Click to learn more about author Ken Steinhardt.
If you are depending wholly on data backup for disaster recovery and business continuity, you may want to take a second look at your strategy. For mission-critical environments, data backup should be, at most, a secondary level of protection and not the primary form of data protection.
The recovery point objective (RPO) and recovery time objective (RTO) for backup after almost any kind of disaster is simply insufficient for meeting today’s service level agreements (SLAs) for virtually all mission-critical environments.
The best way to address this challenge is by using data backup for what it does best, serving as the “backup” to a superior first-line of data protection that uses real-time data replication in an active-active configuration.
Active-active replication can provide zero RPO at zero RTO between different systems. It should be enhanced with frequent snapshots to provide a way to recover data faster (superior RTO) with less data loss (superior RPO) than invoking a recovery from backup.
The Biggest Challenge
The biggest data recovery challenge is not being able to restore from the backup fast enough. When data recovery from backup is required, it probably means that either the primary business continuity protection mechanism has failed, or else specific data has been either corrupted or accidentally deleted. At that point, you have already accepted a potentially bad RPO, and now everything depends on at least minimizing the RTO.
In an effort to reduce cost, many disk-based backup solutions provide extensive data reduction techniques, particularly for data deduplication. Unfortunately, the more advanced the data reduction, typically the more performance-constrained will be the recovery using such systems.
What you might have saved in backup cost could come back to haunt you in the form of a really slow recovery time, when recovery time is what matters most. If the backup is still to tape, then you not only will likely experience a slow recovery but a significant risk that data has been corrupted and might not restore at all.
The best way to address this challenge is to deploy solutions that are capable of rapid recovery for fast RTO by using the intelligence of software and caching while using disk media for the persistent storage of the backup data. This provides an ideal balance of minimal RTO, onboard data protection to prevent data corruption, and low cost.
Tape is cheap but too slow for RTO, and tape can’t protect well against data corruption. Most disk-based backup systems are too slow for recovery when RTO is the most important factor. Flash-based backup systems can provide good RTO but are usually extremely expensive.
Too Much Is at Stake
Never stopping a data-driven business service is high up on most senior business leaders’ priorities within large enterprises. If a disruption were to happen, an existing data backup system might not be enough to ensure that IT can deliver the business continuity. You don’t want to be caught unaware.
The active-active configuration of a real-time data replication solution enables an IT team to maintain business continuity while better managing costs through a flexible approach that simultaneously reduces the risk of an enterprise’s business service going down.