Understanding the fundamentals of the technical components of information systems
is an essential first step in understanding the strategic role of information systems in
modern organizations. An obvious technical component is the physical computing
machinery, known as hardware. In this chapter we will see that hardware is more than
just the computer itself—it includes a variety of related technologies involved with
getting data into and out of the computer. The information in this chapter not only
will provide a foundation for understanding the rest of the text, but it also will help
you make informed decisions about personal and professional computing technology.
CHAPTER PREVIEW
3.1 The Significance of Hardware
3.2 The Central Processing Unit
3.3 Computer Memory
3.4 Computer Hierarchy
3.5 Input Technologies
3.6 Output Technologies
3.7 Strategic Hardware Issues
CHAPTER OUTLINE
1. Identify the major hardware components of a
computer system.
2. Describe the design and functioning of the central
processing unit.
3. Discuss the relationships between microprocessor
component designs and performance.
4. Describe the main types of primary and secondary
storage.
5. Distinguish between primary and secondary storage
along the dimensions of speed, cost, and capacity.
6. Define enterprise storage and describe the various
types of enterprise storage.
7. Describe the hierarchy of computers according to
power and their respective roles.
8. Differentiate the various types of input and output
technologies and their uses.
9. Describe what multimedia systems are and what
technologies they use.
10. Discuss strategic issues that link hardware design and
innovation to competitive business strategy.
LEARNING OBJECTIVES
3
COMPUTER HARDWARE
COMBINING MAINFRAMES AND E-COMMERCE
ABF Freight System, Inc.® (ABF) is a less-than-truckload (LTL) transportation company
in Fort Smith, Arkansas. LTL carriers such as ABF® and competitor Roadway Express
fill the niche between parcel carriers like Federal Express and full-truckload
carriers like JB Hunt that specialize in huge shipments. Competition also comes
from virtual companies, such as Freightquote.com and Transportation.com.
LTL carriers ship general commodities. Their core customers are businesses,
not consumers. These carriers calculate prices for each shipment using
variables such as weight, volume, distance, and the number of boxes. LTL carriers
typically offer discounts on most shipments, often making custom quotes
to win jobs. ABF® wanted to leverage the Internet to be able to keep up with
rapidly changing business conditions and to offer an accurate price to customers
without reinventing mainframe applications.
ABF Freight System, Inc.® (ABF) built an e-commerce infrastructure that runs on its
IBM S/390 mainframe. The same mainframe applications that ABF® had used to calculate
pricing, trace shipments, schedule routes, and review freight bills are now accessible
via the e-commerce Web site, the intranet, devices enabled by Wireless Application
Protocol (WAP), imaging software, and an interactive voice response (IVR) system.
At ABF®’s self-service Web site, dubbed eCenter®, customers map routes, trace
shipments, schedule a pickup, and create a bill of lading (the formal document required
for shipments). ABF® customers generate price quotes that include discounts,
view images of shipment documents, and review damage claim status. The eCenter®
also provides predictive e-mail alerts that offer progress reports of a shipment in transit
and alert the customer if the shipment will be late.
eCenter® has several innovative features. The Shipment Planner™ displays pending
shipments in a calendar format. A feature called Transparent Links lets ABF® customers
incorporate shipping data from ABF®’s mainframe into their own systems via XML.
ABF® Anywhere lets users communicate with ABF with a Palm handheld device or mobile
phone equipped with Internet access. Dynamic Rerouting lets customers change the
destination of an in-transit shipment or recall a shipment by accessing ABF®’s Web site.
ABF® then e-mails a confirmation of the new destination and revised charges.
Drivers en route check in at ABF service stations in 311 locations until they arrive
at the destination terminal, where the shipment is scheduled for delivery to the consignee’s
address. At each checkpoint, drivers submit documents (each with a bar
code), such as the bill of lading, for scanning. The scanned images are uploaded via
FTP over the wide area network (WAN) to a database on the mainframe located at
company headquarters. This process creates a visual record.
ABF®’s e-commerce infrastructure has more than 23,000 registered users from more
than 17,000 ABF customers. These customers generate more than 70 percent of
ABF®’s annual revenue and shipment volume. ABF®’s new e-commerce infrastructure
enhances customer service, and has created a new business line and opened new
markets for the company. The infrastructure also has revamped virtually everyone’s
job at ABF, from how a regional vice president builds customer loyalty to how a customer
service representative spends the day.
Source: Network World (February 26, 2001); abfs.com.
T h e B u s i n e s s P r o b l e m
T h e I T S o l u t i o n
T h e R e s u l t s
abfs.com
ABF Freight
Systems® trucks
can be tracked via
a new e-commerce
infrastructure.
Selecting the right IT infrastructure for any business is a complex decision. Such a decision
often entails “out of the box” thinking—that is, imagining how business
processes could be ideally configured and supported—rather than incremental improvement
of an outdated process. Indeed, ABF®’s e-commerce infrastructure is an
outstanding example of old-to-new economy transformation. The company had a
tremendous amount riding on its IT decision. In the LTL industry, superior system
performance translates very quickly into customer satisfaction.
The same basic issues confront all organizations that use computing technology.
Such decisions about information technology usually focus on three interrelated factors:
capability (power and appropriateness for the task), speed, and cost. A computer’s
hardware design drives all three factors, and all three factors are interrelated
and are much more complex than you might imagine.
The incredible rate of innovation in the computer industry further complicates
IT decisions. The ABF® executives in this case had a difficult decision to make, because
ABF® was already a going concern with an information technology already in
place. Computer technologies can become obsolete much more quickly than other
organizational technologies. Yet, regardless of industry, computer hardware is
essential to survival, and the most modern hardware may be essential to sustaining
advantage over competitors. Evaluating new hardware options and figuring out how
to integrate them with existing (legacy) systems is an ongoing responsibility in most
organizations.
Finally, almost any time an organization makes major changes in its computer infrastructure,
much of its software needs to be rewritten to run on the hardware’s new
operating system. In some cases, all the data that a company has accumulated may
have to be put into a different format. Personnel may have to be retrained on the new
computers. These are very lengthy and expensive undertakings, often dwarfing basic
hardware acquisition costs by tenfold. Therefore, computer hardware choices are generally
made only after careful study. Many of the issues in such decision making involve
employees from all functional areas and are the topics of this chapter.
Most businesspeople rightly suspect that knowing how to use computer technology is
more important to their personal productivity and their firm’s competitive advantage
than knowing the technical details of how the technology functions. But some basic
understanding of computer hardware design and function is essential because organizations
frequently must assess their competitive advantage in terms of computing capability.
