Data is at the heart of any solution. It's what the entire system runs on and without it you're going nowhere fast. Therefore, we put the same high levels of investment into our data protection as everything else. What's more, we don't charge extra for these services. We feel you should be able to expect nothing less from your provider.
The most likely item to fail in a computing system is a hard drive, due to the moving parts. By using RAID systems, we can prevent an individual drive failure having any impact.
RAID works by storing all the data in the system on multiple disks, so that if one fails, there is another copy that the system can keep running on. This is called mirroring. By using RAID 10 we can combine this resilience with a performance boost from multiple disks.
Any drive that does fail is replaced and the system rebuilds in the background. You won't even notice it happen.
With Tier 1 we protected against single drive failure. However, what happens if there is a more serious hardware failure? For every SAN we use, there is an entire unit running as a high availability spare. This is known as 2N protection. Data is written to both devices and in the event of a problem, the spare unit seamlessly continues operation.
Silent data corruption can occur on any storage platform unless proper protection is put in place. It is a problem that is well documented, but rarely acknowledged and even more rarely dealt with properly.
Data can be corrupted at many stages, including every read and write of each bit. This can happen at every level of hardware, unless adequate checks are put in place.
Our systems employ block level checksums to ensure that data is always consistent. Not only that, but we continually walk through data on our SANs, even infrequently accessed data, to check that it is not degrading on disk.
If you want complete data protection, adequate silent data corruption checks are a must have.
The next tier of protection is full system backups. We use the power of our SAN systems to take full images of every server and we do this every hour. We provide the following images as standard:
• Hourly backups, kept for 24 hours
• Daily backups, kept for 14 days
• Weekly backups, kept for 4 weeks