Hardware: Input, Processing, and Output Devices - Principles of Information Systems - page 100

more costly and less portable, are more popular because of their greater storage capacity and
quicker access time. The Iomega REV Loader 560, shown in Figure 3.7, is a storage device
that holds up to eight removable 70 GB discs, each housed in a shock-resistant plastic case
about the size of a 3.5-inch floppy disk. The device works well for anyone rotating off-site
backups of critical data. Tape has been giving way to the external hard drive as the preferred
backup medium for small businesses. However, if you need off-site copies of your data,
hauling home a heavy, fragile hard drive seems less than ideal. You could use an online backup
service, but for more than a few gigabytes, speed and expense become problems. Burning to
DVDs might offer a solution, except a single disc holds only 4.7 GB (or 9 GB for the few
people who have dual-layer drives), making capacity an issue. 26 Hitachi has announced a 500
GB 2.5-inch hard drive for portable computers. 27

Figure 3.7

Iomega REV Loader 560
(Source: Courtesy of Iomega
Corporation.)

RAID
Putting an organization's data online involves a serious business risk—the loss of critical data
can put a corporation out of business. The concern is that the most critical mechanical
components inside a disk storage device—the disk drives, the fans, and other input/output
devices—can break (like most things that move).
Organizations now require that their data-storage devices be fault tolerant—they
can continue with little or no loss of performance if one or more key components fails. A
redundant array of independent/inexpensive disks (RAID) is a method of storing data that
generates extra bits of data from existing data, allowing the system to create a “reconstruction
map” so that if a hard drive fails, it can rebuild lost data. With this approach, data is split
and stored on different physical disk drives using a technique called striping to evenly dis-
tribute the data. RAID technology has been applied to storage systems to improve system
performance and reliability.
RAID can be implemented in several ways. In the simplest form, RAID subsystems du-
plicate data on drives. This process, called disk mirroring , provides an exact copy that protects
users fully in the event of data loss. However, to keep complete duplicates of current backups,
organizations need to double the amount of their storage capacity. Thus, disk mirroring is
expensive. Other RAID methods are less expensive because they only partly duplicate the
data, allowing storage managers to minimize the amount of extra disk space (or overhead)
they must purchase to protect data. Optional second drives for personal computer users who
need to mirror critical data are available for less than $100.
Medkinetics is a small (12 employees) business that automates collecting and submitting
for approval of information about a doctor's qualifications. Jim Cox, founder and president
of Medkinetics, says: “The high availability of data is also really important to us.” The firm
employs terabytes of inexpensive but secure RAID storage. 28

redundant array of
independent/inexpensive
disks (RAID)
A method of storing data that
generates extra bits of data from
existing data, allowing the system to
create a “reconstruction map” so
that if a hard drive fails, the system
can rebuild lost data.

disk mirroring
A process of storing data that
provides an exact copy that protects
users fully in the event of data loss.