6 computer systems

6 Computer systems

To err is human, but to really foul up requires a computer. Dan Rather, US TV newscaster

Overview In unit 1 we looked at what made a system and, in particular what constituted an information system. In this unit we are going to get more specific and consider what is involved with a computer system in action.

To keep our investigation of computer systems relevant we must always be aware that they are operated by people. Throughout our study we will look at both the social and ethical use of the systems and the human-computer interaction that takes place.

To find out about computer systems we will investigate:

the parts of a computer system hardware software networking cabling and protocols case study of a computer system the changing workplace.

Introduction Designing, implementing, and organising a computer system can vary from the very simple, to the very complex; from a student buying a single laptop, to a large organisation setting up a network involving a wide range of technology.

We will investigate what is involved in this process, but before we do we will focus our thinking by looking at a scenario of a typical computer system in development.

Dr. Jill Foote has been in general practice in her doctor’s surgery for nearly ten years now and is growing increasingly annoyed at the inefficiency of the current system of running her office.

Dr Foote has about 900 patients, but many visit only rarely. On average she will see 25-30 patients a day and has an income of around $38 000 per month. While this may seem a lot, out of this she must pay for rental on the surgery, equipment, supplies, insurance and the wages of Kym the receptionist/nurse, and Darryl a cleaner who comes in each evening. Although Kym controls most of

© Kevin Savage 2011 – Licensed to Hillcrest Christian College for use in 2011-12

Computer systems 171

the appointments and clerical work much of the financial management has fallen on Dr Foote and she is finding she has little free time to be with her family.

She is also concerned about the growing complexity of the finances. Most of Dr Foote's income is paid through Medicare or by private health insurers; only a small proportion is paid in cash by patients. Claiming this money involves detailed record keeping and much form filling. The occurrence of even small errors can delay payment by months and errors affecting patients harm the practice.

The recording of GST is also a problem. While most medical services are free of the tax there are some aspects of her business that involve the GST. Keeping track of input tax credits and which services and items are taxed is very demanding, as is preparing the quarterly BAS (business activity statement).

Wages and the payment of accounts owing also falls on Dr. Foote. Patient medical histories, reminder notices for follow up visits, and a stock record of medicines, etc., on hand are other aspects of the business that Dr Foote is interested in computerising.

At home Dr Foote has an Internet connection and is in regular email contact with colleagues. She would like to extend this connectivity to the office. It would also be useful to network the computers in the office. The system to be set up will need to be reliable as she cannot afford the inconvenience of a breakdown; secure so that patient confidentiality is maintained; and simple to use as she has little computer background.

Activity 6.1 – The doctor’s office 1. a Make a list of as many different activities as you can that take place in the doctor’s

office and that are described above (e.g. maintain patient records).

b Tick those activities in your list that you think could best be managed by a computerised information system.

c Are there any activities that could be computerised, but which perhaps might best not be? Give reasons for these.

2. What advantages will Dr Foote gain by introducing a computerised information system to her business? In your answer consider time, money, space, security, efficiency, effectiveness, and patient response.

3. a Identify the individuals who will be the main users of the new system.

b What tasks will each of these users carry out?

c What are some of the human factors Dr Foote will need to consider in establishing this system? (Consider training, security, management, and other similar factors.)

4. Part of the difficulty Dr Foote will have with installing her new system will be the problems involved in changing over from the old to the new, i.e. paper based to computer based.

a Describe some of the problems that might arise.

b Do you think it would be better to make a sudden change (over a weekend say), or phase in the new system over a period of time?

Give reasons to support your answer.


Leading Technology 172

System parts A complete computer system consists of three areas, the human, the processing environment, and interconnectivity. Each of these is important in its own right, each interacts with the other two parts, and each must be considered when putting together a computer system. There should not be a focus on one of these areas at the expense of the other two; each area must be considered independently to properly ensure the usability, reliability and effectiveness of the whole system.

We will start by looking at the human side of a computer system.

Human considerations The users of a computer system must be directly addressed during its development. While it is tempting to focus on hardware and software, the role of the user is crucial – without the user there is no need for the system. People collect the data for the system, they interact with it, and the system’s output is used for their purposes.

The main areas of human interaction involved in a computer system to be considered are:

the work procedures of the day-to-day operation of the computer; these include measures for data collection, methods of entering the data into the computer, and procedures for handling system output; they also include the maintaining of effective interaction between colleagues for shared tasks

technical management which involves installation, customisation and maintenance of hardware, software and networking features; it also often includes user maintenance and user support

training in the use of the computer system; this will include both initial training and on-going re-skilling as changes occur in the technology; training can be either on the job, in seminars off site, or it may be required as a basic qualification prior to obtaining the job

personnel management to ensure the health, safety and wellbeing of workers, the reduction of stress factors, and the use of ergonomic principles in the design of the working environment

security to protect the data in the system from accidental or deliberate errors; this will include the use of company policies, firewalls, virus protection and a regular backup protocol.

In developing any new computer system each of these factors must be considered.

Going back to the doctor’s office scenario we can be more specific as to what each of these involves. In Dr Foote’s office the human systems will involve each of the features identified above and decisions will have to be made as part of the change over:

work procedures – who will do which jobs?, e.g. enter confidential patient data, handle financial matters, manage day-to-day updating of patient records and bookings, and so on; what output is to be obtained, what recording process is to be followed, how is the file system to be set up; how will each of the users work together for best efficiency and without interfering with the others?

technical management – who will decide what hardware, software and networking system is to be purchased, at an affordable price, and ensuring it is suitable for this particular situation; who



is responsible for the system set up, for repairs or contacting repairers, for routine maintenance and purchasing of supplies?

training – who will have to be given instruction in how to use the system and software, who will provide the training, and how will it be delivered so as not to interfere with normal operations? Dr Foote herself will need skilling in the use of the new system and how to use any of the programs she is not familiar with such as a new accounting package; Kym the receptionist/nurse will also need training; time must be found in a busy schedule for both of these to occur

personnel management – what safeguards and procedures will have to be put in place to ensure the change over and operation of the new system occurs smoothly and with as little stress as possible, and that the new system conforms to health and safety standards?

security – what precautions and practices are to be put in place to ensure the confidentially and safeguarding of patient records and other sensitive or critical data in the system?

The overall effectiveness of a computer system relies on how well the users are accommodated during the design and implementation stages of system development. By addressing each of these areas the new system has the best chance of success. The goal is to allow users to do their job as efficiently as possible by removing any obstacles that might be introduced by changes to the computer and network systems or by human factors.

Processing environment Once the human aspects have been determined we can look at the processing involved.

The computer system itself consists of three main sub-systems, the hardware setup, the software layer and the interface system. Together these three provide a complete system for processing. They must be planned for during the design phase, and together take up the majority of the implementation phase.

Hardware is the physical components that make up a computer system. It includes such items as monitors, processors, disc drives, printers, modems, routers, cabling, wireless access points,

and so on. The hardware is physical in that it is real and solid, and can be touched, moved and connected. These components can be selected from off-the-shelf parts, manufactured to fit a required task, or a mix of these options.

Many organisations source all of their hardware from a single supplier such as Dell. Systems that exclusively use a prefabricated option such as this have the advantage of reduced cost, faster implementation, and demonstrated reliability. This

approach however may not address the hardware requirements of some organisations. In this case a more hybrid combination of devices may better fit the particular needs of the enterprise.

The software is the programs that run on the computer. These are not physical and exist only as electrical or magnetic pulses stored on disc or in memory. Software consists of applications to run on the computer, operating systems to run the computer itself, and programming languages used to write specific applications.

Similar to hardware these programs can be purchased as completed packages or can be developed by teams of programmers. It is even common for programs to be extended to add



extra features, or to be directed how to interact with data that the users input. Choosing the best software for the task is the most important part of planning the processing environment.

The final system is the interface. The interface is the means by which a user communicates with the computer system – the link between person and machine. In the case of package software such as Microsoft Office, the interface is often limited to a default style, however this is not always the case and custom interfaces can be employed. Software specially written for an enterprise will need a clean and simple interface developed. (User needs and interface are explored in more detail in the unit on HCI.)

Interconnectivity The final element important to the development of a computer system is the networking in place. For the purposes of a practical system these networks can be grouped into two types, the abstract and the physical.

The abstract networks are those which connect information but do not directly deal with moving information between computers. Some examples of an abstract network includes humans working together to collect and input data into the system, the hardware of a computer talking between its components to process data, and peripheral devices such as printers and routers assisting in both of these tasks.

Physical networks are used to transfer data that exists on one device on the system to another. This data can be anything input into the system, information derived from the original input, or information obtained from outside sources. These physical networks allow the rapid movement of data locally and around the globe.

In all computer systems abstract networks must exist even if physical networks are not required. Identifying these networks allows the organisation of the system to be planned during the design phase and tested during implementation.

System solution In the next sections we will look at hardware, software and networking in more detail, but for now we will propose one possible solution for Dr Jill Foote’s situation in each of these areas:

Hardware: There is probably the need for at least two computers, one for the receptionist and one on the doctor’s desk. Of these one might be a laptop so that it can be taken home in the evening, while the other can be a mid-level desktop computer. The computers will use the normal peripherals such as keyboard, mouse and printer, but direct connection to a fax line would help.