Important decisions about computing capability have to be made, and to a
large degree these decisions turn on an understanding of hardware design. In this
chapter you will learn the basics of hardware design and understand the sources of
this capability.
Our objective is to demonstrate how computers input, process, output, and store
information. We will also look at the hierarchy of computer hardware, from the super
computer down to the handheld microcomputer and even some smaller technologies.
Finally we will consider the dynamics of computer hardware innovation and the effects
it has on organizational decision making.
An important benefit from reading this chapter will be that not only will you better
understand the computer hardware decisions in your organization, but also your
personal computing decisions will be much better informed. Many of the design
W h a t W e L e a r n e d f r o m T h i s C a s e
3.1 THE SIGNIFICANCE OF HARDWARE
principles presented here apply to any size computer, as do the dynamics of innovation
and cost that affect personal as well as corporate hardware decisions.
As we noted in Chapter 1, computer-based information systems (CBISs) are composed
of hardware, software, databases, telecommunications, procedures, and people.
The components are organized to input, process, and output data and information.
Chapter 3 focuses on the hardware component of the CBIS. Hardware refers to the
physical equipment used for the input, processing, output, and storage activities of a
computer system. It consists of the following:
• Central processing unit (CPU)
• Memory (primary and secondary storage)
• Input technologies
• Output technologies
• Communication technologies
The first four of these components are discussed in the following sections. Communication
technologies is the subject of Chapter 7.
The central processing unit (CPU) performs the actual computation or “number
crunching” inside any computer. The CPU is a microprocessor (for example, a Pentium
4 by Intel) made up of millions of microscopic transistors embedded in
a circuit on a silicon wafer or chip. (Hence, microprocessors are commonly referred
to as chips.) Examples of specific microprocessors are listed in Table 3.1.
As shown in Figure 3.1 (on page 58), the microprocessor has different parts,
which perform different functions. The control unit sequentially accesses program
instructions, decodes them, and controls the flow of data to and from the ALU, the
registers, the caches, primary storage, secondary storage, and various output devices.
The arithmetic-logic unit (ALU) performs the mathematic calculations and makes
logical comparisons. The registers are high-speed storage areas that store very small
amounts of data and instructions for short periods of time. (For a more technical
overview of the components of modern chips, see Modern Chip Components on the
Web site.)
H o w t h e C P U W o r k s
The CPU, on a basic level, operates like a tiny factory. Inputs come in and are stored
until needed, at which point they are retrieved and processed and the output is
Section 3.2 The Central Processing Unit 57
Table 3.1 Examples of Microprocessors
Name Manufacturer Word Length Clock Speed (MHz) Application
Pentium III Intel 32 1000 PCs and workstations
Pentium 4 Intel 64 2000 PCs and workstations
PowerPC Motorola, IBM, Apple 32 1000 High-end PCs and workstations
Alpha Compaq 64 1500 PCs and workstations
Athlon Advanced Micro Devices 32 1000 PCs and workstations
3.2 THE CENTRAL PROCESSING UNIT
Intel’s Pentium 4
microprocessor.
stored and then delivered somewhere. Figure 3.2 illustrates this process, which works
as follows:
• The inputs are data and brief instructions about what to do with the data. These instructions
come from software in other parts of the computer. Data might be entered
by the user through the keyboard, for example, or read from a data file in
another part of the computer. The inputs are stored in registers until they are sent
to the next step in the processing.
• Data and instructions travel in the chip via electrical pathways called buses. The
size of the bus—analogous to the width of a highway—determines how much information
can flow at any time.
• The control unit directs the flow of data and instructions within the chip.
• The arithmetic-logic unit (ALU) receives the data and instructions from the registers
and makes the desired computation. These data and instructions have been
translated into binary form, that is, only 0s and 1s. The CPU can process only binary
data.
• The data in their original form and the instructions are sent to storage registers and
then are sent back to a storage place outside the chip, such as the computer’s hard
58 Chapter 3 Computer Hardware
Figure 3.1 Parts of a
microprocessor.
The Microprocessor
Registers
Control
unit
Arithmeticlogic
unit
Primary storage
(main memory)
Secondary storage
Input Output
Communication
devices
The Microprocessor
Control unit
Fetch
Arithmetic-logic unit
Primary storage
(main memory)
1 4 Store
2 Decode 3 Execute
Inputs from
software
Instruction
instruction instruction instruction
Registers Results
Figure 3.2 How the
CPU works.
drive (discussed below). Meanwhile, the transformed data go to another register
and then on to other parts of the computer (to the monitor for display, or to be
stored, for example).
(For a more technical overview of CPU operations, see CPU Operations on the Web
site.)
This cycle of processing, known as a machine instruction cycle, occurs millions of
times per second or more. It is faster or slower, depending on the following four factors
of chip design:
1. The preset speed of the clock that times all chip activities, measured in megahertz
(MHz), millions of cycles per second, and gigahertz (GHz), billions of cycles per
second. The faster the clock speed, the faster the chip. (For example, all other factors
being equal, a 1.0 GHz chip is twice as fast as a 500 MHz chip.)
2. The word length, which is the number of bits (0s and 1s) that can be processed by
the CPU at any one time. The majority of current chips handle 32-bit word lengths,
and the Pentium 4 is designed to handle 64-bit word lengths. Therefore, the Pentium
4 chip will process 64 bits of data in one machine cycle. The larger the word
length, the faster the chip.
3. The bus width. The wider the bus (the physical paths down which the data and instructions
travel as electrical impulses), the more data can be moved and the faster
the processing. A processor’s bus bandwidth is the product of the width of its bus
(measured in bits) times the frequency at which the bus transfers data (measured
in megahertz). For example, Intel’s Pentium 4 processor uses a 64-bit bus that runs
at 400 MHz. That gives it a peak bandwidth of 3.2 gigabits per second.
4. The physical design of the chip. Back to our “tiny factory” analogy, if the “factory”
is very compact and efficiently laid out, then “materials” (data and instructions) do
not have far to travel while being stored or processed. We also want to pack as
many “machines” (transistors) into the factory as possible. The distance between
transistors is known as line width. Historically, line width has been expressed in microns
(millionths of a meter), but as technology has advanced, it has become more
convenient to express line width in nanometers (billionths of a meter). Currently,
most CPUs are designed with 180-nanometer technology (0.18 microns), but chip
manufacturers are moving to 130-nanometer technology (0.13 microns). The
smaller the line width, the more transistors can be packed onto a chip, and the
faster the chip.
