Much to the surprise of VMware, the world is not 100% VMware. Hyper-V is seeing a rapid increase in adoption both in small businesses and for smaller remote offices in enterprises. But small organizations may want to replicate to a cloud provider that uses VMware and many remote offices need to replicate to a VMware based central data center. Other organizations want to leverage Amazon as a disaster recovery location or for backup or test and development. The cloud can be an ideal DR site since Amazon doesn’t charge for compute resources until they’re consumed, which allows for essentially a free standby DR site, other than the cost of storage and compute used during testing. All of this data movement requires cross-hypervisor replication, a function that Zerto in its 4.0 release is addressing, in addition to cross-cloud replication.
Disaster Recovery-as-a-Service (DRaaS) allows a business to start instances of its applications at a cloud provider’s facility without having to buy dedicated hardware. And regional cloud providers can often deliver the facilities and resources required very cost effectively, with the local support that an Amazon can’t. The key is to have the right data there at the right time. Zerto, with its cross-hypervisor replication capability, allows for the business to be running either Hyper-V or VMware replicating to a VMware Cloud Provider. This means that if the business is hit by a disaster they can instantiate their application in their provider’s cloud even if that cloud is running a different hypervisor.
In the event of a disruption a fail over to the cloud can be triggered with the click of a button and the application can be up and running in a matter of minutes, again regardless of hypervisor type. But most disaster declarations are “just in case”, there disaster never strikes, and the data center survives the impending disaster with data intact. If that happens, then rapid fail-back becomes critical. Zerto can automate the fail back process efficiently. When the time to fail back into the data center occurs, a quick verification of data is done with just the newly added or changed data being sent back to the data center. Again, this can be triggered with a single click. There is also a commit before rollback so the successful failover can be verified before shutting down production.
Another key feature in 4.0 is the ability to leverage Amazon as a failover location. While not as rapid of a recovery as a regional provider, a half hour versus a few minutes, Amazon can represent a tremendous cost savings compared to the rapid recovery provider solutions. Amazon does not charge for CPUs that are idle when everything is working, only when there is an actual disaster and those CPUs are needed to perform a recovery. The fail-back from Amazon is not as sophisticated as from a regional VMware base provider, because the entire VM needs to be copied back. But for customers that don’t need sub-30-minute fail-overs, Amazon remains an attractive option.
Organizations are demanding stricter and stricter recovery times, recovery point objectives (RPO/RTO) and disaster recovery capabilities, but the associated costs can be out of reach for many of them. These organizations are looking to the cloud to help them bridge the gap. Zerto is beginning to deliver on its Cloud Continuity Platform which they articulated in a video with us last year. This platform’s ability to replicate data across hypervisors and across clouds should allow organizations to meet DR challenges without many of the costs typically associated with having a standby, secondary data center.