Data is exploding and doubling every two years. Did you know that in 2011 the world created a staggering 1.8 zettabytes? (Source: IDC). By 2020 the world will generate 50 times the amount of data. With this increase in data, Administrators are being pushed to come up with cost-effective solutions to protect this vital information.
One of the most popular backup methods is to use Network Attached Storage (NAS) with RAID 5, as it provides the maximum bang for your buck. Administrators believe that this strategy is a good way to protect their critical data.
But, before selecting RAID as the sole backup solution, think of this:
The whole reason disk is being deployed for backup is to increase speeds so that it can complete within the acceptable backup window. Writing to RAID actually is slower than writing to a single disk, with the exception of using RAID 0. However, RAID 0 does not provide redundancy. When a RAID system suffers a failure on one of its drives, the entire array has to be rebuilt, reading all the drives. As drives get bigger, rebuild times can stretch from a few hours to days. While the system is rebuilding, the performance gets degraded to the extent that the system will not work to its full potential.
Statistics show that drives in a RAID set fail more frequently than individual drives. Besides, the probability of a second drive failing during a rebuild is very high when using large arrays. Disks are commonly specified with an unrecoverable read error rate (URE) of 10^14. This means that once every 100,000,000,000,000 bits, the disk will not be able to read a sector back to you. That is a hard-to-imagine number of bits; it’s is roughly about 12TB. So, as the RAID controller is trying to read through those disks to reconstruct the data from the failed drive, it will most certainly see a URE. When this read fails, the RAID is toast. All of that data is gone.
Off-Site Storage for Disaster Recovery
Best practices advocate keeping a minimum of 3 copies of the backup on 2 different media and at least 1 copy should be off site for disaster recovery purposes. The NAS box does not satisfy this rule in any way since all 3 copies are on the same single system.
Protection against Other Elements
Since data is online, there is always a chance for accidental deletion or user error. There is no protection from viruses or malware. Data corruption because of other failed hardware or power loss also cannot be protected against. An online backup provider lost data for 7,500 users because of RAID issues.
Redundancy is not the only requirement. The first line of defense is keeping an off-site copy of your data. The second equally important requirement in this day and age is to also have an off-line copy of your data so that it cannot be affected by malware, viruses and ransomeware.
A good case study is Codespaces which was a victim of ransomeware. Despite having data in multiple locations and boasting ‘full redundancy,’ they went out of business practically overnight because all their data was online including their backup.
Shamelessly promoting our product
Idealstor addresses all of these issues with our removable disk-based backup system and more. Idealstor uses disk cartridges instead of tape, thereby offering the speed benefits and reliability of disk while maintaining the portability of tape. It enables companies to keep multiple copies on individual disks which spreads the risk in case of any failure. Throughput speeds are faster than cloud, and each disk can become a separate target for the data. The disks are ruggedized to be taken off site for disaster recovery. They can be connected directly to any system to start restoring data immediately in case of a disaster.
While RAID is a popular choice for data backup, Idealstor’s solution provides a truly cost-effective, safe and reliable approach to data storage and recovery for technology professionals.