These four factors make it difficult to compare the speeds of different processors.
As a result, Intel and other chip manufacturers have developed a number of benchmarks
to compare processor speeds. (For a discussion of these benchmarks, see the
section on Processor Benchmarks on the Web site.)
A d v a n c e s i n M i c r o p r o c e s s o r D e s i g n
Innovations in chip designs are coming at a faster and faster rate, as described by
Moore’s Law. Gordon Moore, an Intel Corporation co-founder, predicted in 1965 that
microprocessor complexity would double approximately every two years. As shown in
Figure 3.3 (on page 60), his prediction was amazingly accurate.
The advances predicted from Moore’s Law come mainly from the following
changes:
• Increasing miniaturization of transistors.
• Making the physical layout of the chip’s components as compact and efficient as
possible (decreasing line width).
Section 3.2 The Central Processing Unit 59
60 Chapter 3 Computer Hardware
10M
100M
1M
100K
10K
1978
Transistor Counts for Intel Processors
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 2001 2002
Itanium II
(McKinley)
(214 million)
Pentium 4
(42 million)
Pentium III
(9.5 million)
Pentium II
(7.5 million)
Pentium Pro (5.5 million)
Pentium (3.1 million)
486 (1.2 million)
386 (275,000)
286 (134,000)
8086 (29,000)
Figure 3.3 Moore’s Law as it relates to transistor counts in Intel microprocessors.
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987
Clock speed: 2 MHz
MIPS: 0.64
Transistors: 6,000
Internal bus: 8-bit
Introduced: April 1974
Clock speed: 5 MHz
MIPS: 0.33
Transistors: 29,000
Internal bus: 16-bit
Introduced: June 1979
Clock speed: 16 MHz
MIPS: 6
Transistors: 275,000
Internal bus: 32-bit
Introduced: October 1985
Clock speed: 16 MHz
MIPS: 2.5
Transistors: 275,000
Internal bus: 32-bit
Introduced: June 1988
Clock speed: 5 MHz
MIPS: 0.33
Transistors: 29,000
Internal bus: 16-bit
Introduced: June 1978
Clock speed: 8 MHz
MIPS: 1.2
Transistors: 134,000
Internal bus: 16-bit
Introduced: February 1982
IBM PC
IBM PC AT Compaq Deskpro 386
8080 8088 80386DX 80386SX
8086 80286
Figure 3.4 The lineage of Intel microprocessors. [Diagram and content displayed from 1974–1993 reprinted from PC Magazine
(April 27, 1993), with permission. Copyright (c) 1993, ZD, Inc. All rights reserved. Diagram and content displayed for
dates beyond 1993 based on data from Intel, added by the authors to show Intel microprocessing trends through 2001.]
• Using materials for the chip that improve the conductivity (flow) of electricity. The
traditional silicon is a semiconductor of electricity—electrons can flow through it at
a certain rate. New materials such as gallium arsenide and silicon germanium allow
even faster electron travel and some additional benefits, although they are more expensive
to manufacture than silicon chips.
• Targeting the amount of basic instructions programmed into the chip. There are
four broad categories of microprocessor architecture: complex instruction set computing
(CISC), reduced instruction set computing (RISC), very long instruction word
(VLIW), and the newest category, explicitly parallel instruction computing (EPIC).
Most chips are designated as CISC and have very comprehensive instructions, directing
every aspect of chip functioning. RISC chips eliminate rarely used instructions.
Computers that use RISC chips (for example, a workstation devoted to
high-speed mathematical computation) rely on their software to contain the special
instructions. VLIW architectures reduce the number of instructions on a chip by
lengthening each instruction. With EPIC architectures, the processor can execute
certain program instructions in parallel. Intel’s Pentium 4 is the first implementation
of EPIC architecture. (For a more technical discussion of these architectures, see
Microprocessor Architectures on the Web site.)
In addition to increased speeds and performance, Moore’s Law has had an impact
on costs. For example, in 1998, a personal computer with a 16 MHz Intel 80386 chip,
one megabyte of RAM (discussed later in this chapter), a 40-megabyte hard disk (discussed
later in this chapter), and a DOS 3.31 operating system (discussed in Chapter
4), cost $5,200. In 2002, a personal computer with a 2 GHz Intel Pentium 4 chip, 512
megabytes of RAM, an 80-gigabyte hard disk, and the Windows XP operating system,
cost less than $1,000 (without the monitor).
Although organizations certainly benefit from microprocessors that are faster,
they also benefit from chips that are less powerful but can be made very small and inexpensive.
Microcontrollers are chips that are embedded in countless products and
technologies, from cellular telephones to toys to automobile sensors. Microprocessors
and microcontrollers are similar except that microcontrollers usually cost less and
Section 3.2 The Central Processing Unit 61
1988 1989 1990 1991 1992 1993
Clock speed: 66 MHz
MIPS: 112
Transistors: 3.1 million
Internal bus: 64-bit
Introduced: May 1993
Clock speed: 20 MHz
MIPS: 16.5
Transistors: 1.185 million
Internal bus: 32-bit
Introduced: April 1991
Clock speed: 50 MHz
MIPS: 40
Transistors: 1.2 million
Internal bus: 32-bit
Introduced: March 1992
Clock speed: 25 MHz
MIPS: 20
Transistors: 1.2 million
Internal bus: 32-bit
Introduced: April 1989
1995 1997 2000
Clock speed:
350-400 MHz
Transistors: 7.5 million
Introduced: May 1997
Clock speed:
150-200 MHz
Transistors: 5.5 million
Introduced: Fall 1995
1999
Clock speed:
450-550 MHz
Transistors: 9.5 million
ALR PowerCache 4 Introduced: January 1999
486SX Pentium
486DX 486DX2
Pentium II
Clock speed:
1 GHz to 2 GHz
Transistors: 42 million
Introduced: late in 2000
Pentium 4
Pentium Pro Pentium III
work in less-demanding applications. Thus, the scientific advances in CPU design affect
many organizations on the product and service side, not just on the internal CBIS
side.
Figure 3.4 (on pages 60–61) illustrates the historical advancement of Intel microprocessors.
New types of chips continue to be produced. (For a discussion of advanced
chip technologies, see Advanced Chip Technologies on the Web site.)