Software: The office will employ a range of software to run the computers as well as take care of all of the tasks Dr Foote would like to computerise. A simple practical solution might encompass a Windows 7 based operating system, the Microsoft Office suite of programs, a networking client, and an accounting package such as MYOB or Quicken. A dedicated communications package such as Outlook would also help.

Networking: These computers will need to be linked so that Dr Foote and Kym can share data and resources. The office should also have an Internet connection for browsing, ordering, email and other communication. A network modem/router, possibly wireless, would help achieve this.



Activity 6.2 – Sub-systems 1. a Identify the three areas of importance to be considered in developing a computer

system.

b One of these can in turn be divided into three further sub-systems. Which is it and identify these sub-systems.

2. a What sort of processes are involved in the human system side of computing?

b Choose one of these and explain its role and importance in the effective running of the computer system as a whole.

3 a What is the difference between hardware and software?

b Find out what is meant by the term firmware?

c Identify three different forms of software.

d What is the interface in a computer system? Why is an effective interface so important?

4. What advantages are gained by networking computers together?

5. A possible solution for Dr Foote is given above, but it lacks specific detail.

Prepare your own solution by doing the following:

a Make a list of the hardware you think Dr Foote might need.

b To this list add the software you think would assist her in running the office.

c Suggest a suitable operating system that could be used in the situation described.

d Use computer magazines, adverts or web sites to prepare a cost effective information system for Dr Foote.

In doing this keep in mind:

Dr Foote does not have a lot of money to spend so the system purchased will need to be cost effective (i.e. pay for itself over time)

the computer system must be simple to use and easy to maintain the system will need to be flexible so that it can be upgraded as time goes by.

e In a paragraph justify the decisions you have made in developing this solution.

6. Make a diagram showing an appropriate layout design for the office indicating the waiting area, receptionist, consulting room, etc.

On the diagram position all of the equipment you identified in Q5, as well as desks, filing cabinets, desk diary, phone, fax, photocopier, intercom, etc.

Design the layout so that the most frequently used items are within easy reach, and so that the space is effectively and efficiently used. Ensure that no “clutter” exists between the receptionist and patients at the counter.

For confidentiality the receptionist’s screen should not be visible to patients.



Inside the computer Computer power has risen dramatically since the first electronic computers were built in the 1940s. Current computers are many times more powerful than those early models that filled whole laboratories.

Gordon Moore, one of the founders of the processor manufacturer Intel, coined a guideline which now bears his name. The so-called Moore’s Law proposes that computing power will double every 18 months. This suggests that, on the whole, if you compare current computers with those of a year and a half ago you will see they have about twice the processing power. Moore coined this proposition over 40 years ago, and so far his hypothesis has been borne out!

Despite Moore’s Law, the rising expectations of users are placing greater pressure on hardware resources. Computers are now handling sound, animation and video for multimedia and internet streaming which are placing increasing demands on the power and memory needed.

In this section we will investigate the basic hardware used in a computer system. Before we look at the parts of hardware however it is necessary to gain some idea of what underlies the operation of the computer.

Analogue and digital To understand what a digital computer is we first have to appreciate the distinction between digital and analogue.

The computers we know are devices that work with digits (numbers), and so are called digital computers. Less common but still around are analogue computers such as the neural networks we will see in unit 10. An analogue device is one that handles a continuous range of values rather than distinct number values.

Most of the world is analogue. Our ears can hear a range of tones from under 100 Hz up to over 15 000 Hz – including all of the frequencies in between. Cheetahs travel at speeds of from still up to 80 kph – again including all of the speeds in between. The second hand on an analogue watch sweeps out all of the seconds in a minute, including all of the fractions of a second as it travels over the watch face. Each of these is an example of an analogue measure – a continuous range of values, including all quantities in between any two measures.

A digital device on the other hand deals only in distinct values. A light switch must be off or on. A digital watch records only the specific seconds of a minute (not the in between parts of a second that an analogue watch does). A digital camera image is made up of separate dots. All

A digital watch only shows separate, whole seconds

The sweep hand of an analogue

watch covers all fractions of a second



of these digital values are discrete – it easy to see where one ends and the next begins, there is no in-between part.

Perhaps the best example of the distinction between analogue and digital is with a light that is controlled by both an on-off switch and a dimmer. The on-off is digital (two distinct values) while the dimmer is analogue (the light can be set to any of a full range of levels of brightness).

Analogue signals are usually represented as a wave form, while digital signals by a square wave.

1 1 0 0 Analogue signal Digital signal

As computers are digital, and all forms of input and output humans use are analogue, we must constantly be converting from one form to the other in order to pass values that are useful.

One example of this is where we want to send digital computer values over an analogue telephone line. Using a modem the digital signal is converted into sounds that are transmitted over the connection. At the other end the analogue sounds are demodulated back into digital values by another modem for input back into a computer.

Analogue to digital conversion

While analogue is more common in the real world it is not as convenient to handle as digital. Digital values are simple to transmit, maintain, compress, and to check for errors. For these reasons it is now becoming common to make recordings digital. A music CD uses digital coding, mp3 is a digital format, images are stored as bitmaps (bmp or jpg) and video is digitised for DVD. HDTV is a digital form of television transmission. It seems the world is going digital.

An enlargement of a bitmap showing how analogue curves are represented by digital pixels.



Bits and bytes Computers are electronic and, because of the limitations of electricity, digital computers are restricted to using only the states of on and off. For ease of handling these states are represented as the two digits 0 and 1, and are referred to as bits (short for binary digits). All computer data is stored or transmitted as bits:

0 – off, uncharged, unpolarised, low (or negative) voltage 1 – on, charged, polarised, high (or positive) voltage

For transmission or storage it is usual to group bits together. A byte is a group of usually 8 bits,

e.g.

By using bits and bytes we can electronically store and transmit text (writing) and numbers.

A byte can hold one character such as a K or a % or an 8 . In the ASCII code most commonly used to represent text, the character K for example is represented by the byte 0100 1011 .

Binary numbers 0s and 1s are also used to represent numbers in binary (base 2) arithmetic.

All numbers possible in our decimal (base 10) number system can be represented in binary. The two bytes 1011 0101 1001 0011 hold the decimal value 46 483. Each 1 or 0 in a binary number has a place value of the power of 2,

e.g. 1001 11012 = 1 x 27 + 0 x 26 + 0 x 25 +1 x 24 + 1 x 23 + 1 x 22 + 0 x 21 + 1 x 20 = 15710

In addition to the simple representation of a value, binary can also represent large numbers. By structuring multiple bytes with meta information they can store floating point numbers. These are the binary equivalent of what is called scientific or exponential notation in the decimal system; for example 3.14 x 1023.

Binary memory Early computer memory consisted of magnetic cores woven onto strands of wires in a 32 x 32 grid. Each of these core memory grids could hold 32 x 32 = 1 024 bits of data. As this was close to one thousand they were said to hold one kilobit of data. The unit 1 024 is still used in memory values.

Common units of memory measurement are:

kilobytes (kB – 1 024 bytes) megabytes (MB – 1 024 kB) gigabytes (GB – 1 024 MB) and terabytes (TB – 1 024 GB).

A kilobyte will hold about 20 words of text, a megabyte about 45 pages of writing, a gigabyte over 150 books, and a terabyte a school library of books. In terms of media content, 1TB can store over 250 high definition movies.

Note: These measurements are not to be confused with Mb, Gb, etc., which refer to mega bits and giga bits (not bytes). Units such as these are usually used to measure download speeds. To further confuse the issue the International Electrotechnical Commission (IEC) released a new set of measurements such as MiB, GiB, etc., which are based on a multiple of 1 000 instead of 1 024.

1 1 1 10 0 0 0



Von Neumann architecture Shown below is the traditional, or von Neumann, model of a computer.

Von Neumann model of a computer – showing flow of data

Dr John von Neumann in 1946 was the first to suggest that the computer could be used as a general purpose device by reading into memory a different program for each task. Previously each computer had been hard wired to do one specific task. This development was so effective that this model has been the basis of most past and current computers, and is known by his name. We will look at the parts of the von Neumann computer in the next section.

For the last decade the processor industry has been reaching the limits of miniaturisation and speed that can be achieved with the von Neumann model.

One of the limits to speed is the so called von Neumann bottleneck. (A bottleneck is a narrowing in a channel that restricts flow along it, here data.) The von Neumann bottleneck occurs because in this way of structuring computers instructions and data from memory must travel along the same bus (data pathway). This means that pipelining is not possible, i.e. reading in the next instruction to the processor while the results of the last are being passed out. In the von Neumann model instructions can only be executed sequentially, one at a time.

There are also limits to miniaturisation with this model. Each instruction in the processor relies on a pulse of electricity and, as there are hundreds of thousands of operations per second, heat builds up very quickly. The accumulation of heat disrupts the operation of the processor and can overheat the whole system. There is also a limit to the maximum speed information can travel, especially along pathways as complex as in microchips. This can become a problem when the data simply cannot cover the distances required in the time available, thus slowing down the whole process.

Dr John von Neumann



In addition the extreme miniaturisation in integrated circuits has led to the problem of electrons jumping from one path to another, causing computational errors.

To overcome the limitations in speed and miniaturisation of the von Neumann model, alternative architectures have been suggested. The alternatives range from conceptual, to practical, and even to the style of programming they facilitate. Many of these alternatives can only perform very limited tasks and have been specialised to the point where they cannot be used for more general tasks. In some cases von Neumann systems can emulate or be part of the alternative architecture.

