Backup software vendors are taking great strides to adapt their software to accommodate new recovery point objective (RPO) and recovery time objective (RTO) pressures that increasingly require restoring data immediately and with zero data loss. For example, vendors are investing in block-level backup capabilities to improve the efficiency of backup jobs by identifying and then backing up only the modifications made to a file, as opposed to backing up the entire file multiple times. Replication and instant recovery capabilities are also becoming more common, which enable spinning up workloads on the fly.
The steps that backup software vendors now take to facilitate capabilities and levels of performance that are closer to production storage levels are impressive. However, for a truly modernized backup approach, IT professionals should also look for some additional capabilities. Chief among these additional capabilities is intelligent tiering of data across multiple storage types. It now makes more sense to selectively integrate faster-performing storage media, and even solid-state drives, as a component of the backup storage environment to meet stringent RTOs and the need for fast application performance while in a failover state. At the same time, storage retention requirements continue growing massively due to explosive data growth, increased data utilization, and stricter data privacy regulations. Put simply, more data must be stored longer (and potentially forever). The ability to migrate older data to lower-cost and scalable on-premises object storage or cloud storage services for archive and long-term retention is necessary to meet these large and quickly growing capacity requirements while keeping the budget in check.
Metadata tracking is another important capability. Metadata describes other data, for example documenting which user most recently accessed a file, and when. More granular and easily accessible metadata can be used to set data tiering and retention policies. It also can be used to identify files and to provide a file version history. This can help the enterprise to comply with eDiscovery and “right to be forgotten” requests, whereby a user requests that their personally identifiable information (PII) be stripped from the enterprise’s data stores. Along this same line, staged recoveries can also help to comply with privacy requirements, by avoiding recovering data that is supposed to be forgotten.