The amount and type of memory that a computer possesses has a great deal to do with
its general utility, often affecting the type of program it can run and the work it can
do, its speed, and both the cost of the machine and the cost of processing data. There
are two basic categories of computer memory. The first is primary storage, so named
because small amounts of data and information that will be immediately used by the
CPU are stored there. The second is secondary storage, where much larger amounts of
data and information (an entire software program, for example) are stored for extended
periods of time.
M e m o r y C a p a c i t y
As already noted, CPUs process only 0s and 1s. All data are translated through computer
languages (covered in the next chapter) into series of these binary digits, or bits.
A particular combination of bits represents a certain alphanumeric character or simple
mathematical operation. Eight bits are needed to represent any one of these characters.
This 8-bit string is known as a byte. The storage capacity of a computer is
measured in bytes. (Bits are used as units of measure typically only for telecommunications
capacity, as in how many million bits per second can be sent through a particular
medium.) The hierarchy of byte memory capacity is as follows:
• Kilobyte. Kilo means one thousand, so a kilobyte (KB) is approximately one thousand
bytes. Actually, a kilobyte is 1,024 bytes (210 bytes).
• Megabyte. Mega means one million, so a megabyte (MB) is approximately one million
bytes (1,048,576 bytes, or 1,024 1,024, to be exact). Most personal computers
have hundreds of megabytes of RAM memory (a type of primary storage, discussed
in a later section).
• Gigabyte. Giga means one billion; a gigabyte (GB) is actually 1,073,741,824 bytes
(1,024 1,024 1,024 bytes). The storage capacity of a hard drive (a type of
secondary storage, discussed shortly) in modern personal computers is often many
gigabytes.
• Terabyte. One trillion bytes (actually, 1,078,036,791,296 bytes) is a terabyte.
To get a feel for these amounts, consider the following examples. If your computer
has 256 MB of RAM (a type of primary storage), it can store 268,435,456 bytes
B e f o r e y o u g o o n . . .
1. Briefly describe how a microprocessor functions.
2. What factors determine the speed of the microprocessor?
3. How are microprocessor designs advancing?
62 Chapter 3 Computer Hardware
3.3 COMPUTER MEMORY
of data. A written word might, on average, contain 6 bytes, so this translates to approximately
44.8 million words. If your computer has 20 GB of storage capacity on
the hard drive (a type of secondary storage) and the average page of text has about
2,000 bytes, your hard drive could store some 10 million pages of text.
P r i m a r y S t o r a g e
Primary storage, or main memory, as it is sometimes called, stores for very brief periods
of time three types of information: data to be processed by the CPU, instructions
for the CPU as to how to process the data, and operating system programs that manage
various aspects of the computer’s operation. Primary storage takes place in chips
mounted on the computer’s main circuit board (the motherboard), located
as close as physically possible to the CPU chip. (See Figure 3.5.)
As with the CPU, all the data and instructions in primary storage have
been translated into binary code.
There are four main types of primary storage: (1) register, (2) random
access memory (RAM), (3) cache memory, and (4) read-only
memory (ROM). To understand their purpose, consider the following
analogy: You keep a Swiss Army knife handy in your pocket for minor
repairs around the house. You have a toolbox with an assortment of
tools in the kitchen cabinet for bigger jobs. Finally, in the garage you
have your large collection of tools. The amount and type of tools you
need, how often you need them, and whether you will use them immediately
determines how and where you store them. In addition, one type
of storage area—like a fireproof wall safe—must be completely safe, so that its contents
cannot be lost. The logic of primary storage in the computer is just like the logic
of storing things in your house. That which will be used immediately gets stored in
very small amounts as close to the CPU as possible. Remember, as with CPU chip design,
the shorter the distance the electrical impulses (data) have to travel, the faster
they can be transported and processed. That which requires special protection will be
stored in an exceptionally secure manner. The four types of primary storage, which
follow this logic, are described next.
Registers. As indicated earlier in the chapter, registers are part of the CPU. They
have the least capacity, storing extremely limited amounts of instructions and data
only immediately before and after processing. This is analogous to your pocket in the
Swiss Army knife example.
Random access memory. Random access memory (RAM) is analogous to the
kitchen toolbox. It stores more information than the registers (your pocket) and is farther
away from the CPU, but it stores less than secondary storage (the garage) and is
much closer to the CPU than is secondary storage. When you start most software programs
on your computer, the entire program is brought from secondary storage into
RAM. As you use the program, small parts of the program’s instructions and data are
sent into the registers and then to the CPU. Again, getting the data and instructions as
close to the CPU as possible is key to the computer’s speed, as is the fact that the
RAM is a type of microprocessor chip. As we shall discuss later, the chip is much
faster (and more costly) than are secondary storage devices.
RAM is temporary and volatile; that is, RAM chips lose their contents if the current
is lost or turned off (as in a power surge, brownout, or electrical noise generated
by lightning or nearby machines). RAM chips are located directly on the computer’s
main circuit board or in other chips located on peripheral cards that plug into the
main circuit board.
Section 3.3 Computer Memory 63
a
b
d
c
Figure 3.5 Internal
workings of a common personal
computer. (a) Hard
disk drive; (b) floppy disk
drive; (c) RAM; (d) CPU
board with fan.
The two main types of RAM are dynamic RAM (DRAM) and static RAM
(SRAM). DRAM memory chips offer the greatest capacities and the lowest cost per
bit, but are relatively slow. SRAM costs more than DRAM but has a higher level of
performance, making SRAM the preferred choice for performance-sensitive applications,
including the external L2 and L3 caches (discussed next) that speed up microprocessor
performance.
Cache memory. Cache memory is a type of high-speed memory that a processor can
access more rapidly than main memory (RAM). It augments RAM in the following
way: Many modern computer applications (Microsoft XP, for example) are very complex
and have huge numbers of instructions. It takes considerable RAM capacity
(usually a minimum of 128 megabytes) to store the entire instruction set. Or you may
be using an application that exceeds your RAM. In either case, your processor must
go to secondary storage (similar to a lengthy trip to the garage) to retrieve the necessary
instructions. To alleviate this problem, software is often written in smaller blocks
of instructions. As needed, these blocks can be brought from secondary storage into
RAM. This process is still slow, however.
Cache memory is a place closer to the CPU where the computer can temporarily
store those blocks of instructions used most often. Blocks used less often remain in
RAM until they are transferred to cache; blocks used infrequently stay stored in secondary
storage. Cache memory is faster than RAM because the instructions travel a
shorter distance to the CPU. In our tool analogy, cache memory might represent an
additional box with a selected set of needed tools from the kitchen toolbox and the
garage.