Some examples of alternative architectures include:

distributed computing – work is split up into units or tasks and allocated by a master control to any number of distinct computers

parallel processing – with more than one processor in the one computer clustering – where a group of independent computers is managed as a single system reduced instruction set computer (RISC) processors – a Harvard architecture where

data and instructions travel on multiple buses (lines) simultaneously artificial neural networks (ANNs) that mimic the operations of neurons in the human

brain field programmable gate array (FPGA) – a processor that can be configured by the

designer or user after manufacturing; can be mixed with von Neumann, RISC, or other processors.

One other area of alternative architecture is the quantum computer. This theoretical computer has the ability to solve problems that are not suited to traditional architecture, with much greater efficiency. Practical examples of problem solving using quantum computing methods have been successful and research is ongoing. While quantum computing is not suitable for general purpose processing it could be used in conjunction with current computers.

Activity 6.3 – Fundamentals 1. a What is Moore’s Law?

b The personal computer (PC) was introduced in 1980. Using Moore’s Law, by what factor would we expect today’s PCs to be better than the original IBM PC?

c Use the adverts in a computer magazine compare an entry level computer from 18 months ago with a similarly priced one today. Would the improvements validate Moore’s Law?

2. a Explain the difference between analogue and digital.

b What are four advantages of using digital storage of data over analogue?

c Is optical fibre an analogue or a digital form of communication? What advantage does this form of communication have for the National Broadband Network (NBN) which is based on fibre to the home?

3. ‘The world is going digital’. What does this mean, and why might it be happening?

4. a What is the abbreviation bit short for? © Kevin Savage 2011 – Licensed to Hillcrest Christian College for use in 2011-12


b Why must computer values be stored as bits?

c What decimal value is stored by the binary number 1101 0101?

d Why is a kilobyte 1 024 bytes and not 1 000 bytes?

e How many bytes are there in a gigabyte?

5. a Make a diagram of the von Neumann model of a computer.

b What is the von Neumann bottleneck?

c Choose one of the alternative architectures listed and find out more about it. Write your findings in a paragraph and report them to the rest of the class.

6. What is a quantum computer? In your answer refer to qubits, basic operation, and the potential applications of quantum computing.

Processor The central processing unit (CPU), or processor, in a computing device is contained in an integrated circuit on a silicon chip. It is made up of many millions of transistors in an area not much larger than your finger nail.

The chip consists of microscopic layers of silicon, the circuits of which are made by etching or depositing different materials such as aluminium or chemically impure silicon. The lines of the circuits on the chip are so small that one hundred lines are less than 1 mm wide. Single chips are now being produced with over 2.3 billion transistors on each.

The processor is the device that follows the steps of the program the computer is running. It is made up of a variety of sections, each with a specialised task:

the control unit (CU) directs and synchronises the operations of the processor; it does this by reading instructions from memory, decoding them and then activating circuits to execute each instruction

the arithmetic and logic unit (ALU) carries out the instructions from the CU; it includes components that can add and compare numbers and store partial results; current processors have a floating point unit (FPU – a maths co-processor) as part of the ALU to assist with complex mathematical operations

a level 1 cache which is a small, very fast memory area for immediate use by the processor; many processors have a secondary cache known as the level 2 cache which is bigger and greatly assists media intensive tasks

an I/O unit which consists of registers to control the flow of information to and from the processor and the rest of the computer, and

a variety of accumulators, registers and buffers that hold data as it is switched between the parts of the processor.

To make an analogy with a person, the CU could be seen to be the brain that directs the body, the ALU the liver and stomach that process input, while the I/O system is the senses and voice.

A silicon chip



Taking the analogy further the cache would be a scrap pad for writing quick notes, and the FPU a calculator.

Because of the complexity and level of miniaturisation these processors are extremely expensive to design and develop. However, once production is underway they can be manufactured in batches of thousands. When development costs have been met, the cost of each processor soon drops. The current main producers of processor for desktops are Intel and AMD, and ARM for small mobile computers.

The processor is mounted on a motherboard – a rectangular circuit board that links all of the components inside the computer. The connections are made through pathways called the system bus. The speed of the system bus (e.g. 100 or 400 MHz) affects the rate at which data can pass between the components and hence the speed operations can occur at.

Graphics processors Increasing demands are being placed on processors and memory nowadays. Multimedia computing for example calls for the manipulation of blobs (binary large objects, e.g. video images or sound files) which require very fast and complex processing. This demand seems to be following a spiral pattern. As processors become more powerful developers find applications that push them to the limit so that more powerful processors need to be developed. The expectations of users and the more sophisticated applications being produced lead to a rapid obsolescence in processors.

To counter this trend many computers now employ the use of a graphics processing unit (GPU). The GPU supports the main processor by handling the complex maths involved in images to be output to the monitor, especially in gaming.

Modern computer games require intensive processing to model the complex movement of 2- and 3-dimensional representations, and to handle the realistic display of light, shadow and texture. As the gaming industry has developed and expanded over the past decade, the quality of graphics expected by game players has increased. In an example of consumer expectations and demand driving development, this has led to fierce competition to create more powerful graphics cards.

While processor manufacturers were improving the power of single core processors, the graphics industry chose a different direction to focus their efforts. Graphics cards are massively multi-core, and rely on hundreds of comparatively primitive processors to work. These cores are designed to do very specialised maths, very quickly.

Recently the power of these GPU cards has drawn attention from academics. The inclusion of dedicated physics processors has allowed real world representation of movement in not only games, but in scientific modelling. The idea that these specialised functions could be used more widely caught on quickly, and GPGPU (general-purpose computing on graphics processing

A personal computer motherboard



units) has became a standard feature on many cards. This general-purpose programming ability of GPUs has attracted the attention of universities who see them as a cheap and powerful processor that can be used to solve complex maths problems. Some universities are already building large computer clusters of GPUs.

An Nvidia GPU card

It is possible that in future the graphics card will become as important to personal computers as the main processor. The CPU would deal with general programs while diverting real-world-problem processing to the graphics card.

Main memory All programs and data in computer systems are recorded in some type of memory. This can be on silicon chips linked directly to the processor or on secondary devices such as hard drives, optical discs like DVDs, or even remotely on the internet.

The main memory is the section of the computer where programs and data are stored ready for rapid access when needed. The processor has direct access to main memory and so it must be very fast to avoid slowing down the operations of the processor. Access speeds as quick as 6 or 7 ns (nanoseconds – billionths of a second) are common now. To show how brief this is 6 ns compared to one second is the same as 1 second is to 5 years!

This type of memory is made up of multiple lattices of very small wires or printed paths. At each intersection in this grid is a microchip and a capacitor which can store a single on or off state for access while the system is running. The capacity of these grids is usually 1024 or one kilobit. This was done as it allows base 2 addresses which can be directly linked to physical locations in memory. Other digital storage, such as disc, does not have this quirk which has led to a debate over how units of memory should be described. (As mentioned earlier with the reference to the IECs MiB and GiB units.)

All types of memory can be divided into ROM (read only memory) and RAM (random access memory).

ROM holds static instructions needed to run systems such as the BIOS (basic input output system) and its replacement EFI (extensible firmware interface) on all consumer computers. ROM is permanent and persists even when the computer is turned off.



RAM on the other hand holds current programs and data. These have only been read into the memory since the computer was last switched on. Data in RAM is lost when the power is turned off.

Nowadays this type of memory is principally held by DDR SDRAM (double data rate synchronous dynamic RAM) modules that consist of millions of transistors that can switch and hold a 1 or 0. A typical desktop nowadays will come with at least 4 GB of DDR3 RAM as standard. This can be increased by the addition of more RAM modules that plug into memory sockets on the motherboard. The more memory available the less disc access is required, and the quicker and more smoothly the computer operates. These memory modules are currently 240 pin DIMMs (double inline) and are incompatible with earlier generations of motherboards.

Larger computers used in corporations and universities use other more expensive forms of memory such as magnetic bubble memory or CCDs – charged coupled devices. Many enterprise servers will use RAM capable of correcting errors in data collected from secondary devices.

Most computer systems now use some very fast memory as a cache. The cache holds data recently obtained from disc on the assumption that the same data may be needed again shortly. The processor will check the cache each time it needs data and if it is available (a hit) can get at it very quickly without having to wait for the data to come from the disc. Level 1 cache is on the processor, and while level 2 used to be separate, it is now also on modern processors.

When the processor calls for data the search order is, look in level 1 cache, then level 2 cache, then main memory, then hard disc, and finally any other (e.g. flash drive, internet). If the data is found at any point the search stops. Cache memory is kept small to limit search time (and because it is very expensive), but its use can significantly speed up operation.

Secondary memory Secondary memory includes all of the devices used to store bulk data that is not immediately needed by the processor. Common forms include hard discs, flash drives, Blu-Ray discs, and tape cassettes.

While secondary memory devices are much slower than main memory (taking milliseconds – ms rather than nanoseconds – ns) the relative cost is very much less. As a result slow but cheap secondary memory devices are used for storing bulky data such as programs, files, and data that are not immediately required by the processor. When needed the data is loaded into main memory so that the processor has swift access to it.

