Working on my COMPTIA A+ Cert. and I came across this topic. While I understand how it works, I don’t understand why you would use it over individual drives?
In: 1
Depends on the raid level.
Raid 0 is just spreading data among multiple drives, with the goal of improving speeds (since read/write tasks will be split among drives). However there is zero redundancy: a single drive loss will cause a complete loss of data.
Raid 1 is purely mirroring among drives: there is no performance improvement in write tasks since all data must be written identically to multiple drives. However even in a two drive array, a single loss results in no data loss.
Raid 5 is a bit of both: data is spread among drives, but the equivalent to one drive’s space is used for redundancy, allowing a single drive in a 3+ drive array to be lost without losing data.
RAID provides two primary advantages, and 3 technically but the 3rd isn’t as big a deal
1. increased performance – Having multiple drives in an array allows you to leverage the performance of all of those drives, so you get much faster read and write speeds. There’s a performance penalty associated with RAID so you don’t get 100% of the performance of all the drives, particularly RAID 5+6, but overall there’s a gain in performance.
2. Redundancy – RAID allows you to lose 1 or more drives in an Array (depending on the RAID level) without losing any data. Most RAID arrays allow for drives to be hot-swapped or replaced and the array rebuilt live without downtime. For big business this is critical as it allows the server to continue operating without data loss while it is repaired.
3. Useable Space – Having drives in an array means that you can manage it as one large block of storage, which is much more efficient than trying to manage multiple independent drives.
Redundancy and performance.
For example: instead of having just one hard drive, you have two identical drives which contain the same exact data. This is called RAID 1.
This has two advantages: First, there’s always a backup – if one of the drives dies, you just replace it with a new one and copy the data from the working drive. Secondly, you can read twice as fast, by reading half the data you need from each drive.
Other RAID configurations provide even better performance and better options for redundancy, usually by using more than two drives.
Drives fail. It’s a fact of life and there’s no way to know when each drive will fail.
If I have a hard drive at home that fails and I don’t have a backup, my data is gone forever (or be forced into spending money on an expensive recovery service, which probably won’t be 100% effective).
If I have a RAID 1 setup at home, one drive can fail, and my data is still there on the other drive. I can then replace the failed drive and the mirror is restored. I am protected again.
Another reason for RAID is to have your data spread out on multiple drives. If all of my data is spread out on 10 drives, I will probably be able to read out all of that data faster than if it was all on one single drive. The single drive becomes a bottleneck.
– you can store files larger than the size of one whole drive
– You don’t lose any information if only one drive fails
– You can just change a drive if one fails and keep the array running and readable
> I don’t understand why you would use it over individual drives?
When a drive fails in a computer using individual drives, then it usually takes down either the whole machine (OS is dead) or renders it functionally useless (the data it is supposed to serve/store/process is lost). If you have a backup, you have to spend time restoring it onto a new drive. If you don’t, you also permanently lose data and even more work hours to recreate the settings on the machine.
If you are running RAID1/5/10/etc. (not RAID0), then what happens is that you get a log message/mobile alert/email about a drive failing, you walk up to the still working machine, pop the drive out of the hotswap drawer and put a new one in, then add it into the RAID array in the console. The system will automatically start copying the data from the surviving disk(s) to restore the redundancy. Altogether 0 seconds downtime.
Latest Answers