Cohesity Delivers a New Way to Approach Disaster Recovery

Cohesity has announced an automated disaster recovery offering it says is the only web-scale converged solution to protect applications across tiers, service levels and locations on a single platform.

SiteContinuity, which integrates with the company's backup and continuous data protection (CDP) capabilities, aims to better protect critical applications by converging backup, continuous data protection and disaster recovery (DR) capabilities. According to Cohesity Senior Director Raj Dutt, it will allow organizations to recover as little as a single file or as much as an entire data center quickly, with near-zero application downtime and data loss, through a global user interface and a unified policy framework. It works across application tiers, service levels and environments.

Under the covers, Cohesity is using snapshot-based backup and CDP to capture and/or move applications and data. Once the data lands on the Cohesity platform, SiteContinuity automatically replicates the data to the disaster recovery site based on the user-defined service-level agreements (SLAs).

To achieve near-zero data loss, Cohesity’s CDP intercepts I/Os between the protected VM and its underlying storage. This ensures that every I/O change in the production is captured and replicated almost instantly to the DR site to achieve near-zero recovery point objective and data loss, Dutt explained. Built-in intelligence allows the periodic consolidation of log deltas and the foundational VMDK (virtual machine disk).

Converging backup, CDP and disaster recovery capabilities is kind of like having an "RPO/RTO slider" with unified management, said Christophe Bertrand, a senior analyst at Enterprise Strategy Group. "Having that type of capability allows you to set the dial where you want it based on the criticality of applications and data. It's a pretty unique combination that can cover a lot of ground in one platform."

Dutt said the success of the system is based on the SLA-driven policy framework. For example, the user defines the desired SLAs and the sequence of operations in the event of a disaster. If the user defines near-zero application and data loss, Cohesity captures every I/O change and instantly replicates it to the DR site. This data is ready to be used and requires no restore. To avoid downtime, at the time of a disaster, the user, with just a few clicks, automates the entire failover orchestration process.

It works similarly for non-mission-critical applications and data. Based on the desired SLA, SiteContinuity uses snapshot-based backups and then replicates that data to another backup infrastructure at the DR site. This tier of data does require a restore process, Dutt noted.

Any technology that helps organizations meet critical recovery objectives is a step forward, Bertrand said.

"Our research shows that 15% of organizations require RPO=0, meaning no data loss at all, and the same number — likely the same organizations — need RTO=0, or essentially high availability," he said. "Data is the business, and any interruption is costly directly and indirectly. Data protection is really a business problem when it's not up to par."

Bertrand also commented on the solution's flexibility, which he said is key to keeping pace with the constant pace of change. "Organizations need to have solutions that help them today but also tomorrow, particularly in light of accelerated digital transformation."

SiteContinuity will be generally available in the fourth quarter of 2020.