Main memory – a DDR SDRAM DIMM



Disc storage So called “volume storage” in modern computers takes the form of a hard disc drive (HDD), a solid state drive (SSD), or a mixture of the two. In mobile phones and tablet computers with a large storage need, a less specialised version of SSD is used, usually a single microchip instead of a large array of chips. All of these forms of mass storage tend to be referred to (confusingly) as hard disks or HDDs.

The true HDD consists of one to five aluminium platters (discs) encased in a sealed housing to avoid contamination with dust. Each platter has a thin layer of magnetic material coating it and, as the platter is spun, a read/write head moves over it. To write to the disc a pulse of electricity through the head polarises (magnetises) sections of the disc. The same head can later pass over the disc to read this pattern of 1s and 0s. In consumer computers capacity is measured in terabytes, with 2TB or larger being typical in desktop computers.

SSDs on the other hand are similar to several sticks of standard computer memory joined together. They employ a technique that allows the information stored to survive even when the mains or battery power is not available. SSDs are in the same form factors as standard PATA or SATA HDDs, i.e. 3.5" or 2.5". They can be mounted directly on the motherboard of the device or computer, but are usually in the same removable form factor as HDDs.

HDDs usually stay permanently in the computer, although portable or external HDDs are becoming common for holding or transferring large data files such as backup or media content. These can be moved from computer to computer and connected via USB or ESATA. An alternative portable option is the use of USB Flash Drives that can store gigabytes of data on a small chip. Flash drives use the same technology as SSDs.

Optical storage CDs, DVDs and Blu-Ray are all forms of optical disc used for mass storage of data.

A computer CD (compact disc) works similarly to an audio CD. The data is encoded as a series of depressions or pits cut into the surface of the disc to represent 1s and 0s. These are read by a laser that bounces light off the pits. A CD can store 700 MB of programs, video, music, data, games, etc., in a compact (12cm) form. CDs are rated at speeds such as 52x which are multiples of a nominal rate. (1x is the speed of a standard audio CD.)

DVD (digital versatile disc) works similarly to CD but holds more data. Depending on the format DVD can hold 8.7G, 9.4 GB or more data. Some are even four sided (i.e. two layers of pits on each side).

The Blu-Ray (BR) disc is the next evolution of the optical storage format. The name refers to a “blue” laser that is used to read the data. This format is capable of storage of up to 128 GB depending on the standard and discs used. High definition movies are most commonly distributed in BR format.

Inside an HDD showing the top platter and the read/write head



All three standards of optical disc allow several options for users to create writeable discs. These are denoted by a postfix to the name:

–ROM (read-only memory) discs cannot be changed –R for discs which can be written to once –RW for re-writable discs that allow their content to be changed at a later time.

For all forms of writeable disc the surface has a photo-reactive dye layer that can be burned away by laser to create the pits to store data. This process of recording data is usually referred to as burning.

Network storage In addition to sharing data and resources, networks are also important in providing access to secondary memory for the storage of data.

Networks that require large quantities of data to be available to their users will often use storage clusters known as network attached storage (NAS), or for specialised networks storage area networks (SAN). NAS devices are dedicated storage computers that do nothing but connect to a network and provide a large storage space for data. While this style of storage was once used mainly by enterprises it is becoming more common in everyday households as rich media such as HD video requires an increasing amount of space.

Secure storage is of primary importance, and so most networks will also have some form of data backup in case files are lost for one reason or another. The simplest form of backup employs storage tape using QIC (quarter inch cartridge) or DAT (digital audio tape). Some systems use discs as backup, and those with a more critical data set may use mirrored drives. This is a second set of disc drives that constantly copy everything from the master set of drives. If the original drives go down the mirrored drives can take over immediately.

A RAID (redundant array of inexpensive discs) is used in networking situations also. In this setup a bank of HDDs has data stored, not on one but across a series of the discs. (The data is said to be striped across the discs.) The RAID acts like a single disc but has better fault tolerance, data storage and reliability. There is also increased system performance through better I/O of data. If any of the discs fails it can be replaced with a hot swap (without stopping the server) and data restored from a mirrored section of the RAID.

Types of Computer There is a wide range of computers available from those that will fit into a pocket, to those that fill rooms. Below are listed some of the main categories in order of size and power:

palmtop computer or PDA (personal digital assistant) – pocket sized, used for portability, convenience and on-line connection by engineers, accountants, real estate agents, etc.

A Raid-5 array



laptop or notebook computer – used away from main place of activity by sales people, students, journalists, lawyers, etc.

desktop computer – the standard computer used for general computing activities work station – a more powerful desktop computer used for intensive applications such

as CAD or DTP server – a powerful computer with rapid memory access that acts as a data source for

a network; servers may provide files, print services, web access, email, etc. mini computer – a desk-sized computer used by agencies such as credit bureaus,

travel companies, educational institutions, etc. mainframe computer – a large computer system, these are used by insurance

companies, banks, government bodies, universities, etc. super computer – the largest computer setup available; these are used by the military,

intelligence organisations, or for weather forecasting.

In addition to the above computers can be used in clusters, where several computers together handle tasks. The advantages of using computers in clusters are:

availability and redundancy – if one computer goes down the others can take up the load giving a high level of fault tolerance

performance – tasks can be shared across the cluster to balance work load and improve the level of service

management – the cluster appears as one device which makes for simpler control.

The disadvantage to using a cluster is the initial cost of setup and the complex controlling software.

Activity 6.4 – Inside information 1. Expand the following abbreviations:

USB CPU ALU CU I/O FPU DDR SDRAM CD-ROM DVD TB Gb GPU BIOS RAID ESATA CD-RW

2. What is meant by the term obsolescence in regard to computer processors?

3. a Explain the difference between ROM and RAM.

b What is each used for?

c What is the cache and how does its use help speed up the operations of the processor?

4. a What is secondary memory?

b Give five examples of secondary memory.

c It is possible to hold all of the data and applications a computer could use in main memory, but this is not done.

Why is secondary memory used even though it is much slower?



5. Find out about one of the following experimental forms of data storage. Briefly explain how the one you chose works, its potential storage capacity, and the likelihood of its wide scale use in the near future.

magneto-optical (MO) magnetic random access memory (MRAM)

scanning tunnel microscopy (STM) atomic force microscopy (AFM)

solid immersion lenses (SIL) atomic resolution storage (ARS)

holographic storage

6. What type or category of computer could each of the following be classified as:

a A Dell Optiplex desktop computer used by an office worker.

b A Cray SX-6 used by the NSA spy agency to decrypt strongly encoded intercepted messages from China?

c A BlackBerry used by an architect while she was at a building site.

d A DEC computer used by an insurance company to process 50 000 accounts.

e An iPad that a university student takes with him to lectures.

7. a List the type(s) of computer used in your computer room.

b What processors are used in them?

c How much RAM is in these computers?

d What sort of secondary storage is available? Identify the storage capacity of each.

e What peripherals (printers, mice, etc.) are used?

f What form of network connection is used?

8. Using a computer designated by your teacher carry out the following:

a Switch off and unplug the power cord from the wall outlet.

b Disconnect every lead from the back of the computer. Some may need to be unscrewed using a small screwdriver.

c Carefully examine the ends of the leads and the sockets they go into and note how each connection is different from the others and how each lead and socket connect together.

d Connectors are described as male or female depending on which fits into which. Identify these.

e Each socket or port has a series of pins, each of which is connected to one wire or cable in the lead. Which connector has the most pins? How many is this?

f If permitted by your teacher open the case of the computer and find each of the following components:

motherboard processor processor fan or cooling fins power supply HDD DIMM slots (memory) GPU (if present) CMOS battery peripheral connectors (USB ports, etc.)

g Identify how many cards are in the back of the computer, and attempt to identify what each is used for.



h Look at the slots the cards fit into. Is there more than one type of slot (and if so can you name them)?

i Are there any spare slots for additional memory modules? If so what is the maximum amount of RAM this computer could have?

j Replace the case and reconnect the leads ensuring each goes on the correct port and is not forced into position.

9. Go to an on-line computer vendor such as Dell (dell.com.au) and choose a suitable computer for each of the following individuals:

a A newspaper reporter who travels most of the time; needs a good reliable computer that is easy to transport and can be used to send stories back to the head office.

b A university student who needs to prepare assignments but does not have much money.

c A copywriter who is working with colleagues overseas to prepare television advertisements.

d A school computer coordinator who needs to fill a classroom with networked computers at a reasonable cost.

Software Software is the set of programs used to operate a computer. There are three main forms of software, applications to run on the computer, an operating system (OS) to run the computer itself, and programming languages to write the two other forms of software.

Applications Applications are the type of computer program we are most familiar with. Word processors, computer games and web browsers are all applications that run on the computer.

Some applications, such as databases and spreadsheets, are general purpose in that they give the user a basic format for doing a wide variety of tasks. Once the application is started then the user is free to use the program to develop different types of activities such as invoices, letters, diagrams or web pages.

In addition to general purpose, there are special purpose or dedicated applications that are designed to do one specific task. Examples of these are a company payroll program, a production line control program, or a computer game.

Sometimes applications are grouped together to form a program suite. The best example is Microsoft Office which includes Word, Excel, Access, PowerPoint, etc. Another example is the Adobe Creative Suite package which includes Flash, Illustrator, Photoshop, Dreamweaver, Acrobat, and so on.

