Briefing Update, Cloud, Data Center

CloudEndure Finds Joint Answers to Issues of Disaster Recovery and Cloud Migration

IT departments usually look at the challenge of setting up disaster recovery in the cloud as discrete from the equally large challenge of migrating workloads from an on-premises environment to the cloud. Yet, those two challenges have a lot in common.

One shared characteristic is speed. Whether you’re preparing data for disaster recovery or for a cloud migration, the less time required both to prep the data and to move it from one environment to another, the better. Even more important than speed may be disruption. Either process, be it disaster recovery or migration, should cause little or no downtime to production systems.

The ability to exercise complete control over the sequence of the process through automation and orchestration is another common issue for both DR and migration. Finally, the ability to test that a failover or a migration is going to work in the target environment is on most IT checklists for both processes.

One company that thinks of DR and migration as two sides of the same technical challenge is CloudEndure, a provider of Disaster Recovery and Live Migration for any Windows or Linux application onto most of the big public clouds, including Amazon Web Services, Google Compute Platform, Microsoft Azure and Oracle Cloud. We caught up with Gil Shai, Chief Revenue Officer and co-founder, for an update on the six-year-old company.

Underlying both of CloudEndure’s core solutions are several pillars of capability. The headline technology is operating system-based continuous block-level replication. With an eye to being able to replicate data being generated in the cloud, CloudEndure’s product architects wanted to stay away from replicating at the storage level or the hypervisor level. They viewed both approaches as problematic in cloud environments where those levels may not be accessible. By operating at the OS level, CloudEndure works in any environment and is able to replicate any application, even home-grown ones. After an initial read that doesn’t require a reboot and levies only a light I/O hit, CloudEndure’s agent never impacts disk again. It sits in memory capturing changes, then compresses them, encrypts them, and replicates them over the network.

Another impressive pillar is what CloudEndure bills as its 1-minute Any-to-Cloud Image Conversion. Given that data is moving at the block level, there’s some magic that has to happen in order to bring up a virtual machine on a new platform. Rather than requiring an admin to spend hours figuring out the correct drivers, CloudEndure has built an engine that converts machines from end to end. It injects hypervisor drivers, changes bootloaders, plugs in networking adapters, and overhauls anything else necessary to bring up the workload on the new platform.

A third pillar is an automated orchestration engine with the capability to spin up thousands of machines in parallel, which supports enterprise-scale disaster recovery requirements and also covers migration needs for splitting workloads into move groups.

In addition to those core technologies, CloudEndure offers intriguing capabilities in each of its solutions. On the disaster recovery side, the company uses minimal footprint replication staging areas in the public cloud as a way to reduce the total cost of ownership. The data is left in its compressed state until it’s needed for a recovery scenario. Waiting to inflate the data minimizes public cloud compute, storage, and networking costs until those resources are absolutely necessary. Meanwhile, the continuous replication means sub-second recovery point objectives and recovery time objectives of minutes. Applied to disaster recovery, the image conversion engine means failback is as fast and as effective as failover.

For Live Migration, those base technologies reduce human involvement in the process, shorten cutover windows to minutes, and make cutovers predictable. CloudEndure has added custom APIs and post-scripts for migration-related automation and optimization.

CloudEndure’s technology is getting attention from big cloud providers. Amazon brings CloudEndure into deals and features the company prominently at its conferences. Google uses CloudEndure as an OEM solution for cloud migrations. Even Microsoft, which offers a competitive Azure Site Recovery product, presents CloudEndure’s wider functionality to its field sellers and partners for certain use cases.

CloudEndure, however, is not sitting still. Recent additions include a lower priced disaster recovery tier with slightly more lenient RPO and RTO timeframes, and a new Backup Tier offering that brings continuous data protection to workstations, as well as the servers that are covered in the company’s disaster recovery solutions.

Fitting the company’s vision of any-to-any workload mobility, CloudEndure also recently added disaster recovery support for VMware targets. While VMware represents older technology than the AWS/Azure/GCP targets that CloudEndure started out supporting, Shai points out that most customers still use VMware targets for disaster recovery.

Any organization looking for modern technology that will position them well for future cloud-to-cloud migrations as well as cost-effective disaster recovery in the public cloud would do well to keep CloudEndure on their vendor short list.