There are two types of cache memory in the majority of computer systems—
Level 1 (L1) cache is located in the processor, and Level 2 (L2) cache is located on the
motherboard but not actually in the processor. L1 cache is smaller and faster than L2
cache. Chip manufacturers are now designing chips with L1 cache and L2 cache in the
processor and Level 3 (L3) cache on the motherboard.
Read-only memory. In our previous example, we alluded to the need for greater security
when storing certain types of critical data or instructions. (This was represented
by the wall safe.) Most people who use computers have lost precious data at one time
or another due to a computer “crash” or a power failure. What is usually lost is whatever
is in RAM, cache, or the registers at the time. This loss occurs because these
types of memory are volatile. Whatever information they may contain is lost when
there is no electricity flowing through them. The cautious computer
user frequently saves his or her data to nonvolatile memory (secondary
storage). In addition, most modern software applications have
autosave functions. Programs stored in secondary storage, even
though they are temporarily copied into RAM when used, remain intact
because only the copy is lost and not the original.
Read-only memory (ROM) is the place (a type of chip) where
certain critical instructions are safeguarded. ROM is nonvolatile
and retains these instructions when the power to the computer is
turned off. The read-only designation means that these instructions
can be read only by the computer and cannot be changed by the user. An example
of ROM instructions are those needed to start or “boot” the computer once it has
been shut off. There are variants of ROM chips that can be programmed (PROM),
and some that can be erased and rewritten (EPROM). These are relatively rare in
mainstream organizational computing, but are often incorporated into other specialized
technologies such as video games (PROM) or robotic manufacturing
(EPROM).
64 Chapter 3 Computer Hardware
“Oops! I just deleted all
your files. Can you repeat
everything you’ve ever
told me?”
Another form of rewritable ROM storage is called flash memory. This technology
can be built into a system or installed on a personal computer card (known as a flash
card). These cards, though they have limited capacity, are compact, portable, and require
little energy to read and write. Flash memory via flash cards is very popular for
small portable technologies such as cellular telephones, digital cameras, handheld
computers, and other consumer products.
S e c o n d a r y S t o r a g e
Secondary storage is designed to store very large amounts of data for extended periods
of time. Secondary storage can have memory capacity of several terabytes or
more and only small portions of that data are placed in primary storage at any one
time. Secondary storage has the following characteristics:
• It is nonvolatile.
• It takes much more time to retrieve data from secondary storage than it does from
RAM because of the electromechanical nature of secondary storage devices.
• It is much more cost effective than primary storage (see Figure 3.6).
• It can take place on a variety of media, each with its own technology, as discussed
next.
• The overall trends in secondary storage are toward more direct-access methods,
higher capacity with lower costs, and increased portability.
Magnetic media. Magnetic tape is kept on a large open reel or in a smaller cartridge
or cassette. Although this is an old technology, it remains popular because it is the
cheapest storage medium and can handle enormous amounts of data. The downside is
that it is the slowest for retrieval of data, because all the data are placed on the tape
sequentially. Sequential access means that the system might have to run through the
majority of the tape, for example, before it comes to the desired piece of data. Magnetic
media store information by giving tiny particles of iron oxide embedded on the
tape a positive or negative polarization. Recall that all data that a computer understands
are binary. The positive or negative polarization of the particles corresponds to
a 0 or a 1.
Magnetic tape storage often is used for information that an organization must
maintain, but uses rarely or does not need immediate access to. Industries with huge
numbers of files (e.g., insurance companies), use magnetic tape systems. Modern versions
of magnetic tape systems use cartridges and often a robotic system that selects
and loads the appropriate cartridge automatically. There are also some tape systems,
Section 3.3 Computer Memory 65
Figure 3.6 Primary
memory compared to secondary
storage.
Cost Speed
Semiconductor or
primary memory
Secondary
storage
Register
Cache
RAM
ROM
Magnetic
disk Optical
Size
Magnetic
tape
like digital audio tapes (DAT), for smaller applications such as storing copies of all
the contents of a personal computer’s secondary storage (“backing up” the storage).
Magnetic disks come in a variety of styles and are popular because they allow
much more rapid access to the data than does magnetic tape. Magnetic disks, called
hard disks or fixed disk drives, are the most commonly used mass storage devices because
of their low cost, high speed, and large storage capacity. Fixed disk drives read
from, and write to, stacks of rotating magnetic disk platters mounted in rigid enclosures
and sealed against environmental or atmospheric contamination. These disks
are permanently mounted in a unit that may be internal or external to the computer.
All disk drives (including removable disk modules, floppy disk drives, and optical
drives) are called hard drives and store data on platters divided into concentric tracks.
Each track is divided further into segments called sectors. To access a given sector, a
read/write head pivots across the rotating disks to locate the right track, calculated
from an index table, and the head then waits as the disk rotates until the right sector is
underneath it. (For a more technical discussion of hard disk drives, see Hard Disk
Drives on the Web site.)
Every piece of data has an address attached to it, corresponding to a particular
track and sector. Any piece of desired data can be retrieved in a nonsequential manner,
by direct access (which is why hard disk drives are sometimes called direct access
storage devices). The read/write heads use the data’s address to quickly find and read
the data. (See Figure 3.7.) Unlike magnetic tape, the system does not have to read
through all the data to find what it wants.
The read/write heads are attached to arms that hover over the disks, moving in
and out (see Figure 3.8). They read the data when positioned over the correct track
and when the correct sector spins by. Because the head floats just above the surface of
the disk (less than 25 microns), any bit of dust or contamination can disrupt the device.
When this happens, it is called a disk crash and usually results in catastrophic loss
of data. For this reason, hard drives are hermetically sealed when manufactured.
A modern personal computer typically has many gigabytes (some more than 100
gigabytes) of storage capacity in its internal hard drive. Data access is very fast, measured
in milliseconds. For these reasons, hard disk drives are popular and common.
Because they are somewhat susceptible to mechanical failure, and because users may
need to take all their hard drive’s contents to another location, many users like to
back up their hard drive’s contents with a portable hard disk drive system, such as
Iomega’s Jaz.
Disk drive interfaces. To take advantage of the new, faster technologies, disk
drive interfaces must also be faster. Most PCs and workstations use one of two
66 Chapter 3 Computer Hardware
Figure 3.7 Magnetic
disk drive.
