MAGNETIC DISKS
Magnetic disks are the most popular direct-access secondary storage device. A magnetic disk is a thin, circular plate/platter of metal or plastic. They are also the most popular online secondary storage device.
Its surfaces on both sides have a coating (such as iron oxide) that can record data by magnetization. Data is recorded on its coated surfaces as tiny, hidden magnetized, and non-magnetized spots (representing 1s and 0s). It uses a standard binary code, usually 8-bit EBCDIC, for recording data.
The Magnetic disk itself is stored in a specially designed protective envelope or cartridge, or several of them are stacked together in a sealed, contamination-free container.
Like in the case of magnetic tapes, we can erase old data and record new data on magnetic disks as well. As we record new data, it automatically erases old data in the same area. However, we can read stored data many times without affecting it.
Basic Principles of Operation
Storage Organization
A magnetic disk’s surface has several invisible and concentric circles called tracks. Tracks have numbers consecutively from outermost to innermost, starting from zero.
The number of tracks varies greatly, with low-capacity disks having as few as 40 tracks and high-capacity disks having several thousand’s tracks.
A magnetic disk’s surface also has invisible, pie-shaped segments. If there are eight such segments, each track has eight parts. Each such part of a track is called a sector.
Typically, a sector contains 512 bytes. It is the smallest unit of data accessed by a disk drive. That is, disk drives can access (read/write) an entire sector at a time. Even if a computer needs to change just 1 byte out of 512 bytes stored on a sector, it rewrites the entire sector.
When people refer to the number of sectors a disk has, the unit they use is sectors per track – not just sectors. Hence, if a disk has 200 tracks and eight sectors per track, it has 1600 (200 × 8) sectors, each having a unique number.
To access a piece of data (a record) from a disk, we need to specify its disk address, representing the record’s physical location. It comprises sector number, track number, and surface number (when double-sided disks are used). Often, multiple disks are stacked together as a disk pack to create large-capacity disk-storage systems.
The disk pack is sealed and mounted on a drive consisting of a motor to rotate the disk pack about its axis. The disk drive also has an access arms assembly having separate read/write heads for each recordable surface of the disk pack.
Usually, a disk pack does not use the topmost disk’s upper surface and the bottommost disk’s lower surface because these surfaces get scratched easily. However, modern disk drives use miniaturization and components to eliminate this drawback.
All-access arms (on which read/write heads are fixed) of an access arm assembly for all disk surfaces to move together. Hence, if the read/write head serving the 0th recording surface is positioned over the 5th track, each of the heads on the arms serving other recording surfaces is also placed over the 5th track of respective surfaces.
For faster access to data, disk packs use a concept called a cylinder for data organization. A set of corresponding tracks on all recording surfaces of a disk pack together form a cylinder. For example, the 5th track of all recording surfaces together forms the 5th cylinder of the disk pack.
Hence, if there are 200 tracks on each disk surface, there are 200 cylinders in the disk pack. The disk address of a data record on a disk pack consists of sector number, cylinder number, and surface number (track number is not required because track number and cylinder number are identical). This addressing scheme is called CHS (Cylinder-Head-Sector) addressing. It is also known as disk geometry.
A cylinder-based organization achieves faster access to data by avoiding the movement of access arms when an application has to process many records in sequence. It stores related records of a file on the same cylinder of a disk pack so that in one revolution of the disk pack, the disk drive can read/write all records stored on, say, cylinder 5 of all surfaces.
Storage Capacity
| Storage capacity of a disk system= | Number of recording surfaces × Number of tracks per surface × Number of sectors per track × Number of bytes per sector |
Let us assume that a disk pack has ten disk plates, 2655 tracks per plate, 125 sectors per track, and 512 bytes per sector. Since the disk pack has ten disk plates, there are 18 recording surfaces (excluding the topmost disk’s upper surface and the bottommost disk’s lower surface). Hence, its capacity = 18 × 2655 × 125 × 512 = 3,05,85,60,000 bytes = 3 × 109 bytes (approximately) = 3 GB (3 Giga Bytes). As one character is stored per byte (using 8-bit EBCDIC encoding), the disk pack can store over 3 billion characters.
