In this section, I discuss the most common RAID configurations. However, there are other, more obscure types available to you as well. To learn more about the RAID arrays I talk about below, as well as the ones that I don't, I recommend visiting http://www.acnc.com/04_01_00.html.
A RAID 0 array is what's known as a stripe set with no fault tolerance. The basic idea is that multiple physical hard drives are treated as a single volume. Rather than only combining the storage capacity of those drives though, a RAID 0 array also combines the performance of each drive by simultaneously writing to each one.
For example, suppose that you needed to write a file to a RAID 0 array consisting of three hard drives (the minimum requirement for a RAID 0 array is two drives). The file would basically be broken into three pieces, with a third of the file being written to each drive. Since the write operations occur simultaneously, the file is written to disk roughly three times faster than it could be written to a comparable single disk.
The advantages of a RAID 0 array are its speed and that the full capacity of each drive in the array is available for storage. The disadvantage is that RAID 0 offers no fault tolerance. If a drive within the array fails, the entire volume is destroyed and all data is lost.
RAID 1 involves a technology known as mirroring (sometimes called disk duplexing). The idea behind RAID 1 is that any data written to one drive is also written to another. If a drive fails, you've got a copy of your data already stored on another drive.
There are a couple of disadvantages to RAID 1 though:
- The first disadvantage is speed. When it comes to writing data, RAID 1 is no faster than a comparable single disk. Although, some RAID 1 implementations allow you to read data at double the speed of a comparable single disk by reading half of the data from each drive in a simultaneous read operation.
- The other disadvantage to RAID 1 is overhead. You've got an entire hard drive that is being used as a spare. Therefore, it can't be used for data storage in the traditional sense.
RAID 0+1 is basically a mirrored RAID 0 array. RAID 0+1 provides the same level of fault tolerance as RAID 5, and the same overhead for fault tolerance as RAID 1. In plain English, this means that if a drive in your RAID 0+1 setup fails, the array continues to function.
However, because you're mirroring an entire array, there is a lot of overhead. For example, if your RAID 0+1 array consists of 10 hard disks, you are only going to be able to use five of those disks for storing data. The other five act as a mirror.
The minimum requirement for setting up a RAID 0+1 array is four drives (a mirrored set of a two drive stripe set).
A RAID 5 array is very similar to a RAID 0 array, but with one major difference: fault tolerance. RAID 5 arrays are usually referred to as stripe sets with parity. What this means is that a RAID 5 array is a stripe set just like a RAID 0 array. But unlike a RAID 0 array, if a disk within the array fails, the volume will not be corrupted. In fact, the volume can continue to function while the system is waiting for you to replace the failed disk.
So what's the catch? The fault tolerant abilities of RAID 5 come at a cost of one of the drives in your array. For example, if your array consists of five hard drives (three is the minimum), then you will only have access to the capacity of four of those drives. That's because a portion of each drive stores parity information used to keep the array going should a drive fail.
When the failed disk is eventually replaced, this parity information is used to replace the data that previously existed on the failed disk. Regardless of the size of your RAID 5 array, the parity information consumes the equivalent of one disk worth of data.
Writing parity information and data to each drive also has an effect on throughput. A RAID 5 array is not as fast as a comparable RAID 0 array -- but the decrease in speed is often worth it given the array's fault tolerant nature.
CRASH COURSE: EXCHANGE SERVER 2003 STORAGE MANAGEMENT
Part 1: Microsoft's Exchange Server storage recommendations
Part 2: RAID configuration options for Exchange Server storage
Part 3: Using a SAN for Exchange Server storage
Part 4: Using NAS for Exchange Server storage
Part 5: Using DAS for Exchange Server storage
Part 6: Related resources on Exchange Server storage management
|ABOUT THE AUTHOR:|
| Brien M. Posey, MCSE
Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Exchange Server, and has previously received Microsoft's MVP award for Windows Server and Internet Information Server (IIS). Brien has served as CIO for a nationwide chain of hospitals and was once responsible for the Department of Information Management at Fort Knox. As a freelance technical writer, Brien has written for Microsoft, TechTarget, CNET, ZDNet, MSD2D, Relevant Technologies and other technology companies. You can visit Brien's personal Web site at http://www.brienposey.com.