Read/write head
Magnetic disk
high-performance disk interface standards: Enhanced Integrated Drive Electronics
(EIDE) or Small Computer Systems Interface (SCSI). EIDE offers good performance,
is inexpensive, and supports up to four disks, tapes, or CD-ROM drives. SCSI drives are
more expensive than EIDE drives, but they offer a faster interface and support more
devices. SCSI interfaces are therefore used for graphics workstations, server-based storage,
and large databases. (For discussions of other interfaces, including fibre channel,
firewire, Infiniband, and the universal serial bus, see Other Interfaces on the Web site.)
Magnetic diskettes. Magnetic diskettes, or floppy disks as they are commonly called,
function similarly to hard drives, but with some key differences. The most obvious is
that they are not rigid, but are made out of flexible Mylar. They are much slower than
hard drives. They have much less capacity, ranging from 1.44 megabytes for a standard
high-density disk to 250 megabytes for a disk formatted for a Zip drive (on which
the data are compressed). Further, although they are individually inexpensive, floppy
disks are less cost-efficient than hard drive storage. However, the big advantage of
floppy disks has been that they are portable. Hard disk drives are usually permanently
installed in a computer, but the small, removable diskette (installed in its thin plastic
housing) can fit into a shirt pocket and can be easily mailed.
Optical storage devices. Unlike magnetic media, optical storage devices do not store
data via magnetism. As shown in Figure 3.9, to record information on these devices, a
pinpoint laser beam is used to burn tiny holes into the surface of a reflective plastic
platter (such as a compact disk). When the information is read, another laser, installed
in the optical disk drive of the computer (such as a compact disk drive), shines on the
surface of the disk. If light is reflected, that corresponds to one binary state. If the
light shines on one of the holes burned by the recording laser, there is no reflection
Section 3.3 Computer Memory 67
Figure 3.8 Read/write
heads.
Read/write heads
“fly” over disk surfaces
11 disks
20 recording surfaces
7,200 RPMs
Write operation
uses high-powered
laser beam
Read operation
uses low-powered
laser beam
Figure 3.9 Optical storage
device.
and the other binary state is read. Compared to magnetic media, optical disk drives
are slower than magnetic hard drives. On the other hand, they are much less susceptible
to damage from contamination and are also less fragile.
In addition, optical disks can store much more information, both on a routine
basis and also when combined into storage systems. Optical disk storage systems can
be used for large-capacity data storage. These technologies, known as optical jukeboxes,
store many disks and operate much like the automated phonograph record
changers for which they are named.
Types of optical disks include compact disk read-only memory (CD-ROM), digital
video disk (DVD), and fluorescent multilayer disk (FMD-ROM).
Compact disk, read-only memory (CD-ROM) storage devices feature high capacity,
low cost, and high durability. However, because it is a read-only medium, the CDROM
can be only read and not written on. Compact disk, rewritable (CD-RW) adds
rewritability to the recordable compact disk market, which previously had offered
only write-once CD-ROM technology.
The digital video disk (DVD) is a five-inch disk with the capacity to store about
135 minutes of digital video. DVD provides sharp detail, true color, no flicker, and
no snow. Sound is recorded in digital Dolby, creating clear “surround-sound” effects.
DVDs have advantages over videocassettes, including better quality, smaller
size (meaning they occupy less shelf space), and lower duplicating costs. DVDs can
also perform as computer storage disks, providing storage capabilities of 17 gigabytes.
DVD players can read current CD-ROMs, but current CD-ROM players cannot
read DVDs. The access speed of a DVD drive is faster than a typical CD-ROM
drive.
A new optical storage technology called fluorescent multilayer disk (FMD-ROM)
greatly increases storage capacity. The idea of using multiple layers on an optical disk
is not new, as DVDs currently support two layers. However, by using a new fluorescent-
based optical system, FMDs can support 20 layers or more. FMDs are clear
disks; in the layers are fluorescent materials that give off light. The presence or absence
of these materials tells the drive whether there is information there or not. All
layers of an FMD can be read in parallel, thereby increasing the data transfer rate.
Memory cards. PC memory cards are credit-card-size devices that can be installed in
an adapter or slot in many personal computers. The PC memory card functions as if it
were a fixed hard disk drive. The cost per megabyte of storage is greater than for traditional
hard disk storage, but the cards do have advantages. They are less failureprone
than hard disks, are portable, and are relatively easy to use. Software
manufacturers often store the instructions for their programs on a memory card for
use with laptop computers. The Personal Computer Memory Card International Association
(PCMCIA) is a group of computer manufacturers who are creating standards
for these memory cards.
Expandable storage. Expandable storage devices are removable disk cartridges. The
storage capacity ranges from 100 megabytes to several gigabytes per cartridge, and
the access speed is similar to that of an internal hard drive. Although more expensive
than internal hard drives, expandable storage devices combine hard disk storage capacity
and diskette portability. Expandable storage devices are ideal for backup of the
internal hard drive, as they can hold more than 80 times as much data and operate five
times faster than existing floppy diskette drives.
Advanced storage technologies. (For an overview of advanced storage technologies,
see the section on Advanced Storage Technologies on the Web site.)
68 Chapter 3 Computer Hardware
E n t e r p r i s e S t o r a g e S y s t e m s
The amount of digital information is doubling every two years. As a result, many companies
are employing enterprise storage systems.
An enterprise storage system is an independent, external system with intelligence
that includes two or more storage devices. These systems are an alternative to allowing
each host or server to manage its own storage devices directly. Enterprise storage systems
provide large amounts of storage, high-performance data transfer, a high degree of
availability, protection against data loss, and sophisticated management tools. (For a
technical discussion of enterprise storage systems, see the section on Enterprise Storage
Systems on the Web site.)
There are three major types of enterprise storage subsystems: redundant arrays of
independent disks (RAIDs), storage area networks (SANs), and network-attached
storage (NAS).
Redundant array of independent disks. Hard drives in all computer systems are susceptible
to failures caused by temperature variations, head crashes, motor failure,
controller failure, and changing voltage conditions. To improve reliability and protect
the data in their enterprise storage systems, many computer systems use redundant
arrays of independent disks (RAID) storage products.
RAID links groups of standard hard drives to a specialized microcontroller. The
microcontroller coordinates the drives so they appear as a single logical drive, but
they take advantage of the multiple physical drives by storing data redundantly, thus
protecting against data loss due to the failure of any single drive.
Storage area network. A storage area network (SAN) is an architecture for building
special, dedicated networks that allow rapid and reliable access to storage devices by
multiple servers. Storage over IP, sometimes called IP over SCSI or iSCSI, is a technology
that uses the Internet Protocol to transport stored data between devices within
a SAN. Storage visualization software is used with SANs to graphically plot an entire
network and allow storage administrators to view the properties of, and monitor, all
devices from a single console.