Most modern programs give the user the option to customise the interface. The menu structure, form and position of toolbars, and screen appearance can all be altered to suit the working style of the user. In turn this gives the user more control over their working environment. Current application software is also very user friendly. The use of on-screen, context sensitive help (the message refers to what the user has highlighted) means users do not have to refer to manuals to



find out what to do next. Most applications also include tutorials or wizards that lead the user through an operation they wish to carry out.

One aspect of current applications that is a result of this customisation and user friendliness is that they are becoming increasingly large and complex. The increased elaborateness of programs is in turn making greater demands on hardware in processor requirements, memory capacity, and data storage devices. This trend has become known as feature creep.

An application should be linked to the task the user is required to carry out, and the user should be aware of which application is best suited to the task at hand. For example, while the distinction between word processing and desktop publishing applications has blurred, in certain situations one is better to use than the other. When it comes to web publishing certain programs will only go so far, and it is up to the user to know when to move onto a specialist web publishing package to achieve what they want to in the best way. It is the user’s responsibility to choose the correct application for the job.

Remote applications A recent trend has been for an outside company to provide the software for an organisation. An application service provider (ASP) will own and host on its own server each of the applications an enterprise uses. As a user needs to use a program he or she will access it from their desktop as normal however, instead of coming from the local HDD, the parts of the program needed (and only those parts) will come from the ASP’s server. This might be over the internet or a dedicated link. If more parts of the application are needed they are downloaded as required.

At first sight this on-demand software might seem an inefficient method, but it can be very effective. Instead of buying dozens of copies of software and then loading these onto each machine in a network, the ASP provides the programs only as needed. The maintenance, correcting and updating of dozens of copies of the software does not have to be done. Any changes in setup can be made at one central location. Security and virus control is centralised. Costs are also reduced. Instead of paying for all of the features of complex applications, the enterprise pays for (rents) only the parts of the programs that are actually used. In some cases savings can also be made on the type of computer being used with a thin-client (minimal computer) running the remote applications.

A variation on the ASP in more general use are web-apps. These are programs ordinary users can access on-line and run in their browser. The most popular of these web-apps is Google Docs which includes word processing, spread sheeting and presentation software, though there are many other apps available. Most web-apps can also store the user’s files on-line, and so take up little local disc space.

To be effective both ASPs and web-apps rely on reliable broadband links between the provider and user. This is so that programs are always available, and run as smoothly as if they were hosted on the local machine.

Operating systems When using a word processor we type at a keyboard and are not surprised to see letters appear on the screen in front of us. Clicking on Print causes a copy of the screen to appear on paper. But what determines how the keys we press are converted into images that are placed on the screen, or emerge on paper? The answer is the operating system of the computer we are using.



The operating system (OS) is the program (or set of programs) that manages the flow of information in a computer system. Data must come from the keyboard or the disc drive, into memory or the processor, and from there must be sent to the screen, the printer or other computers. This does not happen by accident, and in fact must be carefully controlled. The movement is very fast and complex, and to avoid clashes, losses, or errors the operating system must determine what goes where and when.

There are as many different forms of operating system as there are types of computer. The system selected for your computer depends on both the level of user friendliness required, and the purpose the computer is to be used for.

Currently common OSs include:

Microsoft Windows 7 Macintosh OS X Unix and GNU/Linux Nintendo, Playstation, etc. for games consoles VMS and SAA/SNA for mainframe computers.

Despite the wide variety of operating systems there is little standardisation. Software can often only run on one type of OS, e.g. Xbox games will not run on a Playstation.

At the personal computer level one effect of this is that Macintosh software will not run on Windows computers without special emulation software. One way of emulating another OS is to set up a virtual machine, i.e. where one computer is used to carry out or mimic the actions of another. An example is Parallels Desktop for Mac, an application which permits the Windows OS and programs to run in a window on the Macintosh.

Types of OS The types of OS we are most familiar with are those that manage the standalone and the networked computer.

A standalone is independent of other computers and has the advantage of being able to function by itself even if there are problems with other nearby computers. The standalone is however becoming rarer nowadays with the advantages networking offers such as the sharing of information and resources.

More sophisticated OSs offer the options of multiuser, multitasking, or parallel processing.

A multiuser operating system is one in which more than one user is accessing the computer at a given time. A good example is an ATM network for a bank. Here a mainframe computer located in a capital city can control up to hundreds of ATMs. In sequence the mainframe polls each ATM to see if it wants to do some processing (deposit, withdrawal, etc.) and if it does, gives it a few milliseconds of processing time. (This is an activity called time slicing.) To each user at a multiuser terminal they appear to have exclusive use of the computer, yet all of the terminals are sharing the same processing.

In the past multiuser systems were common, especially in places such as universities or offices where users all accessed one mainframe. Nowadays they have been replaced by networked computers, with the move toward decentralisation and more local autonomy in processing.



A multitasking OS allows one processor to perform several jobs at once, to gain the maximum benefit from a processor. The operating speeds of a processor are in nanoseconds (ns, billionths of a second) while disc access is in milliseconds (ms, thousandths of a second). Rather than have a processor stand idle for several milliseconds waiting for data from disc, it is often better to get it to work on another job in the meantime. In this way a fast, powerful processor will switch jobs in and out, working on several programs at once.

Another development is parallel or distributed processing where an OS uses more than one processor to handle its task or tasks. This may be in one computer, or alternatively on computers spread across a network (perhaps at night when they are idle). Naturally the OS that can handle a series of tasks over many different processors must be very sophisticated and complex.

File maintenance Murphy’s Law states: “If anything can go wrong, it will go wrong, at the worst possible time”. This is never more apparent in computing than when vital files are lost.

Backing up files should be part of your working strategy in operating a computer. It is important to make regular backups; the more significant the files – the more regular the backup. It is also a good idea to either leapfrog your backups (i.e. keep one or even two generations of earlier backups before overwriting them) or to use incremental backups. This is where you make a regular full backup, and then make more common smaller backups of files that have changed since the last full backup.

Backup can be made by using special backup programs or by compressing the required files using a zip (or similar) compression utility. Zipping can also be useful for compressing files for transfer over the Internet as it reduces file size.

If you lose some files it will be time to call on your earlier backup and restore the lost data.

A virus is a self duplicating program that copies itself onto computers and performs an action the computer owner does not intend. Again it is part of your working strategy to regularly check your hard drive(s) for viruses, and most importantly to check every flash drive, and every internet download or email for viruses before you give them access to your computer.

There are often unwanted files on disc that it is a good idea to remove every now and then to remove clutter and save disc space. Two places these can be found are in the recycled/trash bin or temp folders.

When files are deleted from a hard disc they are sent to a special folder (the recycled folder in Windows). This is a precaution to prevent accidental deletion, or to hold files in case you later find you still want them. This folder needs clearing out every now and then.

Many applications, as they are running, write temporary files to the hard drive. These files hold data that will not fit into memory (RAM). When the application is closed the temporary files are normally deleted automatically. If, for any reason, the application does not close normally (Reset button pressed, power failure, etc.) the temp files remain on disc. After a while these can clutter up the hard disc and should be removed. In Windows computers these files are recognised by the extension .tmp or by having a tilde ~ in front of the file name. In deleting any files be very sure they are not files needed for the computer or applications to run.



Finally after time a disc will need to be defragmented. When files are written to disc then are placed in the nearest available spare slot. As some files are deleted they leave gaps that parts of new files are written into. Over time there will be parts of files inefficiently spread all over the disc. The process of defragmentation collects these parts of files together in one continuous band that frees up disc space and speeds up file access.

Activity 6.5 – The computer at work 1. What type of application is each of the following programs?:

CorelDraw Word MYOB Excel

FireFox Access AutoCAD MS Project

DreamWeaver PowerPoint Flash Photoshop

2. a Re-read the task description about Dr Jill Foote and identify the areas of office and financial procedures that could be computerised in her office.

b Suggest general purpose applications to do each of these tasks.

c What type of OS would best suit the operations of this office? Suggest a specific operating system that could be used in the situation described.

3. Mike is about to start at UQ in Brisbane studying engineering. His parents have offered to buy him a laptop to assist in the preparation of assignments, the formation of diagrams, carrying out calculations and the recording of information he will need to be familiar with. Mike will need to purchase software to do this.

a By looking on-line, through computer magazines, or by visiting computer shops obtain a list of the applications (with prices) that you think he will need.

b As Mike is on Austudy he has limited money available. Find a list of free programs that could be used as an alternative to the above list.

4. In your own words describe the function of an operating system.

5. Suggest a suitable operating system for each of the following situations:

a A high school student with a laptop computer.

b An insurance agent who wishes to communicate with clients.

c A bank system manager who wants to manage her IBM mainframe.

d A Nintendo DS handheld.

e An advertising copy writer preparing a newspaper advert.

f A classroom network.

6. a What is an app store?

b In what ways does this form of software distribution differ from more traditional methods?

c What are the advantages in obtaining software in this way?

7. Explain what a virtual machine is, and explain why it might be needed.



Networking The standalone computer is becoming a thing of the past. The convenience and efficiency of connecting computers together has led most organisations to alter their work practices and to form networks of computers. Home networks are now also common.

A network is a set of computers that are linked together to share data and communications. There are three principal advantages of using a network in place of standalone computers:

data and programs can be shared between many computers so that for example several people can work on one report, all users can access a central database, or alternatively networked games can be played

resources such as printers, scanners, CD drives, or a broadband router do not have to be attached to each computer, but can be communal

the ability to communicate with colleagues electronically through email, messaging or video conferencing.

