1. Introduction

All information systems need to store data. This may be done temporarily whilst inputs are processed to produce outputs or for much longer periods of time.

For example :

• A school uses a database package to keep a record of the achievements of its students. These records are kept for the seven year period that each student attends the school.
• A student is using a graphics package to draw a logo on the screen. The graphics package must store the colour of each pixel in the logo so that it can be printed.

A computerised information system must be able to store both programs (e.g. a word processing package) and data (e.g. a word processed letter).

A storage device stores programs and data either temporarily or permanently. All information systems contain two different types of storage :

Immediate Access Store (IAS)	Immediate access store holds programs and data that the user is currently working with.
Backing Store	Backing store keeps data and programs when the computer is turned off.

Immediate access store is also known as main store or primary store. Backing store is also known as secondary store.

The capacity (amount of data) that a storage device can hold varies significantly between different devices. Units such as bytes, kilobytes and megabytes are used to describe a storage device's capacity. Other factors such as speed of data access, cost and portability will also determine which storage device is the most appropriate one to use for a particular application.

Select a storage device from the table below to find out about it or read through the rest of this topic to learn more about all of the different storage devices.

Immediate Access Store
RAM
ROM
Backing Store
Magnetic Disks		Floppy Disk
		Hard Disk
		Zip Disk
		Jaz Disk
Optical Disks		CD-ROM
		CD-R
		CD-RW
		DVD-ROM
		DVD-RAM
Magnetic Tape
Flash Memory Devices

2. Device Capacity

Various units of measurement exist to describe the capacity of storage devices. The smallest unit of storage is the bit. A computer can only store data as binary numbers (0s and 1s). A bit of storage can hold either a 0 or a 1. Devices such as computers which can only use binary information are described as digital devices.

Eight bits are grouped together to form a byte. Using 8 bits it is possible to represent any character on the keyboard with a byte. The characters are represented in 0s and 1s using a code called the ASCII code. Here are some example codes :

To store the sentence "Hello World" in a computer would require 11 bytes (one for each character), which is 11*8 = 88 bits.

Describing the capacity of a storage device in bytes would be silly. A typical home microcomputer can store 536870912 bytes in its immediate access store and 128849018880 bytes in its main backing store (probably a hard disk). Therefore there are a number of other related units that are used :

Typical Storage Capacity of a Microcomputer

A typical new home microcomputer has 512Mb of immediate access store. This is equivalent to :

• 512*1024 = 524288 Kb
• 512*1024*1024 = 536870912 bytes
• 512/1024 = 0.5Gb

The same computer is likely to have three backing storage devices, a floppy disk drive (capacity 1.44Mb), a CD/DVD drive (capacity 650Mb-18Gb) and a hard disk drive (capacity 120Gb+).

3. Immediate Access Store

Immediate Access Store ( IAS ) holds programs and data that the user is currently working with. For example :

• A word processed document that is being edited will be loaded into IAS.
• An email program that is currently transmitting a message will be loaded into IAS.

There are two different types of IAS :

• Read Only Memory (ROM) : The contents of ROM is permanent. It can not be altered by the user. The content is written onto the ROM when it is first made. ROM keeps its contents even when the computer is turned off and so is known as non-volatile memory. On some computers a special piece of software called the BIOS which is used to configure the computer is stored in ROM. ROM is also often used in embedded systems where a small built-in computer is used to control a device such as a washing machine. The program that controls the machine is stored on ROM.

• Random Access Memory ( RAM ) : RAM is used to store programs and data that are being used by the computer. When the computer is turned on the RAM is empty. Data and programs can be put into RAM from either an input device or backing store. The data in RAM is lost when the computer is turned off so it is known as volatile memory. To keep data the user must save it to backing store before the computer is turned off.

  • The process of transfering data/programs from backing store into RAM so they can be used is known as loading.
  • The process of transfering data/programs from RAM to backing store so that they will be retained when a computer is turned off is known as saving.

  It is easy to add extra RAM to a microcomputer by inserting extra RAM cards called SIMM or DIMM cards :

  

IAS is located inside the computer. Data can be written to and read from IAS electronically at very high speeds, much faster that it can be written to or from backing store. IAS is much more expensive to buy per Mb than backing store is.

4. Backing Store

If data needs to be kept whilst a computer is turned off then it must be stored on backing store. Any programs or data that are not currently being used by a computer will be kept on backing store. When programs or data are used they are copied (loaded) into immediate access store for faster access.