For larger storage capacity, designers prefer to increase storage capacity by increasing the number of tracks per inch of surface and bits per inch of the track rather than increasing disk size. Hence, a constant goal of designers of desk systems is to increase the data density of each square inch of the disk surface.
Access Mechanism
A disk drive records data on a spinning disk surface and tracks and reads data from the surface using one or more read/write heads on its access arms assembly.
Most disk drives use a single read/write head for each disk surface. However, some faster disk systems use multiple heads on each access arm to service several adjacent tracks simultaneously.
As the access arms assembly moves in and out, read/write heads move horizontally across the surfaces of the disks.
In this manner, the disk drive positions the read/write heads on any track on/from which it wants to record/read data. In the case of a disk pack, each usable surface has its own read/write head, and all heads move together.
Hence, the disk drive can simultaneously access information stored on the tracks constituting a cylinder through the disk pack. Recall the cylindrical storage arrangement of information in a disk pack.
Read/write heads are of flying type. They do not have direct contact with disk surfaces. There is a separation of about 0.00002 inches between a read/write head and its corresponding disk surface. It prevents the wear of a disk surface.
However, read/write heads fly so close to the disk surface that if a dust particle (typically of 0.0015inch size), smoke particle (typically of 0.00025inch size), fingerprint (typically of 0.00062inch size), or a human hair (typically of 0.003inch size) is placed on the disk surface, it will bridge the gap between reading/writing head and the disk surface, causing the head to crash. A head crash, in which the head touches the disk, destroys the data stored in the impact area and can also destroy a read/write head.
Access Time
Disk access time is the interval between the time a computer requests data transfer from a disk system to primary storage and the time this operation completes. To access data stored on a disk, we need to specify the disk address of the desired data in terms of surface/head number, track/cylinder number, and sector number. It is because a disk always stores and reads information from the beginning of a sector and tracks it. Hence, disk access time depends on the following three parameters:
Seek time
When a disk receives a read/write command, it first positions the read/ write heads on the specified track (cylinder) number by moving the access arms assembly in the proper direction. The time required to position the read/write heads on the specified track/cylinder is called seek time.
Seek time varies depending on the position of the access arms assembly when a read/write command is received. For example, if the access arms assembly is on the outermost track and the specified track is the innermost one, seek time will be maximum.
On the other hand, it will be zero if the access arms assembly is already on the specified track. Average seek time is thus specified for most systems. It is of the order of 10 to 100 milliseconds.
Some disk systems use multiple read/write heads on each access arm to lower the seek time. For example, a disk system having two sets of read/write heads for each surface (one for inner tracks and another for outer tracks) will reduce average seek time by half. It is because each read/write head needs to cover and move across only half the total number of tracks.
Latency
Once the disk drive positions read/write heads on the specified track, it activates the head on the specified surface. Since the disk continuously rotates, this head should wait for the specified sector to come under it. This rotational waiting time, i.e., the time required to spin the specified sector under the head is called latency.
Latency, also known as rotational delay time, is a variable that depends on the distance of the specified sector from the initial position of the head on the specified track. It also depends on the disk’s rotational speed, ranging from 300 rpm (rotations per minute) to 7200 rpm.
Modern ultra-fast disks can reach 10,000 to 15,000 rpm or more. Thus, disk systems usually specify an average latency, which is of the order of 5 to 80 milliseconds.
Note that the average latency of a disk system is equal to half the time taken by the disk to rotate once. Hence, the average latency of a disk system with a rotational speed of 3600 rpm will be 0.5/3600 minutes = 8.3 milliseconds.
Transfer rate
Transfer rate is the rate at which a computer reads/writes data from/to a disk into memory. Once the disk drive positions the read/write head on the specified sector, it reads/writes data at a speed determined by the disk’s rotational speed.