A group of computers in one location (an office, a school, a work site) is called a local area network (LAN). Computers that are connected but not on the one site are called a wide area network (WAN). Examples of WANs are the computers of an organisation located in different cities or even countries, and bank ATMs. The ultimate WAN is the internet

Network configuration Each device connected to a network is called a node. Nodes include computers, printers, CD drives, modems, laptops, scanners, and so on.

There are two principal configurations for a network:

peer-to-peer – each computer on the network is equal to others and can be configured to share peripherals (such as printers) and disc space, including disc space on other computers

client-server – one computer acts as a file server or master computer that directs data, communications or printing tasks for a group of other computers; the file server may be dedicated (do no other job) or may be able to also act as a work station itself.

Peer-to-peer networks are cheaper to establish (no file server) and easier to set up and administer. They are normally used in a LAN with up to 25 nodes. A peer-to-peer network is ideal for situations where the main activity is using general applications such as word processing and spreadsheets as in a small business, a classroom, or in the home. There is usually a form of password access to log into the system but generally not much greater security.

The disadvantages with a peer-to-peer network are:

it is limited to about 25 nodes a particular work station with a peripheral attached can be slowed down as others use

the processor in, for example, a print job there is also usually a high RAM overhead in running a peer-to-peer network, and all machines need to stay switched on all the time for users to access all facilities and

data.



The client-server model can handle up to 200 nodes and can also operate over a WAN. While more expensive to set up and more complex to manage the client-server configuration is more suitable for businesses in that it can provide centralised data on the server. This data (e.g. stock, employee data, order entry, billing system, etc.) can then be accessed by each user in the organisation.

Security on a client-server network is obviously more of an issue with many people able to use one set of data. This is generally handled by having levels of access so that most users are only able to read data and only a few are able to modify or delete it.

Both peer-to-peer and client-server can have a printer sever attached. This is a printer linked directly into the network. As such it does not work through a computer and hence does not tie up one.

Over recent years with the influence of the internet the distinction between these two network configurations has started to blur. It is common for systems in business to be a mixture of peer-to-peer and client-server as each has advantages for some shared tasks and not others. In the same way home computers are increasingly being used to connect to remote networks or meshes to obtain work related or rich-media content using the best available method.

(As a side note peer-to-peer (P2P) file sharing refers to the fact that equal partners contribute to the data exchange, and not to how the networking is structured.)

A variant of client-server that some organisations are using is called thin client. A thin client is a computer that acts as little more than a screen and keyboard. A user’s keyboard or mouse actions on the client are transmitted to the server which does the work. Applications, data storage and processing are carried out on the server. The results are then sent for display on the screen of the thin client. Depending on the type of client, the amount of processing it can undertake may vary from nil, to being able to act as a standalone if the network connection goes down.

The advantages of a thin client network are that they are relatively cheap to set up, and they are very simple to administer with all programs and processing being in one central place.

Packet switching Information that travels over a computer network is usually sent by a process called packet switching. There are other forms of network transmission (circuit or message switching) but packet switching is by far the most common. It is the mode used over the internet and most intranets.

In this form of transmission each message to be sent is divided into parcels or chunks of data called packets. Each packet has a header of addressing information and a tail to mark the end of the packet.

Packet switching is a very efficient form of data communication because the packets of many messages can be interwoven onto one line so that full use is made of the bandwidth.

tail

tail

data

data

data

tail

header

header

header



The process of packing parts of several messages onto one channel is called multiplexing.

C B A

4 3 2 1 3 B q r 2 1 A

p q r

Multiplexing packets

In the diagram the packets from three separate messages have been multiplexed onto the one communication channel. On receipt, the information in the packet header will be used to direct each packet to its required destination.

Multiplexing small packets is very efficient, but packet switching has another advantage. All packets do not have to follow the same pathway to their destination. If one communication line is crowded or broken then some of the packets can be re-directed along other less busy channels.

Bridges When a LAN gets too crowded it can be divided into segments (sections) to prevent traffic overload. A bridge is a device used to link LAN segments. LAN segments that are connected in this way are sometimes called BLANs – bridged LANs.

A bridge is a combined hardware/software device that maintains a list of node addresses. Each packet that reaches the bridge is checked and if necessary sent onto another segment. In this way packets will stay inside of their own segment unless they need to reach a node in another segment. This reduces the network traffic because instead of every device checking the header of every packet on the network it will only “see” packets on its own segment. All other packets are filtered out by the bridge.

A simple bridge will have a look-up table of addresses of devices. Each new device on the network will have to have its address added to the table. This can be a time consuming task for network managers. Smart bridges are more common nowadays. These build their own table of addresses by noting the source address where packets come from and add these to a dynamic



database of device locations. If they detect a new or unknown address they will broadcast it over the whole network until a device responds. They will then update the table with the responding address. Smart bridges use method called a spanning tree algorithm and so they are sometimes called spanning tree bridges.

Bridges can also act as repeaters in that they can regenerate a signal as it passes through them.

Nowadays bridges are not used much on local networks, but are still important on larger networks, including the internet.

Routers While a bridge is effective in linking segments of a LAN together they are not sophisticated enough to work over a wide area network. To work effectively over a WAN packets must be sent directly to a node. Routers read the header information in a packet and extract the network address of the destination device. They then direct (route) the packet to its destination. Routers forward messages rather than filtering them as bridges do. Using routers, packets can be distributed over wide scale networks – in fact without routers the Internet could not work.

Some routers have additional functions such as prioritising some messages, handling security or multicasting (sending the same packet to many destinations).

For the home user access routers or modem/routers are used to link several computers or laptops to the one internet connection, sometimes using a WiFi (wireless) connection.

Switches Bridges and routers tend to be slow and relatively expensive and have now mostly been replaced by switches in general purpose settings like homes and small business. (Routers are still important in enterprise networks with most sites having at least one, and often hundreds.)

Switches combine many of the functions of bridges and routers, but operate at very much faster speeds. Instead of waiting for a whole packet to be received before sending it on a switch has a flow through facility. The addressing information in a header is read, and the packet directly piped into the required output port even before the rest of the packet arrives. This is described as minimal latency.

Using specially designed circuitry switches can rapidly direct packets, assign different bandwidth to various output ports or even handle parallel data flows. Advanced switches incorporate data management functions, prioritisation of data flow and bandwidth allocation.

The main disadvantage a switch has, compared to a router, is that it does not support dynamic routing. This means that they cannot transfer data from one network to another. They also have no support for NAT (network address translation – the remapping of one IP address into another) or advanced filtering.

A wireless router

A simple network switch



Network management As networks with their associated advantages are becoming standard for computer systems they also have a cost in terms of management. Someone has to take over the role of manager, and depending on the type, size and complexity of the network, this may involve a full time position, the need for a management team, or even an out-sourced management arrangement.

While each computer system may have different forms of management, the role is consistent. It involves controlling users’ access to network equipment and resources, overseeing resource usage, responding to bottlenecks in the system, planning expansions of hardware, and general maintenance as well as trouble shooting and fault fixing.

The role of network manager specifically involves:

user management, e.g. establishing users’ accounts and level of access, and the reporting of usage statistics

installation, customisation and maintenance of software, hardware and the operating system

level of access, i.e. which individuals or groups can be permitted to read, create, modify or delete which data

network file maintenance, e.g. the clearing away of unused files establishing remote access (e.g. via modem) or email if required preserving copyright backup either to a tape backup unit or through using a mirrored HDD array (one that

copies the main drive constantly and is available if it goes down) network security: preventing unauthorised access, physical security of the hardware,

data security, establishment of firewalls and avoiding viruses user confidentiality and privacy.

Overall the network manager is responsible for the quality of service (QoS) of the network i.e. how well it supports the operation of the enterprise. The manager may also be responsible for training and user liaison and therefore good communication skills are important.

Activity 6.6 – Inter-connectivity 1. a Why might a school computer coordinator decide to change the configuration in a

computer room from standalone computers to a network? Give at least three reasons with examples.

b Would the introduction of thin client work stations be an advantage or a disadvantage in a school situation?

2. a What is the distinction between a LAN and a WAN?

b Give an example of each that you know about.

3. a Explain what packet switching involves.

b What sort of information is stored in the header and tail of a data packet?

c What is multiplexing and how does it improve the efficiency of communication. Use diagrams to assist in your answer.



4. Explain with examples why a client-server configuration is more suitable than peer to peer for a medium sized business such as a local newspaper.

5. Explain the difference between bridges, routers and switches.

6. a Identify the areas that a network manager is responsible for.

b Of these areas which do you consider to be the three most important? Justify your choices.

c What are the aspects of QoS that can be affected by poor management practices?

7. Bunyip high school has several hundred computers. Recently students have been downloading games and playing them during class time. There has also been a spate of derogatory comments sent via the school’s email system.

While you may not see a problem with the odd game, and value student email as a vital form of communication, it is your job as the system administrator to define what is acceptable on the school computers. You have been asked by frustrated staff to write an acceptable use policy (AUP) that defines what students should do on the network.

As it is not a formal agreement, the policy can be written as short bullet points, however it must clearly state the limits on students. These should include what can be downloaded, what behaviour is appropriate, and what non-class related activities are permitted.

Prepare the AUP.

