (DR), testing, application development, or reporting.
Fabric and appliance-based replication software makes use of intelligent switches and heterogeneous array products to provide block-level replication within the SAN. The intelligent switches have technologies that perform the volume management and replication process and eliminate the overhead on the host while providing any-to-any replication.
Array-based replication software makes a block-based point-in-time block copy or snapshot of storage to disk used for disaster recovery, testing, application development, reporting, and other uses.
Okay, but what’s the difference?
According to Zerto, “before virtualization, replication was managed at the storage layer, which made perfect sense because that’s where the information was…but in a virtual environment, the boxes aren’t (or aren’t all) physical, so putting a physical sensor on a virtual box isn’t going to help you protect its contents.”
Hypervisor-based replication moves replication up into the virtualization/hypervisor layer – above the resources abstraction layer. The benefits of this approach, according to Analyst George Crump, can include:
Being VM aware, which can simplify storage setup and minimize storage requirements at a DR site
Gaining a hardware-neutral solution
Reducing DR-related costs
Who are the players?
Well you’ve heard me mention one already – Zerto. Analyst firm IDC identifies Zerto as “the first to launch a hypervisor-based replication approach in June 2011. VMware followed suit in August 2011, with its vSphere SRM 5.0 data replication feature.” Given the potential of this technology we expect to see more firms targeting this space.