CDP vs. Snapshots: Which Provides Better Data Protection?

Need a kick in the butt to bolt down your data protection strategy? Look no further than the tumultuous world of cybersecurity. According to an annual study conducted by Ponemon Institute, the average cost of a data breach reached $3.86 million in 2018, a 6.4 percent increase over the previous year. The study also reported an increase in the average cost of  individual stolen records, suggesting what most us already know ― our data is more valuable than ever. When it comes to safeguarding your data, strategical options span well beyond conventional backups. In this post, we explore two of the most popular data protection technologies, continuous data protection (CDP) and snapshots.

CDP Explained

Some business processes are so critical that the slightest amount of downtime could prove detrimental to the organization. CDP is designed with these mission-critical processes in mind. Continuous data protection copies or replicates data as it is written to disk. Whereas replication in general does not create recovery points, CDP keeps track of all changes that take place over time. As a result, organizations have the luxury of point-in-time (PIT) capabilities that make it possible to recover data from multiple points in the history of the system. The 'always-on' component of replication enables continuous data protection to offer a substantially lower recovery point objective (RPO) than traditional backup solutions. CDP essentially captures each point in time, which in turn, creates granular recovery possibilities. While achieving zeros across your recovery points is a lofty objective, this technology can help organizations incur considerably less data loss when disaster strikes.

Snapshots Defined

Speaking of PIT, it is literally what snapshots are made of. A snapshot is basically the state of a storage system captured at a given point in time. Preserving the system state not only allows data to be recovered in the event of failure but restored to known working points. This is helpful in various scenarios. For example, snapshots can be used to roll back to a time before installing and uninstalling software, updating the system, or upgrading hardware components. They're also useful for developers when testing and validating application code in virtual machines.Snapshot offer a quick and effective way to rewind a system to desirable points in time. While they can be considered a fail-safe of sorts, they are not a reliable backup alternative. Snapshots reside on the systems they are designed to protect and thus depend on those systems to perform recovery operations. They can also lead to performance issues over time as more points are created on the system. Conversely, backups can existence independently, travel across multiple storage systems, with little to no impact on live production environments.

The Verdict

We often close these comparison posts by stating something along the lines of “there is no right or wrong”. Dedicate some time to figuring out what's best for your organization. That would be downright disingenuous in this case. Snapshots offer convenient, on-demand recovery and are increasingly being integrated into commercial backup solutions. Still, snapshots alone do not provide enough protection to replace backup. Disaster recovery demands the utmost retention and flexibility that provides easy access to data across multiple platforms. On the other hand, continuous data protection is more like snapshots on performance enhancers. Unlike replication technologies that employ synchronous mirroring, CDP is designed to prevent data affected by malware, corruption, or human error from being restored into live production environments. The ability to restore a system to any given point in time makes it an ideal data protection solution for servers, databases, and database-dependent applications alike. If you think about it, continuous data protection is an ongoing diary of complete snapshots. CDP clearly has the edge in this battle. However, snapshots have their place in a comprehensive data protection strategy. They could potentially play an even bigger role moving forward as organizations increasingly have more data, yet less time to protect it. Whether individually or paired together, these technologies can have a major impact on how quickly and effectively data can be recovered.