Data Recovery Digest

Traditionally, data recovery plans relied on tapes or mechanical hard drives, often in facilities that are located off-site. However, the technology and storage landscapes have changed radically, as they always continue to do so, and things are moving more towards cloud backup. More enterprises are phasing out on-premises and off-site tape and favouring online cloud storage.

It’s all well and good backing up to the cloud, but the question remains as to what the recovery process is. Applications need to run efficiently and can send and receive data quickly – with local speeds still far greater than those over an internet connection, uploading recovery data can take a long time if there’s a lot being produced.

While the simple answer to this would be to run the recovery in the cloud and move the apps there too, this isn’t an easy thing to do. While modern applications are likely developed with the cloud in mind, and can fully support all its capabilities, a lot of enterprises make use of legacy applications that just don’t have the framework to support the cloud.

To work towards this, one option might be to use a telecommunications facility as a standby. These will likely have quick links to all major cloud data centres, and have the mainframe gear available to support it, which would then reduce data access speeds to a user-friendly level.

Cloud providers themselves are also making improvements, moving away from tape and favouring disk drives, ideally solid state ones. When recovering from a backup, it isn’t necessarily every single file that you need. In fact, it might only be a few. If this was on tape, searching for that data would be a very slow process because it will need to be run sequentially.

Solid state drives are different – not only do they not have any physical parts, meaning they are quicker and have less wear, but they can also pull up data anywhere on the drive without needing to search through everything else in turn. Recovery processes, therefore, operate much more quickly, meaning you can recover a specific set of data with ease rather than pulling from the entire set.

If you are to go with a cloud provider, the location you opt for is important. Or, at least, the location of their servers is important, as is their redundancy plan. If the server is in the same city as your primary data, you might want to consider what would happen if a physical disaster were to strike – your backup would be taken out and so would your actual data.

As network speeds get quicker and more applications support the cloud, it’s likely that most of our recovery is going to take place over the internet. And that’s a good thing if the speed and reliability is acceptable. But remember that, as with any recovery solution, you need to account for security too – ensure that the network you’ll be recovering from is encrypted and safe, otherwise you could find that your valuable data is intercepted and falls into the wrong hands.