I hope everyone had a good holiday weekend last week!
I spent the majority of last week (and a portion of this week) working with Ryan on server installation and maintenance. This takes a bit of work with a technology called RAID. Those of you who have a server in your business may very well have heard the term, but not know anything about it, so I figured I’d talk a little about what RAID is in this week’s post.
RAID stands for Redundant Array of Independent Disks. It is a system used for, as the name suggests, redundancy purposes in data storage systems. Normally you’ll see it in use on a server, though a home computer may have it as well if it’s high-end or custom built. The system uses an “array”, which is a term meaning “a group of hard drives connected together”, managed by a RAID controller, to arrange data over said hard drives in specific patterns.
The pattern used depends on why you want to use RAID in the first place. There are several options – I will only list the most commonly used ones here:
- RAID 0 is used to essentially turn two physical hard drives into one “logical” drive. As a result, once the computer is running they appear to be a single, larger drive. RAID 0 can also be used to almost double the speed of each individual drive due to how it handles reading and writing data. This array does not have any redundancy – if one drive fails, all your data will be lost because it stores parts of it on both drives.
- RAID 1 is used for extremely basic redundancy. The hard drives are made into exact copies of each other. Any changes made on one are mirrored on the other. This array is able to lose one drive and still retain its data. However, it is only meant to be a failsafe, not a true backup.
- RAID 5 is one of the most common arrays in use in small businesses. The drives are arranged in such a way that one can fail, and the system will keep running. This gives the system administrator (or your friendly neighborhood computer shop!) time to replace the failed drive without downtime or loss of data.
- RAID 6 is an extension of RAID 5. Instead of one drive, the system can lose two before it will fail. This gives the administrators more time to get the equipment replaced.
There are a few other RAID options, but most of them are just combinations of the ones listed here, and don’t offer a huge benefit over one of these. I should also note that this is not a technical description – there are a lot more details involved in how it works! If you would like to read more, dig in to the Wikipedia Article on RAID.
It’s important to note that having a RAID does not mean your data is backed up. Yes, it’s more likely that you will be able to keep a server running if a hard drive fails. But recovering data from a failed server array is much, much harder than with a standard failure. In many cases it’s nearly impossible. So it’s still a great idea to have a dedicated backup set that keeps your information on a separate drive (or drives) in a safe place.