Network-attached storage. A network-attached storage (NAS) device is a specialpurpose
server that provides file storage to users who access the device over a network.
The NAS server is simple to install (i.e., plug-and-play), and works exactly like
a general-purpose file server, so no user retraining or special software is needed.
Table 3.2 (on page 70) compares the advantages and disadvantages of the various
secondary storage media.
S t o r a g e S e r v i c e P r o v i d e r s
Storage service providers (SSPs), also called storage-on-demand or storage utilities,
provide customers with the storage capacity they require as well as professional services
including assessment, design, operations, and management. Services offered by
SSPs include primary online data storage, backup and restorability, availability, and
accessibility.
SSPs offer the advantages of implementing storage solutions quickly and managing
storage around-the-clock, even if the storage devices are located at the customer’s
data center. However, there is some increased security risk associated with moving an
enterprise’s data off-site.
Section 3.3 Computer Memory 69
B e f o r e y o u g o o n . . .
1. Describe the four main types of primary storage.
2. Describe different types of secondary storage.
3. How does primary storage differ from secondary storage in terms of speed, cost,
and capacity?
4. Describe the three types of enterprise storage systems.
70 Chapter 3 Computer Hardware
Table 3.2 Secondary Storage
Type Advantages Disadvantages Application
Magnetic storage devices:
Magnetic tape Lowest cost per unit stored Sequential access means slow Corporate data archiving
retrieval speeds
Hard drive Relatively high capacity Fragile; high cost per unit Personal computers through
and fast retrieval speed stored mainframes
RAID High capacity; designed for Expensive, semipermanent Corporate data storage that
fault tolerance and reduced installation requires frequent,
risk of data loss; low cost per rapid access
unit stored
SAN High capacity; designed Expensive Corporate data storage that
for large amounts of requires frequent, rapid
enterprise data access
NAS High capacity; designed Expensive Corporate data storage that
for large amounts of requires frequent, rapid
enterprise data access
Magnetic diskettes Low cost per diskette, Low capacity; very high cost Personal computers
portability per unit stored; fragile
Memory cards Portable; easy to use; less Expensive Personal and laptop computers
failure prone than hard drives
Expandable storage Portable; high capacity More expensive than hard Backup of internal hard
drives drive
Optical storage devices:
CD-ROM High capacity; moderate cost Slower retrieval speeds than Personal computers through
per unit stored; high durability hard drives; only certain corporate data storage
types can be rewritten
DVD High capacity; moderate Slower retrieval speeds than Personal computers through
cost per unit stored hard drives corporate data storage
FMD-ROM Very high capacity; moderate Faster retrieval speeds than Personal computers through
cost per unit stored DVD or CD-ROM; slower corporate data storage
retrieval speeds than hard
drives
The traditional way of comparing classes of computers is by their processing power.
Analysts typically divide computers (called the platform in the computer industry)
into six categories: supercomputers, mainframes, midrange computers (minicomputers
and servers), workstations, notebooks and desktop computers, and appliances.
Recently, the lines between these categories have blurred. This section
presents each class of computer, beginning with the most powerful and ending
with the least powerful. We describe the computers and their respective
roles in modern organizations. IT’s About Business Box 3.1 gives an example
of several different types of computers used in Formula One auto racing.
S u p e r c o m p u t e r s
The term supercomputer does not refer to a specific technology, but to the
fastest computing engines available at any given time. Supercomputers generally
address computationally demanding tasks on very large data sets.
Rather than transaction processing and business applications—the forte of
mainframes and other multiprocessing platforms—supercomputers typically
run military and scientific applications, although their use for commercial applications,
such as data mining, has been increasing. Supercomputers generally operate
at 4 to 10 times faster than the next most powerful computer class, the mainframe. (For
Section 3.4 Computer Hierarchy 71
A bb oo uu tt B uu ss ii nn ee ss ss
Box 3.1: Where the computers meet the road
Formula One racing is big business. Each racing team
has a multimillion-dollar budget each season. The cars
are high-tech, but the real high-tech machines are the
computers and other information technologies that go
into designing—and increasingly, controlling—the race
cars. Formula One racing is perhaps the most technologically
advanced sport in the world.
In 1991, the Williams British racing team introduced
a revolutionary car combining a computer-controlled,
semiautomatic gearbox with electronic traction control.
That computerized car won the 1992 world championship.
For today’s top Formula One racing teams, engineers
draft car designs on numerous Sun workstations
running computer-aided design software (discussed in
Chapter 4). The car models are run through virtual wind
tunnels simulated on a high-end Sun server.
On race day, roughly 120 sensors in the car monitor
everything from engine temperature to the position of
each wheel. The data from the sensors are relayed by microwave
radio to servers that each team keeps trackside;
about 1.2 gigabytes of information are recorded on each
lap. Engineers study the data and make instant decisions
about when to bring the car in for a pit stop and what adjustments
to make. The crew and the engineers can relay
advice to the driver. Formula One teams have installed
controls on the steering wheel that let the driver make
mid-race changes to the car’s transmission and power
train.
The race data from the trackside servers is sent via
high-speed Internet or satellite links to team headquarters.
At the headquarters lab, laps are recorded and later
replayed on a mainframe computer—or with a real car
mounted on a chassis dynamics rig—to fine-tune the
car’s engineering. During races, the lab can send tips for
tweaking a car’s configuration to crews at the track.
Source: Business 2.0 (October 2001).
Questions
1. Identify the different types of computers used by
Formula One racing teams.
2. Does what is learned in Formula One racing transfer
to regular automobiles? Give examples.
‘s
A supercomputer.
3.4 COMPUTER HIERARCHY
a more technical overview of supercomputers, see the section on Supercomputers on
the Web site.)
Supercomputers help analyze the Earth’s crust. Scientists know that the Earth’s
continental plates are constantly moving at a glacial pace over the planet’s surface,
but the complicated internal dynamics that cause this movement are unresolved. The
Earth rids itself of the intense heat within its core through a massive circulation system
of molten earth and solidified crust, powering the movement of the continental
plates. Earlier computer models viewed the process in two dimensions—depth and
horizontal extension—which is not a full 3-D representation of the Earth.
