Anatomy of a Hard Disk

Almost all computers today store their digital data as magnetic areas on a device called a hard disk, hard drive or fixed disk.

Basically, all hard drives work the same way: Information is encoded and "written" onto a round, spinning aluminum or glass platter that's been coated with magnetic material. The writing is done by a magnetic head, mounted at the end of an arm that pivots in such a way that the head can be positioned over any part of the platter. The same head also reads the stored data. Special software or firmware on the disk drive and the computer keep track of where any piece of information is stored. Older disk drives devoted one entire side of one platter, along with its head, as a servomechanism to calibrate and regulate platter and arm motion, but current technology doesn't require nearly so much space.

Remember when music came on vinyl records? A disk drive operates a lot like the phonograph. Each has a motor that spins a platter containing information that is written or retrieved by a special device mounted on the end of an arm that pivots across the disk.

There are considerable differences, of course. The LP record was plastic and 12 in. in diameter, and it spun at 33-1/3 rpm. The computer hard drive, once 14 in. or more across, is now no bigger than 3.5 or 5.5 in. in diameter, with those in laptops and handheld devices at 2.5, 1.8 or even 1 in. Hard disks spin at speeds ranging from about 4,000 to 15,000 rpm, and those speeds are likely to increase in the future. And where the phonograph needle physically touched the record groove, the drive heads don't touch the spinning media at all, though they get very close while flying on a cushion of air.

Today's disks can store immense amounts of data: About the smallest 3.5-in. hard drive being made today will store 10GB, and capacities for individual drives have reached 100GB. Drive makers have two ways of increasing the capacity of a disk drive. The simplest method is to add additional platters along with a separate head for each side of each platter, and this has been done up to about 16 platters. The second, more basic, way is to increase the amount of data that can be stored on a single area of the magnetic material. This has been the subject of considerable research. Today, IBM has drives that store 25.7GB per square inch, and the company has demonstrated technologies that can quadruple that, to 100GB of data in a single square inch.

The very first disk drive was IBM's RAMAC. Introduced in 1956 the RAMAC's 50 24-in. platters held 5MB of data; the cost was $50,000. In 1980, a 14-in. minicomputer disk cartridge could hold perhaps 5MB or 10MB of data. The original IBM PC in 1981 didn't support a hard disk. When DOS Version 2 came out, the first disk drives appeared for PC-class machines, using 5.25-in. platters that could store 5MB or 10MB and eventually more than 40MB of data.

By 1990, it was common for PCs to come with 40MB disk drives. Five years later, the typical new desktop computer had a 1GB or 2GB hard drive. Nowadays, you can buy laptop computers with 30GB drives, and 48GB 2.5-in. drives have now hit the market.

And as for price, in 1992 I bought an 80MB, 5.25-in. drive at a computer flea market for $300; today's market will deliver a 20GB 3.5-in. hard drive for a little more than $100 retail; that's 250 times the capacity at one-third the price. Put another way, the 1956 disk drive was priced at $10,000 per megabyte. In 1992, I paid just $3.75 for each megabyte of storage; today, my price for that same megabyte is a half-cent.

The combination of low price and high capacity came together in 1990, when IBM assembled a group of these inexpensive drives into the first RAID systems that offered security and error recovery to the mix.

Even in today's world of storage-area networks and network-attached storage, the basic building block is the individual magnetic disk drive, and that's perfectly exemplified in the currently popular acronym JBOD - just a bunch of disks.

Inside a Disk Drive
In this exploded view, you can see the major components that go into a typical drive:

A Platter: Stores the data
B DC spindle motor: Spins the platter
C Head: Reads or writes data from or to the platter
D Actuator: Causes the arm to move
E Printed-circuit cable: Connects arm and head to electronics
F Arm: Moves across the disk, positioning the head
G Chassis: Cast metal base on which other components are mounted
H Protective cover: Seals the mechanism against dust
J Logic circuits: Handle address translation, data buffering and I/O requests
Inside a Disk Drive
Disk or Disc?
For magnetic media used on computers, disk is the preferred spelling. For other round, flat objects, including optical storage media such as CDs—audio or data—and DVDs, the proper spelling is disc. And where, you ask, did the “hard” come from? It’s used in distinction to floppy disks, whose magnetic medium was quite flexible but could only be spun at 360 rpm. The rigid platters in early PC hard drives could spin 10 times as fast—at 3,600 rpm—while today speeds of twice that are common and several makes have drives that turn at 15,000 rpm.

Join the newsletter!

Error: Please check your email address.

More about IBM Australia

Show Comments

Market Place