Information stored on backing store is placed on a storage medium. The most common media which are used for backing store are :

• Magnetic Disks

• Magnetic Tapes

• Optical Disks such as CD-ROMs and DVDs

• Flash Memory Devices

The data is read from or written to the storage medium by a piece of hardware known as a drive or a storage device.

It takes much longer to access data which is on backing store than data which is in immediate access store, typically 100 to 1000 times as long. This is because most backing storage devices operate mechanically. Computers have much more backing store than immediate access store for two reasons :

1. IAS only needs to store programs and data that are currently being used whereas the backing store needs to hold all of the programs and data that can be used on the computer.

2. Backing store is much cheaper per Mb than IAS.

Programs and data can not be used directly from backing store. They must be copied (loaded) into immediate access store before they can be used. Any data which needs to be kept must be transferred back to the backing store from the immediate access store before the computer is switched off. This is called saving.

Sometimes data is compressed before it is stored. Compressing data reduces the storage space that a file uses without losing any of its contents.

5. Magnetic Disks

Magnetic disks are the most common backing storage device. Data is stored using magnetised spots called domains on the disk. Each domain can store one bit of data (a 0 or a 1). A 3.5" 1.44Mb floppy disk used on a PC contains 16,777,216 such spots. A 3.5" 120Gb hard disk drive contains over 120 billion spots!

Data stored on disks is arranged along a series of concentric rings called tracks. Each track is divided up into a number of sectors. Data is read to and written from a disk one sector at a time. A sector usually contains 512 or 1024 bytes of data.

The process of dividing a disk up into tracks and sectors so it can be used on a computer is known as formatting. You must format a new disk before you can use it.

Data is read from the disk using a disk head which moves mechanically about the disk (rather like a record player tone arm). The disk head can move directly to any sector on the disk. Because of this a computer system can load a file or a record from a file very quickly. The system can move directly to the location of the record/file and read it without having to read any other data from the disk. This is known as direct access. For most applications using a direct access medium is much faster than using a serial access medium.

The two main types of magnetic disks are floppy disks and hard disks.

6. Floppy Disks

Floppy disks are magnetic disks. They are portable (can be moved between computers) but have a small storage capacity. Reading and writing data from a floppy disk is very slow. The most common type of floppy disk is the 3.5" disk that can store 1.44Mb of data when it is used on a PC. Older disks were 5.25" or 8" in size but could store much less data.

A floppy disk is manufactured from a flexible plastic disk. This disk is coated with a magnetisable material. For protection the disk is encased in a plastic shell. All sizes of floppy disk have a write protect tab built into the shell. If this tab is set then data can be read from the disk but not written to it. The write protect tab can be used as a security measure to prevent important data being deleted or changed accidentally.

Floppy disks are used for three main purposes :

Some hardware companies now produce storage devices which are very similar to floppy disks but can store 100Mb or even 750Mb of data. These devices are also much faster than standard floppy disk drives. An example is the Zip disk.

7. Hard Disks

Hard disks are magentic disks. They have much larger storage capacities than floppy disks. Data can be transferred to and from a hard disk much more quickly than from a floppy disk. Hard disks are usually fixed inside a computer and can not be moved between different machines. Some expensive hard disks can be moved between computers. These are called exchangeable hard drives. An example of this is the Jaz drive.

A hard disk is made of a rigid disk which is coated with a magnetisable material. The magnetic material used is of a much higher quality than that found on floppy disks. Hard disks spin much more quickly than floppy disks and the disk head is positioned very close to the disk (thousandths of a millimetre away). Because the disk head is positioned so close to the disk hard drives can easily be damaged by dust or vibration. Therefore the disk, the drive head and all the electronics needed to operate the drive are built together into a sealed unit. This picture shows a hard disk drive with the case removed.

Usually (as in the picture above) several physical disks are contained in one hard disk unit. Each disk is known as a platter. Typical hard disk capacities for a home PC now start at 80Gb and units storing up to 300Gb are available. If larger storage capacities are required then multiple hard drives can be combined. Using appropriate hardware or software many hard disks can be made to look like one very large hard disc to the user.

8. Zip and Jaz

Zip and Jaz disks are examples of removable magnetic disks. They can be used to transfer data from one computer to another, to backup data, or as an alternative to a hard disk for storing files.