If the rotational speed of a disk is 3600 rpm and the disk has 125 sectors/track with 512 bytes/sector, the amount of data transferred in one full revolution of the disk is 125 × 512 =64,000 bytes (approximately).
Hence, the transfer rate of the disk system is 64,000 × 3600/60 bytes/second = 38,40,000 bytes/second = 3.8 Megabytes/second (approximately). Notice that the transfer rate of a disk system depends on the density of stored data and the disk’s rotational speed.
Since data transfer time is negligible (due to the high transfer rate) compared to seek time and latency, the average access time for a disk system is it’s average seek time plus its average latency. Average access time varies significantly from one type of disk system to another and ranges from 10 to 600 milliseconds.
Since access time for a piece of data stored on a disk depends on its physical location, it is nice to refer to a disk system as direct access storage device instead of a random-access storage device. Random access refers to a storage device in which the access time of any data is independent of its physical location. For example, primary storage is random access storage.
However, we do not observe this distinction strictly and refer to disk systems as random-access storage devices.
Disk Formatting
Magnetic disks come in different sizes. The size of a disk is usually referred to by its diameter. Typical disk sizes include those with 14-inch, 9-inch, 8-inch, 51/4-inch, 31/2-inch, and 31/4-inch diameter.
Different size disks require other disk drives for a proper match of dimensions. Even for disks of the same size, all disk drives are not the same because disk drives of different computers often have their way of defining tracks, sectors, and sector sizes (number of bytes/sector) to match their way of organizing data.
Therefore, it implies that computer manufacturers should also manufacture the disks used in their computer systems. It has a severe limitation because it prevents using disks manufactured by third-party vendors in one’s computer system.
To overcome the problem, earlier computer systems provided low-level disk formatting utilities. It enabled a user to prepare (format) a new disk before using it with the computer system.
For this, the user had to insert the raw (unformatted) disk into the computer system’s disk drive and initiate the disk-formatting command. The read/write head of the disk drive then laid down a magnetic pattern on the disk’s surface, making it compatible for use.
Modern disk drives do not require low-level formatting because modern disk controllers present a consistent view of disk geometry to the computer while hiding the internal hardware arrangement.
As a result, they accept and can work directly with factory-formatted disks by their manufacturers. However, running a generic low-level format on these disks can cause irreparable damage to drive hardware. For this reason, modern computer systems do not provide low-level disk formatting utility.
The Operating System (OS) creates a file system on the disk to store files and data on a disk. It maintains a table with the sector and tracks the data locations on the disk. This table, known as the File Allocation Table (FAT), enables the computer to locate data quickly. The Operating System’s disk formatting command scans and marks terrible sectors to create the FAT and sets aside sufficient disk space.
Disk Drive
We have to mount a magnetic disk on a disk drive for reading/writing data from/to it. A disk drive contains all mechanical, electrical, and electronic components for holding one or more disks and reading/writing data from/to them.
It includes a central shaft on which the disks are mounted, an access arms assembly, a set of read/write heads, and motors to rotate the disks and move the access arms assembly. Although disk drives vary greatly in their shape, size, and disk-formatting pattern, we classify them broadly into two types:
Disk drives with interchangeable disks
These disk drives allow the loading and using of different disks in the same disk at other instances, enabling offline data storage on disks. In addition, they provide virtually unlimited capacity to a disk system because as many disks as required can be used to store massive data sets.
Disk drives with fixed disks
These disk drives come along with a set of permanently fixed disks. The disks, read/write heads, and access mechanisms of the disk drive are housed permanently in a sealed, contamination-free container. Sealed packaging allows the disks to operate in a dust-free environment.
In addition, it enables designers to provide increased data density (using smaller magnetized spots) on each square inch of disk surface by reducing the distance between a read/write head and its corresponding disk surface.
