Before the turn of the 21st century, backups were simple—mostly because the amount of data enterprises utilized was a tiny fraction of what is used today. The increasing volume of data in databases and email, unstructured data and other data types has been evidenced by orders of magnitude growth in the amount of data requiring backup—from gigabytes to terabytes to petabytes in just a few short years.
Most organizations require that several retention points be kept for legal discovery, regulation audits, financial audits and other business and legal needs—for example, daily, weekly and monthly backups. Thus, it is common for an organization to maintain around 40 backup retention points, with the ability to restore any of them if and when needed.
As a result, data deduplication has become critical in helping reduce the amount of data center backup storage required to meet these ever-growing demands. However, all deduplication was not created equal: Results can range from 2:1 to 20:1 depending on the technique used. Additionally, many solutions perform deduplication in line, which slows down the backup and restore process and affects backups in other ways.
Here are the top challenges organizations may see as backups grow.
- The economic impact of the amount of storage required to meet data center backup storage needs.
- The need to provision enough WAN bandwidth for backup and disaster recovery—and the bottom-line impact of mostly unused bandwidth.
- The effects of inline deduplication on backup performance.
- The data rehydration penalty, which can slow down restore performance by a factor of 20.
- Shrinking backup windows or a lack of scalability when growth in the amount of data being backed up causes the backup windows to grow in unison. If the backups are outside the backup window, then backups need to be halted before being complete.
- An increased chance of successful ransomware attacks due to the fact that network-facing backup storage can be deleted and backup data is unavailable to restore after a primary storage encryption event.
The Tiered Storage Difference
ExaGrid combines the best of two worlds, both writing direct to disk and providing a deduplicated data repository to deliver tiered backup storage, which solves the six big challenges outlined above. Tiered storage provides:
- Disk-speed backups to complete backups in the proscribed window.
- The fastest possible restores and virtual machine boots to keep users productive—meaning VM boots in minutes vs. hours with other architectures.
- A unique scale-out architecture that adds processor, memory, network and storage at the same time, so the backup window never grows as data grows.
- A separate retention storage tier that is not network facing, supports delayed deletes and contains immutable data objects that are ready for recovery from a ransomware event.
- Aggressive deduplication that economizes both storage and WAN resources to lower costs and that scales out to eliminate forklift upgrades—regardless of their age, size or model number, ExaGrid appliances can be mixed and matched in a single scale-out system.
Future articles will take a more in-depth look at a few of these challenges, the impact they can have on the organization and how to overcome them at scale.