9. Optical Disks

Optical disks store data by changing the reflective properties of a plastic disk. Binary computer data (0s and 1s) are represented by the way the disk reflects light when a low power laser is shone at it. A 0 stored on a disk reflects light differently to a 1 stored on a disk.

Like floppy disks, optical disks can be moved from one computer to another. They have much larger storage capacities than floppy disks but can not store as much data as a hard disk. Data can be read from an optical disk more quickly than from a floppy disk but hard disks are much quicker. As with a hard disk the drive head in an optical drive can move directly to any file on the disk so optical disks are direct access.

There are five types of optical disks that are currently in use. They are all the same physical size. It is possible to purchase a Combo drive that can read and write most of these different types of disks.

The five types of optical disk are :

CD-ROM (Compact Disk - Read Only Memory)

CD-R (Compact Disc - Recordable)

CD-RW (Compact Disc - Rewriteable)

DVD-ROM (Digital Versatile Disk - Read Only Memory)

Because of their high capacity, DVD-ROM disks are used to store high quality video such as complete movies. Often extra data such as information about the making of the film or the actors and actresses who star in it are also stored on the disk. Unlike movies recorded on video tape, DVD-ROM movies can be interactive. The user can make selections on the screen and change what they see.

DVD-RAM (Digital Versatile Disk - Random Access Memory)

DVD-RAM disks have all of the benefits of DVD-ROM disks and can be written to as well. These very high capacity disks are ideal for producing backups. They are starting to replace video tapes for recording television programmes. DVD-RAM disks typically store about 5Gb of data.

10. Magnetic Tapes

Magnetic tape comes in two different forms :

The storage capacities of tape cartridges are now so great that reels of tape are not used in new computer systems.

A computer tape is divided up into many blocks. A block on a tape is like a sector on a disk. To read data from a particular block on the tape all of the blocks that come before it must be read first. This means that finding a particular file/record on a tape can take much longer than finding a particular file/record on a disk. However once the file or record has been located it can be read more quickly from a tape than in could be from a hard disk. Having to read through all of the files/records that are stored before the file/record you want to access is known as serial access.

Often files or records are stored on a tape in a particular order (e.g. sorted alphabetically by a key field). If this is the case then the tape is described as having sequential access. Because locating data on a tape takes a long time, magnetic tapes are not used as general purpose storage devices. They are only useful for a few applications. The two main applications tape is used for are :

Applications for Magnetic Tapes
Backup	Often a tape streamer is used to make copies of data stored on a hard disk in case the data becomes corrupted. If this happens then the correct data can be restored from the tape. A backup copy of the contents of the hard disk could be made once every week. Tapes are more suitable for making backups than floppy disks or CD-ROMs. This is because the entire contents of a hard disk can be written onto one tape, producing the backup will be much quicker and tapes are cheaper to buy. It does not matter that tapes are serial access because when a backup is made or restored every file on the tape has to be accessed.
Batch Processing	In a batch processing system when processing occurs the whole master file and the whole transaction file are read to produce a new master file. Because the master file and transaction file are both sorted into the same order the data can be read sequentially from each file. A batch processing system does not have to move backwards and forwards through a tape looking for records. The system is designed so that the next record to be read is stored immediately after the current record on the tape. Therefore using a serial access medium instead of a direct access medium will not slow down processing. Tapes are preferred to disks in batch processing systems due to their relative low costs and fast read/write speeds.

11. Flash Memory

Flash Memory devices store data electronically in a type of memory that can have data written to it and (unlike RAM) can retain the data when the power to the memory is turned off. Commonly flash memory devices store data in Flash Erasable Programmable Read Only Memory (FEPROM).

Several different types of flash memory device are available. These include Compact Flash, SmartMedia, Secure Digital and USB Pen Drives.

Flash memory devices are very compact, most being a similar size to a postage stamp or coin. For this reason they are often used as backing storage in small devices such as digital cameras and PDAs. Typical storage capacities are from 32Mb up to about 2Gb. The cost per Mb of flash memory storage is high compared to other backing storage devices.

12. Disk Cache

Accessing data stored permanently in backing store is much slower than accessing data in RAM. A common technique that is used to speed up access to data on backing storage devices is caching.

A cache is a part of a computer's RAM that is set aside to store copies of frequently used sectors on a backing storage device such as a hard disk drive. As files are accessed from the hard disk the computer will determine which sectors of the disk are being used most frequently. Copies of the data that is located in these sectors are transferred into the cache.