Hence, compared to disk drives with interchangeable disks, these disk drives provide large storage capacity with the same size and number of disk surfaces. However, the disks are not removable from their disk drives because of sealed packaging. Thus, the storage capacity of these disk systems is limited.
Disk controller
A disk controller controls a disk drive connected to it. It interprets the commands for operating the disk drive. Since the disk is a direct access storage device, typically, a disk controller supports only Read and Write commands.
Therefore, a user has to specify the disk address (consisting of surface number, cylinder/track number, and sector number) as a parameter to Read and Write commands. Often, a disk controller controls more than one disk drive connected to it. In this case, the user must specify the disk drive number as a parameter to Read and Write commands.
Types of Magnetic Disks
All magnetic disks are round platters. They come in different sizes, use different types of packaging, and are made of rigid metal or flexible plastic. Based on these differences, there are many types of magnetic disks available today.
However, they are classified broadly into two types – floppy disks and hard disks. Floppy disks are packaged individually in protective envelopes or plastic cases, whereas hard disks are packaged separately or in multiples in cartridges or contamination-free containers.
Depending on the type of packaging, hard disks are classified further into zipping/Bernoulli disks, disk packs, and Winchester disks.
Floppy Disks
A floppy disk is a flat, circular piece of flexible plastic coated with magnetic oxide. It is encased in a square plastic or vinyl jacket cover. The jacket gives handling protection to the disk surface.
Moreover, it has a unique liner that provides a wiping action to remove dust particles, which are harmful to the disk surface and read/write head. Floppy disks are so-called because they are made of flexible plastic plates (not rigid plates) that can bend. They are also known as floppies or diskettes. IBM introduced them in 1972.
Until recently, they were one of the most popular and inexpensive portable secondary storage mediums used in small computer systems. Two popularly used sizes of floppy disks were 31/2-inch and 51/4-inch.
However, portable hard disks, optical disks, and pen drives have replaced them because these devices outperformed floppy disks in storage capacity, access speed, and reliability.
Hard Disks
Hard disks are the primary online secondary storage device for most computer systems today. They are made of rigid metal (frequently aluminum) platters and come in sizes ranging from 1 to 14 inches in diameter. The most commonly used sizes are 1.0-inch, 2.5-inch, and 3.5-inch.
Types of hard disks
Depending on packaging, hard disks usually are of three types:
1. Zip/Bernoulli disk
It consists of a single hard disk platter encased in a plastic cartridge. Depending on the disk drive and size, storage capacity varies from 8 GB to 500 GB. Its disk drive called a zip drive, might be of portable or fixed type. The fixed type is part of a computer system connected to it permanently.
A user can bring and connect the portable type to a computer system for the duration of use and then disconnect and take it away. We can quickly load/unload a zip disk into a zip drive, just as we insert/remove a video cassette player.
2. Disk pack
It consists of multiple (two or more) hard disk platters mounted on a single central shaft. All disks revolve together at the same speed. Its disk drive has a separate read/write head for each usable disk surface (recall that when a disk device uses multiple disks, the upper surface of the topmost disk and the lower surface of the bottommost disk are sometimes not used).
The disk drive is interchangeable and allows users to load/unload different disk packs whenever required. In addition, a user stores a disk pack offline in a plastic case when unused. It gives virtually unlimited storage capacity to disk packs.
Depending on the disk drive, size of the disk, and several disks in a pack, the storage capacity of a single disk pack varies from a few hundred GB to several thousand GB.
3. Winchester disk
A Winchester disk consists of multiple (two or more) hard disk platters mounted on a single central shaft. However, unlike a disk pack drive, a Winchester disk drive is of a fixed type.
Therefore, its hard disk platters and disk drive are sealed together in a contamination-free container and cannot be separated from each other. Hence, Winchester disks have limited capacity.
However, for the same number of disk platters of the same size, Winchester disks have a larger storage capacity than disk packs due to the following reasons:
(i) As the disk platters and disk drive are sealed together permanently, all surfaces of all disk platters (including the upper surface of the topmost platter and the lower surface of the bottommost platter) are used for data recording in a Winchester disk.
