Magnetic Disks


Magnetic disks provide the bulk of secondary storage for modern computer systems. Physically, disks are relatively simple (see figure below). Each disk platter has a flat circular shape, like a phonograph record. Its two surfaces are covered with a magnetic material, similar to magnetic tape. Information is recorded on the surfaces.


img5.JPG (27103 bytes)

Moving-head disk mechanism.


When the disk is in use, a drive motor spins it at high speed (usually 60 revolutions per second). There is a read-write head positioned just above the surface of the platter. The disk surface is logically divided into tracks, which are subdivided into sectors. We store information by recording it magnetically on the sector under the read-write head. There may be hundreds of concentric tracks on a disk surface, containing thousands of sectors. The platter itself may be between 1.8 inches and 14 inches wide. The larger sizes are common on large systems because of their higher storage capacities and transfer rates. The smaller sizes are found on PCs, since they have lower cost.

A fixed-head disk has a separate head for each track. This arrangement allows the computer to switch from track to track quickly, but it requires a large number of heads, making the device extremely expensive. Much more common, is only one head, which moves across the disk to access different tracks. This moving-head disk, or simply hard disk, requires hardware to move the head! but only a single head is needed, resulting in a much less expensive system. The disk platters, mounted on a spindle and surrounded by heads driven by a motor, are known as head-disk assemblies and come in a complete package.

Disks were originally designed for file storage, so the primary design criteria were cost, size, and speed. To provide additional storage capacity, developers took several approaches. They made the primary gain by improving the recording density, allowing more bits to be put on a surface. The density is reflected by the number of tracks per inch, sectors per track, and bits per sector. In addition, with separate heads on each side of the platter, disk capacity can be doubled at minimal cost. We can extend this approach by stacking several disks, each with two recording surfaces, on one spindle. Since all the disks rotate together, only one drive motor is needed, although each surface still needs its own read-write head. The most common disks, used in systems from portable PCs through mainframes, have this configuration. These disks vary in data transfer rate from 1 to 5 megabytes per second. The average access time, including the time for a head to be positioned over the requested data, is from 10 to 40 milliseconds. The capacities range from 10 to 7500 megabytes.

Finally, the disk can be removable, allowing different disks to be mounted as needed. Removable disk packs may consist of one or several platters on one spindle. Generally, they are held in hard plastic cases to prevent damage while they are not in the disk drive.

Disks are rigid metal or glass platters covered with magnetic recording material. Each platter is divided into small sections, and each such section may be changed by the disk head to be in a charged or not charged state. Each section represents a bit and is 0 or 1 depending on its charge state. The smaller the changeable sections are, the more bits can be put on a platter and thus the higher the density. The read-write heads are kept as close as possible to the disk surface to increase this density. Often, the head floats or flies only microns from the disk surface, supported by a cushion of air. Because the head floats so close to the surface, platters must be machined carefully to be flat.

Head crashes can be a problem. If the head contacts the disk surface (due to a power failure, for example), the head will scrape the recording medium off the disk, destroying the data that had been there. Usually, the head touching the surface causes the removed medium to become airborne and to come between the other heads and their platters, causing more crashes. Under normal circumstances, a head crash results in the entire disk failing and needing to be replaced.

Floppy disks take a different approach. The disks are coated with a hard surface, so the read-write head can sit directly on the disk surface without destroying the data. Thus, the disk itself is much less expensive to produce and use. The floppy disk must rotate much more slowly than a hard disk, due to the resulting friction. Also, the coating (and the read-write head) will wear with use, and need to be replaced eventually. Because they are more rugged, floppy disks are removable. The disks are not permanently mounted in a head-disk assembly. Instead, the disk is slipped manually into a slot which contains a spindle to rotate it, and a head and motor to access it. This arrangement keeps the cost of floppy disks low because one drive can be used to access hundreds or thousands of disks.

Floppy disks usually have a much lower capacity than do hard disks because they have much lower storage densities, have only one platter, and spin slower. They hold from 100 kilobytes (a kilobyte is 1024 bytes) to a few megabytes per disk. They come in many variations (single-sided, double-sided, single-density, and double-density) and sizes (5¼ inch, 3½ inch, and so on). Generally, they are formatted and used in the same way as are hard disks, except that all floppy disks are removable and may therefore be used conveniently to transfer data between computers.

A disk drive has a disk controller that determines the logical interaction between the device and the computer. The controller takes instructions from the CPU and orders the disk drive to carry out the instruction. Some disk controllers have a built-in cache, which holds data recently read from or written to the disk. If data are currently in the cache, the need for a disk transfer is obviated.


Last Updated Jul.28/99