Network cabling and protocols Any device on a network that can receive or send data is called a node. Every node in a network requires an adaptor or NIC (network interface card) installed in it. The adaptor or NIC enables communication on the network by reading and sending packets using a network protocol.

A protocol is a standardised way of communicating. Protocols are necessary because there are so many variables in connecting devices that unless there is a common, agreed way of doing things, only a garbled message will get through, if at all.

A network interface card

We will investigate protocols shortly, but the most common way of linking the nodes to each other is to use cabling, which we will look at next.



Cabling Nodes can be connected by cable so that signals can pass between them.

The principal forms of cable in use are:

twisted pair – this is basically unshielded copper telephone line (UTP), but may be shielded to prevent ‘noise’ on the line (STP)

optical fibre – long threads of glass strand; information is sent by a flashing LED (light emitting diode) at one end, and read at the other.

The most familiar form of UTP nowadays is the blue CAT5 (category 5) cable used for networking computers. While CAT5 is most commonly used for 100 Mbit/s networks, such as 100BASE-TX Ethernet, it can also be used for internal phone networks or for streaming video.

Optical fibre cable, on the other hand, is now becoming the standard for faster, more widespread data communication, especially with the roll out of the national broadband network (NBN).

Despite being expensive optical fibre has the advantages of:

being fast (up to, and even exceeding, 40 Gbps) having a wide band width, and can carry voice, video and data at the same time being secure in that it emits no radiation and is difficult to “tap” (intercept messages)

without disrupting the cable with a physical connection

being immune to electrical interference, so that it can be used even around generators,

etc., in factories, and is not affected by lightning can be used over long distances.

The expense with optical fibre arises from the complexity of converting electrical signals to light, and then back again.

The principal alternative form of connection to using cables is wireless (radio) communication between nodes. We will look at wireless shortly, but before we do we will look at some of the transmission protocols in use.

Ethernet Ethernet is currently the protocol most widely in use for cabling and network access. It is fast, reliable, cheap, and easy to install. As such has become the industry standard for LANs.

Ethernet is a contention (or collision detection) based protocol. This means that nodes listen for breaks in the network traffic and, if quiet, they then transmit their message and check to see if it gets through. Special software is used to detect if there has been a signal collision if another device transmitted at the same instant. If there is a collision all messages cease and are re-sent

Optical fibre cable

CAT 5 cable



at random times. This first-come first-served method may seem inefficient, but in fact works well if the network is working at anything below 80% capacity.

Ethernet originally ran at 10 Mbps over UTP (called 10Base-T) or thin ethernet (coaxial STP). 10 Mbps is supposedly 10 million bits per second (equivalent to about 900 A4 pages of text) but in practice most 10Base-T networks only achieve about 4 Mbps. This is adequate for file and printer sharing, but not for streaming video.

Quicker standards with faster packet rates include Fast Ethernet (100Base-T) and Gigabit Ethernet (1 000 Mbps). (For comparison most HDD access is in the 80-200 MBps range.) A mixture of 100 Mbit and Gigabit is now standard for home networking.

TCP/IP TCP/IP is the basis of communication over the Internet (and most intranets). It consists of two parts: IP (internet protocol) and TCP (transmission control protocol).

IP takes care of the addressing of packets. An IP address consists of four numbers less than 256 separated by dots (e.g. 204.19.141.35). The first two numbers usually specify the segment address while the latter two point to an individual node on this segment. Every device on the network is assigned its own IP, enabling other devices to send packets to it.

Since large numbers are not user friendly internet IPs have been associated with words through the Domain Name System (DNS). When a user types www.yahoo.com into a browser a DNS server converts it to the IP 204.71.200.68 that the routers delivering the packets of data can understand.

The TCP part of the protocol specifies how packets of data are constructed from a message to be transmitted, and how they are recombined at the receiving device. It also checks for errors in packets, makes sure all arrive, and arranges for re-transmission if necessary.

Wireless A wireless network is a group of devices that are linked by microwave communication. Typically there is an access point (AP or WAP) that connects to a wired network and the internet. This broadcasts a signal that can be received by a wireless adaptor in devices such as computers, laptops, printers, MP3 players, smart phones, game consoles, and other appliances.

Wireless connection between devices can be over a local area (WLAN) or a wide area network (WWAN).

WLANs generally operate on the IEEE 802.11 protocol, generically referred to as Wi-Fi. In addition to private use in homes and offices, Wi-Fi can also provide contact with other users, or on-line access, at what are called hotspots. These are publicly available access points that are provided either free, e.g. at coffee shops or restaurants like MacDonald’s, or to users who pay an access fee, e.g. at airports or hotels.

WWANs generally use mobile phone technology, and protocols such as GPRS or 3G, for signalling and data transmission.

The advantages of wireless is there is no need for cabling, cheaper set up costs, and the flexibility of being able to move the receiving device (e.g. a laptop) freely within the access area. There are some limitations in using wireless in that there may be interference or blockage



to the signal which can restrict the range of coverage. There may also be problems of channel pollution where there are too many wireless signals in the one location.

The biggest drawback to wireless is security. It is relatively easy to tap into a wireless network without the owner being aware. This unofficial access can be used for purposes as benign as web surfing, to those more damaging such as breeching security, stealing information, or planting malware. The best defence against unauthorised access is the use of encryption, whereby any data transmitted on the network is coded so that only verified users can read it.

Bluetooth Bluetooth is a short range, low power, wireless communications protocol with good security.

Using Bluetooth fixed and mobile devices can exchange data over short distances creating what is sometimes called a personal area network (PAN). This PAN can include devices such as computers, laptops, mobile phones, printers, faxes, GPS receivers, digital cameras, game consoles, kitchen appliances, and so on.

Bluetooth is a packet-based protocol with a master-slave structure. One device acts as master and may communicate with up to seven slave devices in a piconet (tiny network). Generally data is transferred between the master and one other device at a time, switching from one slave to another in a round-robin fashion. While there is a broadcast option, this is rarely used. If required at any time the devices can switch roles, and the slave can become the master.

Other protocols ATM (asynchronous transfer mode) is a very high speed protocol mainly used by telcos (telecommunication providers) such as Telstra. Because of its speed and reliability it is used mainly for high-bandwidth backbone services such as ADSL. ATM is still too expensive for individual use but may one day become the dominant networking protocol.

Token Ring was a protocol devised by IBM that has largely fallen into disuse. In this protocol an imaginary (electronic) token is passed between nodes. Only the node with the token may transmit, the rest remain silent until they receive the token. It is usually works through a hub and hence forms a star based network (despite the “ring”).

The major advantage of the token ring protocol is that selected nodes may be assigned a priority and have longer use of the token or more frequent access to it. This format is more expensive than Ethernet and operates at 4 or 16Mbps. Cabling is usually UTP, although STP is better. With the dominance of Ethernet it is rarely used nowadays.

IPX/SPX was the protocol used for local area communication before TCP/IP, and was important up to the early 1990’s. It is still used in some router and network software.

Activity 6.7 – Getting together 1. a What advantages does optical fibre have over copper wire for connecting computers?

b Optical fibre is still mainly used for network backbones. Why is it not used for direct connections to computers in a home or office?



c The National Broadband Network (NBN) is currently rolling out optical fibre to the home throughout Australia. In what ways will fibre optic cable enable more access to information than conventional cable?

2. a What is a protocol?

b Why are protocols needed in computer communication?

c List five different protocols used for networks.

3. a How is a token used in a token-ring network?

b What advantages does this type of network offer?

4. Using IPv4 an address consists of four groups of numbers, each less than 256, separated by dots, e.g. 23.134.229.97 .

a What is the largest number of devices that can be connected this way so that each has a unique IP? (256 x 256 x 256 x 256)

b Will this number be enough to supply a unique IP for every device that will ultimately be connected to the Internet?

c Use the Internet to find out what IPv6 is and how it overcomes this limitation. Also see if you can find if any ways have been found to overcome IPv4’s limitations.

5. Research and, in a paragraph, report on one of the following:

a Wardriving.

b Wi-Fi hotspots.

c Bluetooth vs Wi-Fi.

Internet cafe As stated earlier any investigation of computer systems must be aware that they are operated by people, and so we have been mindful of the social and ethical implications, and the human-computer interaction that takes place.

To bring together much of what we have looked at with computer systems we will now investigate the following scenario:

Two friends plan to open and manage a cafe for computer gaming and general internet access. Paul 'Zander' Price and James 'Higgs' Lloyd will be equal partners and have secured significant capital to purchase equipment and pay expenses for the first several months of operation. They have chosen a location in the centre of town for maximum exposure and they expect the cafe to be very popular with the capacity for 35 users.

While no hardware has been purchased yet, and the business model is still being developed, they have settled the role of manager, the software management package to be used, and customer acceptable use policies. The grand opening is in just a month.

When compiling the business case for the cafe several surveys were taken to ascertain what the local population expect to find when they come to the cafe. Several important points were extracted from these surveys:

the majority of customers would be students from the local high schools and university, with shoppers expected to make up only a tenth of the clientele



customers coming for internet usage wanted a range of software titles with the 15 most popular costing collectively $700 per computer and satisfying 90% of those surveyed

those interested in gaming asked for an extensive list of software involving over 350 different applications; of these the top 15 cost $1 800 per computer but would only satisfy 40% of those surveyed

prospective customers under 35, in addition to standard computers, wanted gaming consoles like the Wii and Playstation 3 to be available, on the understanding that these would be free to play if there was ever a wait for access to a computer

local shoppers did not want loud or offensive music, while students wanted a mixture of techno, top 100, and alternative music.