After this whenever the data in these sectors is read or changed the computer will use the copies of the sectors in the cache rather than the original sectors on the disk. Since the cache is located in RAM these accesses will be many times quicker, so the operation of the computer system will be speeded up.

The contents of the cache are lost when the computer is turned off. This is because cache is located in RAM which is volatile. To ensure that any changes made to sectors in the cache are saved permanently to the hard disk drive the computer system will copy these sectors back to the hard disk drive at a time when the drive is not being used or when the computer is properly shut down.

13. Access Methods

There are three different methods which can be used to access (locate) data stored on a backing storage device. They are direct access, serial access and sequential access.

Direct Access

When direct access is used the head that reads data from the storage medium can move directly to any point on the storage medium. If a particular record or file must be loaded or saved then the head can move directly to the record/file's position on the storage medium and read the data. Records and files can therefore be located very quickly.

Magnetic disks such as hard disks and floppy disks and optical disks such as CD-ROMs use the direct access method. Direct Access is required if transaction processing is taking place.

Serial Access

When a serial access medium is being used, the head that reads data from the storage medium has limited freedom of movement. The only serial access medium is magnetic tape.

To read a particular record/file from a serial access medium, all of the data that comes before the record/file must be scanned through. The tape head can not move to a particular record/file on the tape without reading through all of the other records that come before it.

Because of this it can take a very long time to locate a record/file on a tape and so tapes are only used for specific applications such as backup and batch processing. For these applications the speed of locating data is not important and the other advantages of magnetic tapes outweigh the slow access speed.

Sequential Access

Sequential access is a slight modification of the serial access method. This method is identical to serial access except that the records are stored on the storage medium in a particular order, e.g. by customer number. Sorting the data into an order may speed up operations such as searching the tape. The sequential access method is usually used by batch processing systems.

14. Compression

When data is stored it occupies space on a storage device such as a hard disk or floppy disk. Buying storage space costs money. To reduce the cost of storing data users often attempt to reduce the size of the files they want to store without losing any of the data.

Data compression is the act of reducing the size of a file without losing any of the data it contains. Decompression (also known as uncompression or expansion) is the act of taking a compressed file and converting it back to its original size and format. Because compression alters the format of the data in a file a compressed file must be decompressed before it can be used.

How much a file can be compressed by will depend upon the method being used to compress the file and the data that is stored in the file.

The three most common applications for which data compression is used are :

• Distributing Software
• Hard Disk Compression
• Data Transmission

Distributing Software

Software is often distributed on CD-ROMs. The software must be installed onto a hard disk before it can be used. Software manufacturers usually compress the files that make up a package on the CD-ROM. When the software is installed onto a hard disk the files are decompressed by the installation software.

Hard Disk Compression

Files stored on hard disks can also be compressed. Some utilites will compress and decompress individual files whilst others can compress an entire hard disk automatically as files are saved and loaded.

Compressed files may load more quickly or more slowly than files that are not compressed. This will depend upon how well compressed a file is and the speed of different components of your computer hardware.

Data Transmission

Compression is also used when data is transmitted over a telephone line. Modems will only allow the transmission of data at very slow speeds (e.g. 3Kb per second). To transmit a picture the size of an A4 page can take over ten minutes. If data that is to be transmitted can be compressed before it is sent then there can be considerable advantages :

There are a variety of agreed standards for compressing data that is sent over telephone lines. Some modem manufacturers claim compression ratios of 1:4 (i.e. data can be reduced to a quarter of its original size before being transmitted). Such high compression rates are rarely achieved.

15. File Processing

Two common methods of processing saved files are sorting and merging. These are often used by batch processing systems which store data as records in files. For example a file might contain records about people who are employed by a company. There would be one record per employee.

Sorting

Sorting is arranging the records in a file into a particular order, such as by employee number.

Merging

Merging is combining the contents of two files into one file. The simplest method of merging two files together is to just take the contents of one file and add them to the end of the other. This is known as appending :

If the records in a file must be kept in a particular order (a sequential file) then the merge process is more complex. To carry out a merge the algorithm shown in this flowchart must be followed :

The results of the merge are also shown in the above diagram. Note that unlike a simple append, this more sophisticated merge has produced a new file that maintains the required order. The records in two files that are to be merged must have the same record structure.

GCSE ICT Companion 04 - (C) P Meakin 2004