RAID is an acronym and stands for Redundant Array of Indipendent Disks. It’s a technique which allows to use serveral disks together to increase the reliability of a storage system, the performance or both at the same time.
There are many kinds of raid configurations which are usually identified by a number, some are common some are very rare. The most common are:
RAID 0 (stripping), RAID 1 (mirroring), RAID 5 (a mix of stripping and mirroring).
Just to give you and idea here’s how they work:
in mirroring it’s very simple, every data that’s written to a disk is written to another one as well which serves as backup, so if one disk fails you still have a online copy and you can keep working without loss of data. When you replace the faulty disk the spare one gets copied on it and you’re back in a safe configuration.
In stripping you split your disk read and write operation across several disks which will carry it out simultaneously. In this way you vastly increase the performance of your system. But the risk of failure is higher than with a single disk because if you have for example a stripe set of three disks and a single one fails you loose all the three disks contents.
RAID 5 is the most common configuration and uses a redundancy technique called XORing, I’ll spare you the details but with it you take 3 disks (at least) and use two of them like they were in stripping (so you have an increase in performance) but for every two bits that you write on the disks you put an extra one on the third disk. This extra bit is calculated in such a way that if one of the disks fails you can rebuild its contents using the redundancy bit. So in RAID 5 you get the benefits of stripping perfomancewise but with the safety of mirroring using less disks than you would otherwise need.
I hope this clarifies you things 