From the outset Paul and James decided that all programs available for use on the system would be licensed. Users would be able to request new applications, though there would be no way to install unauthorised software. The exception to this would be add-ons and custom game configurations.

The network as planned is to consist of a simple client-server arrangement with a central server holding copies of all the software to be accessed. From this server each computer would load the required software across the network. Every computer on the network would have full internet access with a firewall monitoring traffic and a routing device slowing down users who exceeded standard access limits.

Activity 6.8 – Game time 1. a Describe a typical client you expect will use this internet cafe.

b List probable games this user would expect to play.

c In addition to games, what other activities would users carry out? Identify software that would support these activities.

2. What advantages will Paul and James gain by having software on a central server? Consider time, money, space, security, efficiency, effectiveness, and customer satisfaction in your answer.

3. a Make a list of the hardware you think Paul and James will need to support the software you identified in Q1.

b Use computer magazines, adverts, or web sites to prepare a cost effective system for the internet cafe.

In doing this keep in mind:

Paul and James do not have a lot of money to spend so the system purchased will need to be cost effective (i.e. pay for itself over time)

the computer system must be simple to use and easy to maintain the system will need to be flexible so that it can be upgraded as time goes by.

4. a Suggest suitable hours of business for the internet cafe.

b Make a list of the different tasks you think Paul and James will find themselves carrying out during the day. Include as many as you can including user support and equipment maintenance.

c Will Paul and James be able to run this cafe by themselves? If not how many other staff, doing what, do you think they will have to employ?



5. a What is meant by the term total cost of ownership?

b What items other than hardware and software might be included in the cost of setting up this internet cafe?

6. What are some of the human factors Paul and James will need to take into account in establishing this system? (Consider monitoring, security, management, OH&S, and other similar issues.)

7. Make a diagram showing an appropriate layout design for the internet cafe indicating the reception area, computer room, point of sale, and so on.

Activity 6.9 – Computer system case study In a small group undertake a case study of a working computer system in a local enterprise.

Suitable sites will need to be contacted beforehand and arrangements made to visit the various systems. Your school’s library is a good alternative if there is no outside system available.

To carry out the case study you will work through three stages:

as a group prepare a questionnaire to query the organisation your group is to visit visit the computer system comment on the system you have visited.

In completing this study you will follow the DDE cycle.

Design Form a group of 3 or 4 who will visit one of the enterprises.

As a group prepare the questionnaire. It should address the following areas:

purpose – what is the main activity of the organisation (its core business); in what ways does the computer system support this; what other supplementary activities (if any) does the system support (WP, accounting, inventory, etc.); who are the clients of this enterprise; what procedures are in place for dealing with these clients; is any outsourcing used

hardware – what type of computers are used in the enterprise; determine the variety of input and output devices in use; what forms of secondary storage are used; if possible cost the system; what physical security is in place

software – what operating system(s) is used; list some of the major programs employed; identify the backup system (form, frequency, storage); determine software maintenance and programming (if any)

human systems – determine the variety of tasks performed in operating the system; how is the system administered and maintained; what training of staff occurs; what level of user support is in place; how are new staff recruited; what form does HR management take; investigate user comfort (hardware, furniture, programs)

networking – what type of network is in use; what is the number of users; how good is network performance; is wireless networking used; establish the role of email and internet in the organisation; what protocols are supported; what forms of network security are in place.



Develop Prepare the questionnaire. As a courtesy forward a copy to your host before you visit the system.

Visit the system and have at least one member of your group (preferably two) complete the questionnaire. Collect sufficient information to also be able to answer the evaluation questions (see below). Upon completion thank your host.

On return prepare a neat copy of the questionnaire with details filled in.

Prepare and send a letter of thanks addressed to the host at the organisation you visited.

Evaluate In addition to your completed group questionnaire you (individually) should submit answers to the following:

1. Assess the overall effectiveness of the system you have visited. To do this you must identify the purpose and function of the system. In your opinion does the system support the operation of the enterprise significantly? If so, in what ways, if not how could it be improved?

2. In what ways is the use of networking an advantage to the organisation?

3. Discuss the ease of use of the system. In your answer refer to the hardware and software ergonomics, the support given to users, and the functionality of the user interface.

4. Comment on the security system in place. Is it in your opinion effective, or could it be improved in some way?

SEI 6 – The changing workplace Automation is where human workers have some or all of their jobs taken over by machines or a new computer system. Automation can take place in factories with the introduction of robotic devices and computer controlled processes. It can also occur in offices and businesses with the increased use of computers and other technology.

Automation can be costly. Installing new devices and systems has an initial set-up cost, and there is usually a disruption to on-going activities while the new system is installed. It can also be costly in human terms. If a person’s job is automated and they have to undertake a new activity they are described as being displaced. Workers who cannot make the transition or whose job disappears may be made redundant.

Automation however does not automatically mean workers lose their jobs. Often the boring or dirty or dangerous jobs are taken by machines and the human worker takes a position of more responsibility. Usually, if this happens, the worker will need retraining in the use of the new technology.

In the next activity we will explore some of the social and ethical effects that automation can have in the work place.



Activity 6.10 – Automation Read this description of job displacement and answer the questions that follow:

Peter Harris is over 50 and for the first time in his life he is afraid of unemployment. He is a carpenter and for over 30 years has worked for The Furniture Factory in Brisbane. He knows at his age it will be almost impossible to get another job if he is made redundant now.

The reason Peter fears for his job is the automation that has taken place in the factory. He first felt the impact five years ago when the company converted furniture making to an assembly line process to reduce costs and speed up production. The assembly line is fully automated, with robotic devices cutting, holding and securing sections of furniture together. The whole process is computer controlled.

Although the union objected at first, the company argued that robotic devices on the assembly line could work around the clock without tiring or making mistakes. They felt that to stay competitive they had no choice other than to install the assembly line and robots. After discussion and some concessions the union agreed to the process. The initial cost of the assembly line, central computer and robots was high, but they were able to pay for themselves within three years. The Furniture Factory became a leader in its field and profits rose.

At the time Peter was not laid off but he was displaced, that is his job disappeared and he had to take on a new position. The new job was in quality control where he had to inspect the furniture and ensure no poor joints were passed onto the next section. Although the new position required less skill and gave little personal satisfaction he took the job as he had a family to support.

Now five years later Peter is again facing job displacement. The quality of work by the robotic devices has been so high that it can now be checked by an automated system. The company has offered to retrain Peter as a technician to service and maintain the robotic devices.

Peter knows very little about technology, and feels he is too old to learn a new trade. He wants to keep working for the company, but as a carpenterthe job he has taken pride in most of his life. He also has his financial responsibilities.

Faced with what he wants to do and the fear that he cannot handle the technician’s position he does not know what to do.

1. a What advantages do automated workplaces have over non automated?

b What disadvantages are there for people who work in an automated environment?

2. a How can companies justify the high set-up costs of automating?

b Why did the Furniture Factory feel they had no choice other than to use robots?

3. a Which types of jobs are most likely to be taken over by robotic devices?

b Which are the least likely?

4. a Why might unions argue against automation?

b Why might unions argue for automation?

5. What is the difference between being made redundant (laid off) and experiencing job displacement?

6. Describe how you think Peter would have thought:

a When he was first displaced – refer to deskilling, lack of job satisfaction, and the morale effect of having to work alongside a machine.

b With his latest job displacement – where he has to undergo retraining and will end up as virtually the servant of the robotic devices.



7. What do you think Peter should do? Why?

The next extract describes a case of redundancy. Read it and answer the questions that follow: The principal of Bunyip Primary School is faced with a dilemma, he has to make a difficult choice. Much of the clerical work in the school is carried out by two teacher’s aides. Maud is a supporting parent in her late thirties with children at the school. While she has worked at the school for a number of years, and is very capable, she is not prepared to use computers, instead relying on her electric typewriter and calculator. Jenny is a single woman of twenty two who gained her position after two years unemployment. While unemployed she completed several Job Skill courses in computers. Both are valuable workers and are cheerful and efficient.

The school has been offered a new computer-based administrative package developed by the education department. This will greatly improve the efficiency of the running of the school, lightening the load on existing staff and in turn improving the general quality of education given to the students. The package is optional; only those schools that choose to take it up need do so. A condition of the package is that the number of teacher aides must be halved, in this case from two to one.

The principal is keen to adopt the package which will bring greater effectiveness to the running of the school, but is unable to offer an alternative position to the redundant aide. Neither would be likely to find other employment. Maud depends on her salary to help support her family, while Jenny hopes to get married soon if she can afford it.

8. a What options does the principal face?

b Explain why he would not wish to choose either option.

9. Decide (with reasons) which choice you think will be best for the school and the people involved. (If you cannot come to a decision attempt to explain why you could not.)

10. Join a group of 3 to 4 other students and discuss the dilemma, each person to put forward their view and reasons. As a class list the various reasons and decisions reached.

Handling data is crucial, so next we will see how this can best be done


6 computer systems

Technology

complete computer system

typical computer system

new system

system parts

current system

humancomputer interaction

dr foote gain

home dr foote