A range of independent tests have shown that for typical single user desktop use, a RAID (0, 3 or 5) does not improve real world performance and may actually be slower (
www.storagereview.com).
3. RAID helps multi-user applications far more than it does single-user scenarios. The enthusiasm of the power user community combined with the marketing apparatus of firms catering to such crowds has led to an extraordinarily erroneous belief that striping data across two or more drives yields significant performance benefits for the majority of non-server uses. This could not be farther from the truth! Non-server use, even in heavy multitasking situations, generates lower-depth, highly-localized access patterns where read-ahead and write-back strategies dominate. Theory has told those willing to listen that striping does not yield significant performance benefits. Some time ago, a controlled, empirical test backed what theory suggested. Doubts still lingered- irrationally, many believed that results would somehow be different if the array was based off of an SATA or SCSI interface. As shown above, the results are the same. Save your time, money and data- leave RAID for the servers!
Real-world performance is not about maximum throughput numbers or XBench scores!
Striped RAIDs come into their own when there is intense multiuser traffic (as in a file or database server) or in specialized cases where the streaming of continuous high volume data (such as HD video) can saturate an individual drive system. Even then, the advantage only really plays when there are multiple drive busses to support the RAID -- so for example a software RAID with two drives on a single Firewire controller (and despite there being 2 physical Firewire ports, they are still handles by one controller) is of marginal use, because the controller becomes the chokepoint.
As has been mentioned RAID 0 has (for me) unacceptable risks. RAID 1 offers no performance benefit (and a slight write speed / CPU penalty) but does offer redundancy in case of a single drive failure. Note that this does not equal a backup, because if data is lost through deletion, overwriting, file corruption from the software, malware, etc, the damage is done to the mirror at the same time as the master. RAID 3 or 5 or 10/1+0/0+1 is expensive, and, as above, of dubious value for single user applications.
My opinion is that for maximum performance, a multi-drive system should be deployed as individual disks, ideally placing each of the System, Scratch, Application and Data functions on a separate SATA drive, then have periodic online backup to massive external Firewire or eSATA drives that are portable/dockable.
