As the repository of last resort, backups are unique. They essentially store every piece of data in the company, at least all the data that’s worth saving. In order to do this, modern backup systems, which are often disk-based appliances, leverage data reduction technologies to increase effective capacity. But data is changing and backup systems are being asked to do things they really weren’t designed for, like archiving, a practice that’s causing problems for IT.
Click to register for the upcoming webinar "Four Assumptions that are killing your Backup Storage"
Backup collects and stores a copy of the most active data which allows users or applications to get back the current copy and often previous versions as well. Backup provides insurance against data loss and, to the extent that it can restore data quickly, a mechanism to maintain productivity.
An archive, on the other hand, is a system for storing file data, often files that aren’t used as much but still have value. Archives are designed for long-term storage of those files and typically provide reasonably fast access, usually without the intermediate step of being restored to a cache or faster disk tier as backup data requires. For this reason archives are becoming more disk-based.
Backups are changing
While databases still contain much of the critical data, the real growth in backup data volume is coming from unstructured or file data. These digital content data sets are often large files that don’t change much, if at all, and represent significant value so they’re kept for a long period of time. As backup systems start to fill up with these new digital assets, they look more and more like archives. Unfortunately, they’re still designed for the more traditional backup data sets; therein lies the problem. Disk-based backup appliances need to be redesigned to support this more archive-like use case.
New designs for enterprise backup
Data sets are expanding as file sizes increase and to make matters worse, retention times are increasing as well. This is creating a ‘perfect storm’ for the backup system since it’s being asked to keep up with this exploding data growth. The new enterprise backup system needs to scale incrementally into the PB range while remaining economical.
Doing this requires a different architecture than traditional RAID, one that can support the high capacity drives needed to keep costs down but not be subjected to long RAID rebuild cycles. An object storage architecture using erasure coding can provide that resiliency without generating extra copies and still has the scalability that businesses need to accommodate data growth.
To learn more about how changing backup environments are driving new design requirements for disk backup appliances, join Storage Switzerland and NEC for the live webinar: “Four Assumptions that are killing your Backup Storage”.