Home Articles Big Data The End of Backup, As We Know It

The End of Backup, As We Know It

The End of Backup, As We Know It
The End of Backup, As We Know It

Organisations must always be prepared for a disaster scenario. Whether it’s a cyber-attack, a natural disaster or a system failure, companies of all sizes need to formulate disaster recovery plans that will allow them to continue to operate when something goes wrong.

The traditional approach to disaster recovery, particularly for large enterprises, has been to build and maintain a separate datacenter at a safe distance from the primary datacenter site, maintained with expensive high-speed WAN connections between the two locations. The doubled cost of maintaining a fully operable secondary datacenter just for DR was never really viable for companies with smaller IT budgets. 

Another more economical choice is employing a backup solution to ensure data is not lost. The backup solutions employed by most enterprises involve shipping incremental daily backups to off-site storage. This approach leads to recovery times and recovery point objectives measured in hours or days.

But backup solutions, as we know them, are built on an archaic principle of scheduled backup windows and lengthy, productivity-halting restore sessions. And waiting hours or even days for availability is simply unacceptable in modern enterprise standards.

The problem with the backup and restore approach is that it is error-prone, slow, and tends to fail when needed the most. The most common comment from companies that have been hit by a major data loss incident such as a ransomware attack, is that they realised only in retrospect that their backups were not as good as they thought. This caught them off-guard with a recovery process that was way more complicated and time-consuming than they had imagined. 

The takeaway from this is that in a world of rampant cyber-crime, ransomware, and state sponsored attacks, legacy backup solutions are woefully inadequate at keeping data secure.  Organisations need to shift their emphasis from backup and restore to intelligent storage solutions that are self-defending. A new security strategy is needed, one that is based on three major innovations in data storage: Continuous data protection, instant disaster recovery, and proactive disaster prevention.

Continuous data protection

Periodic backups, based on backup windows, are being phased out for modern storage solutions that provide continuous data protection in the cloud. Continuous data protection is a technique that continuously captures changes to data in real-time and streams these changes to a cloud-based, immutable repository. This means a file in use can be changed and that change is immediately captured and stored. This is in contrast to traditional backup techniques, which typically only capture data once or twice per day. The benefits of having continuous data protection are that it allows for recovery from data loss much more quickly and easily, and it allows recovery to any point in time. For example, if an organisation’s data was damaged by ransomware, it can simply revert to a very recent previous version, minimising the amount of lost work.

Previous articleWhat ‘flavour’ of cloud is right for your organisation?
Aron Brand, CTO of CTERA, has more than 22 years of experience in designing and implementing distributed software systems. Prior to joining the founding team of CTERA, Aron as Chief Architect and member of the founding team of SofaWare, where he led the design of cybersecurity products and appliances until its acquisition by Check Point Software.