Which RAID setup is right for your business

RAID stands for Redundant Array of Independent/Inexpensive Drives. This setup is used by companies – and sometimes individuals – for two reasons: to improve the performance (i.e. the read/write speeds) of their storage media or to improve reliability by creating an automated backup plan that provides an effective and prompt form of retrieving data from a failed RAID should it be necessary. It is even possible for RAIDs to improve both performance and reliability if certain setups are utilised, though these tend to be more expensive than frameworks designed to serve just one of these purposes.

All in all, this means that choosing the right RAID setup for your business will differ depending on your requirements and budget. With this in mind, we thought we’d discuss several of the most common RAIDs – along with their various pros and cons – to help you make the best decision for your business.

RAID 0

A RAID 0 setup is often simply referred to as a ‘striped array’ and is designed to offer pure speed. It features several drives and any data written to it is split across each of these. The result is exceptional performance – but this comes at a price.

As the individual bytes that form any file are split across several drives when stored in this type of array, all of the data stored on here will be lost if just one of these fails. Should one drive fail, a percentage of the data that forms the file is lost and – just as a jigsaw with missing pieces cannot be completed – the file cannot be reconstructed.

Whilst RAID 0s offer impressive read/write speeds, they are rarely used alone due to the risk of data loss.

RAID 1

RAID 1 is setup in such a way that all data is stored to two separate drives. Essentially, each drive used within such arrays all contain a clone of one another i.e. if a RAID 1 contains just two HDDs, these drives will hold identical sets of data.

In the event that one drive fails, this setup is able to immediately read the data from the relevant backup drive and continue working as usual. This means that RAID 1 is a fantastic way to store your company’s most important data, particularly as it also offers impressive read/write speeds (even if they do pale in comparison to RAID 0).

Sadly, RAID 1 is in no way a cost-effective option as it halves the potential storage capacity of an array. Say, for example, the array could hold eight separate drives, if eight 10TB HDDs were used, the total capacity would still be just 40TBs.

RAID 5

Like RAID 0, RAID 5 uses striping in order to improve performance. Unlike RAID 0, however, it also offers some protection against data loss.

RAID 5 not only copies all files that are saved to it to at least two separate drives but also has a parity drive present in each block of drives. Should one of the drives fail, the parity drive/s can be used to rebuild the relevant data and restore the failed drive.

Whilst this is a sensible means of achieving superior performance and simultaneously create a backup plan, it should be noted that rebuilding a drive’s data can take more than 24 hours and that should another drive present within the array fail during the recovery process, all of the data contained in the array will be lost.

RAID 6

With RAID 6 setups, the number of parity drives used in each block of drives is doubled to two. This means that two drives can fail in such an array, resulting in added security. There is a price to pay, however, as this setup does adversely affect write speeds.

RAID 10

A RAID 10 setup is, to put it simply, a combination of a RAID 1 and a RAID 0. Basically, data is split across several drives but each also has their own mirror drive. As a result, this setup provides both excellent performance and a reduced risk of redundancy.

But please don’t forget to backup

Whilst RAIDs are often a fantastic way of reducing the risk of data loss, they do not offer a comprehensive disaster data recovery plan so please, please, make sure you use a backup plan as well as RAIDs.