For a Winchester disk with four platters, there are eight usable surfaces as opposed to six surfaces in the case of a disk pack with four platters.
(ii) The contamination-free environment allows Winchester disks to employ more excellent data recording and accessing precision, resulting in greater data storage density than interchangeable disk packs.
Winchester disks were named after the 30-30 Winchester rifle because the early Winchester desk systems had two 30-MB disks. The storage capacity of today’s Winchester disks ranges from a few gigabytes (109 bytes) to a few terabytes (1012 bytes).
Advantages of Magnetic Disks
| Support direct access to data | Usage for sequential applications is less efficient than magnetic tapes |
| Shared simultaneously by multiple users | Difficult to maintain the security of information |
| Suitable for both online and offline storage of data | Disk crash or drive failure |
| Unlimited storage capacity | |
| Low cost and high data recording densities | The cost of magnetic tapes is even lower |
| Compact size | Must label Zip disks and disk packs logically and adequately |
| Portable and used to transfer data | Winchester disks are not as easily portable |
| Superior data access rate is to a tape system | |
| Less vulnerable to data corruption | Stored in a dust-free environment |
1. Magnetic disks support direct access to data, making them more suitable for a broader range of applications than magnetic tapes, which only support sequential access to data.
2. Due to the random access property, computers often use magnetic disks as shared device, which is shared simultaneously by multiple users.
For example, computers often use Winchester disks and disk packs as online secondary storage devices storing data of numerous computer system users. A tape is unsuitable for such usage due to its sequential-access property.
3. Magnetic disks are suitable for both online and offline storage of data. For example, Winchester disks and disk packs are often used as online secondary storage devices, whereas; floppy disks and zip disks are used as offline secondary storage devices.
High-capacity Winchester disks have made it possible for today’s most personal computer users to enjoy having all data and software readily accessible.
4. Except for fixed-type Winchester disks, the storage capacity of other magnetic disks is virtually unlimited because many disks are required for storing massive data sets.
5. Due to their low cost and high data recording densities, the cost per bit of storage is low for magnetic disks. An additional cost benefit comes from the fact that we can erase data on a magnetic disk and reuse it many times.
6. Due to their compact size and high data recording densities, they enable the storage of a vast amount of data in a small storage space.
7. Zip disks are easily portable, and users often use them to transfer data and programs between two unlinked computers.
8. A magnetic disk system’s data access rate is usually much superior to a tape system.
9. Magnetic disks are less vulnerable to data corruption due to careless handling or unfavorable temperature and humidity conditions than magnetic tapes.
Disadvantages of Magnetic Disks
1. Although magnetic disks are usable for both random and sequential data processing applications, their usage for sequential applications is less efficient than magnetic tapes.
2. It is more difficult to maintain the security of information stored on magnetic disks used as shared, online secondary storage devices than data stored on magnetic tapes or other magnetic disks.
3. For Winchester disks, a disk crash or drive failure often results in the loss of entire data stored on it. It is not easy to recover the lost data. Hence, data stored on Winchester disks requires suitable backup procedures.
4. Some magnetic disks, such as disk packs and Winchester disks, are not as easily portable as magnetic tapes.
5. On a cost-per-bit basis, the cost of magnetic disks is low, but the price of magnetic tapes is even lower.
6. Magnetic disks are stored in a dust-free environment.
7. We must label Zip disks and disk packs logically and adequately to remember what data is on which disk/disk pack and to prevent the erasure of valuable data by mistake.
Uses of magnetic disks
Usually, we use magnetic disks for the following purposes:
1. For fandom data processing applications.
2. As a shared, online secondary storage device, we often use Winchester desks and disk packs for this purpose.
3. We often use zip disks and disk packs as backup devices for offline data storage.
4. Archiving of occasionally used data. We often use zip disks and disk packs for this purpose.
5. Transferring data and programs between two unlinked computers. We often use zip disks for this purpose.