ATA RAID Information

What is ATA RAID?

What is RAID? RAID stands for Redundant Array of Independent Disks, a way of combining harddisks together to enhance performance and backing up of data. I wrote this simplified article because I recently saw motherboards incorporating ATA RAID controllers. My friends are asking me what's RAID, thus this article.

Previously, whenever we talk of RAID, it's always in terms of SCSI (Small Computer System Interface) based harddisks. However, with the introduction of ATA RAID controllers, we are now able to use the cheaper ATA harddisks to link an array. Do we need a RAID controller for normal users? I will say no. The most simple RAID configuration requires at least 2 harddisks which are exactly the same. Although harddisk prices are going down, I still think it's not a worthwhile investment unless the data stored is extremely important. However, in that case, you're not a normal user are you?

In ATA RAID controllers, there are only 3 types of configuration. RAID 0, RAID 1, and RAID 0 + 1. The motherboards that I saw usually only support RAID 0 and RAID 1. I will explain both of the configurations. RAID 0 + 1 just combines the 2 previous methods. It costs the most (in terms of the number of harddisks), but it's the fastest, and the most reliable configuration. You will also need to check what kind of RAID controller do you have? An ATA100 RAID controller? This will determine what type of harddisks do you need. You need a minimum of 2 harddisk to form an array.

RAID 0

RAID 0 is useful when you're dealing with a huge file. Writing huge files to harddisks takes quite some time. A simplified explantion... RAID 0 configuration splits the huge file into a number of "smaller files", equal to the number of harddisks you have linked to the RAID 0 array. It will then write the files into the separate harddisks at the same time. This increases the speed of reading and writing huge files.

What are the Pros? Reading a 50MB file is definitely faster than reading a 100MB file. Accessing huge amount of data in the harddisk is faster.
The Cons? The harddisks linked to the array should have exactly the same capacity. The reason is because the total capacity of this configuration is equal to the smallest capacity of the harddisks array, multiplied by the number of harddisks you have. Example: 4 10GB hardisks, and 1 50GB harddisks, will have a total capacity of 50GB (10GB * 5 harddisks). 40GB of space is wasted.

Frequent reading and writing of harddisks will wear the harddisk down. A failure of any of the harddisks, means that all your data is lost.

What's the use of this configuration? Well... Most people uses it for retrieval of information from big databases. I've been told that capturing video is also one of its uses, as the access speed of the harddisk is critical to the speed of the video capturing.

RAID 1

RAID 1 is useful for backing up of data. It mirrors all the information in the harddisk into the other harddisks in the array. If you have a array of 4 harddisks, you will have 3 backups.

What are the Pros? Data security... Gone are the days where you worry about what will happen when a virus strikes your harddisk and wiped out everything. :) When 1 drive fails, the other will take over automatically.
The Cons? Similar to the RAID 0 array, the smallest capacity of the array is the total capacity of the array. Example: 2 10GB harddisk, 1 20GB harddisk, will have a total capacity of 10GB, with 2 backups.

This is an expensive way of backing up of data. Basically, backing up a 50GB harddisk requires another 50GB harddisk.

This configuration is used for backing up important data. It backups and recovers data seamlessly, useful for web servers and file servers.