As Maxwell mentioned, Raid 0 is striping (i.e. 1 big drive). You can stripe across more than 2 drives. This can improve performance (mileage may vary). The minimum array size is 2 drives. There is NO redundancy. If one drive fails, all data is lost.
RAID 1 is mirroring (i.e. one drive mirrors the contents of a second drive). Depending on your controller, you can get up to double the throughput of a single drive if the controller is able to simultaneously read alternate sectors of both drives. The redundancy is provided by the fact that if one drive fails, you can break the mirror and continue with the remaining drive having a full set of data. The downside is capacity. You can only use half of the potential storage capacity.You can combine RAID 0 and 1, essentially having a mirrored striped array. The downside is that the minimum array size will be 4 drives and you can only add capacity by adding 2 drives at a time. With either RAID 0 or 1 or a combination, if a drive fails, you must essentially bring your system down and spend some time to bring back to full capacity (Break mirrored set; break failed striped set; recreate striped set with new drive; recreate mirror). You would not be able to use a fifth drive by itself in the array. Your maximum capacity (based on the drives you mention) will be 240GB.
RAID 5 is also known as striping with parity check. Minimum array size is 3 drives. Additional drives can be added at any time, 1 or more at a time.The error checking information is spread across all drives in the array (because of this, you lose some potential storage capacity to this information). Because of the error checking, you do take a performance hit (although for MOST operations other than the most intense I/O operations, this will be negligible).
(continued)