The scale is so huge that, until recently, scientists could not assemble computers
powerful enough to process the immense amount of data necessary to realistically
simulate the movement of the plates. Now, using a massively parallel supercomputer
(4 gigaflops—4 billion floating point operations per second—of processing power) and
specially designed modeling software, researchers at Princeton University in New Jersey
have been moving toward the answer. They are working to understand the cycle
of convection occurring deep inside the planet. The researchers’ work could someday
help scientists accurately predict earthquakes and volcanic eruptions. ●
M a i n f r a m e C o m p u t e r s
Although mainframe computers are increasingly viewed as just another type of server,
albeit at the high end of the performance and reliability scales, they remain a distinct
class of systems differentiated by hardware and software features. Mainframes remain
popular in large enterprises for extensive computing applications that are accessed by
thousands of users. Examples of mainframe applications include airline reservation
systems, corporate payroll, and student grade calculation and reporting. Analysts predict
that Internet-based computing will lead to continued growth in the mainframe
market.
Mainframes are less powerful and generally less expensive than supercomputers.
In 2000, mainframe capacity was priced at approximately $2,260 per MIP (millions of
instructions per second), down significantly from $9,410 in 1997. Prices are expected
to fall to $490 per MIP by 2004. This pricing pressure has forced two vendors of mainframe
systems, Amdahl and Hitachi, out of the mainframe market, leaving only IBM
as a vendor of traditional mainframe systems. IBM calls its mainframe computer series
the eServer zSeries.
A mainframe system may have up to several gigabytes of primary storage. Online
and offline secondary storage (see the discussion of Enterprise Storage Systems on
page 69) may use high-capacity magnetic and optical storage media with capacities in
EXAMPLE
72 Chapter 3 Computer Hardware
A mainframe computer.
the terabyte range. Typically, several hundreds or thousands of online computers can
be linked to a mainframe. Today’s most advanced mainframes perform at more than
2,500 MIPs and can handle up to one billion transactions per day.
Some large organizations that began moving away from mainframes toward distributed
systems now are moving back toward mainframes because of their centralized
administration, high reliability, and increasing flexibility. This process is called
recentralization. The reasons for the shift include supporting the high transaction levels
associated with e-commerce, reducing the total cost of ownership of distributed
systems, simplifying administration, reducing support-personnel requirements, and
improving system performance. In addition, host computing provides a secure, robust
computing environment in which to run strategic, mission-critical applications. (Distributed
computing and related topics are discussed in detail in Chapter 6.) (For a
more technical discussion of mainframes, see the section on Mainframes on the Web
site.)
Merrill Lynch’s online information and customer service. Merrill Lynch, with
total client assets of more than $1.5 trillion, has a long history of mainframe computing
for providing innovative client services. A new mainframe system plays a key
role in delivering information via the company Web site, satisfying nearly 750,000
requests daily. To run its Web presence and to provide support for online Internet
trading, Merrill Lynch uses an IBM mainframe with 8 gigabytes of random access
memory. The mainframe also supports market data capture, feedback pages from
customers, and secure portfolio downloads. Mainframe workloads at Merrill Lynch
have increased 30 percent over the last few years. Internet activity, extended trading
hours, and moving to stock-pricing decimalization are expected to further increase
workloads. ●
M i d r a n g e C o m p u t e r s
There are two types of midrange computers, minicomputers and servers. Minicomputers
are relatively small, inexpensive, and compact computers that perform the same
functions as mainframe computers, but to a more limited extent. These computers are
designed to accomplish specific tasks such as process control, scientific research, and
engineering applications. Larger companies gain greater corporate flexibility by distributing
data processing with minicomputers in organizational units instead of centralizing
computing at one location. Minicomputers meet the needs of smaller
organizations that would rather not utilize scarce corporate resources by purchasing
larger, less scalable computer systems. IBM is the market leader in minicomputers
with its eServer iSeries (formerly the AS/400).
Automated Training Systems moves to the Internet. Automated Training Systems
(ATS) produces training products for midrange computer users, which consist of audiocassette
media and workbooks. ATS packages, developed for in-house training,
allow learning when convenient for students, and they virtually eliminate many of the
problems normally associated with training new users (time away from the job, scheduling,
and travel). ATS uses an IBM iSeries minicomputer to run its Web site. The
company modified its order-entry application with a browser interface to provide its
customers with information and the ability to place orders over the Internet. Since the
implementation of its Internet project, ATS is receiving orders and inquiries from all
EXAMPLE
EXAMPLE
Section 3.4 Computer Hierarchy 73
over the world. In response to demand, ATS is currently working to translate its
courses into Japanese and Chinese. ATS has significantly improved its customer service
and support while reducing its overall costs. The company’s president notes that
ATS received enough orders in one day over the Internet to recover its entire investment
in hardware. ●
Smaller types of midrange computers, called servers, typically support computer
networks, enabling users to share files, software, peripheral devices, and other network
resources. Servers have large amounts of primary and secondary storage and
powerful CPUs.
Servers provide the hardware for e-commerce. They deliver Web pages and
process purchase and sales transactions. Organizations with heavy e-commerce requirements
and very large Web sites are running their Web and e-commerce applications
on multiple servers in server farms. Server farms are large groups of servers
maintained by an organization or by a commercial vendor and made available to
customers.
As companies pack greater numbers of servers in their server farms, they are
using pizza-box-size servers called rack servers that can be stacked in racks. These
computers run cooler, and therefore can be packed more closely, requiring less space.
To further increase density, companies are using a server design called a blade. A
blade is a card about the size of a paperback book on which memory, processor, and
hard drives are mounted.
Immunet: Using the Web to combat AIDS. Immunet uses the Web to help fight
HIV and AIDS. Because HIV and AIDS research evolves so rapidly, medical personnel
must rely on accredited Continuing Medical Education (CME) courses to keep on
top of the latest developments, medications, and treatment protocols. Immunet offers
accredited CME courses online.
In addition, while AIDS treatment is covered by health plans, many of these do
not specify which doctors specialize in AIDS. Patients are often required to seek referrals
from multiple doctors before finding the right one. In some cases, these issues
must first be discussed with human resources personnel who administer health benefit
plans—an uncomfortable option for many seeking help. Immunet provides automated
searches and matches between patients and doctors.
Having acquired two new domain names, aids.edu and aids.org, Immunet chose
IBM Netfinity servers because they provided reliability and availability to enable the
company to manage its Web sites. Immunet now has a comprehensive Web site that
monthly serves more than 80,000 visitors from more than 155 countries. ●
W o r k s t a t i o n s
Computer vendors originally developed desktop engineering workstations, or workstations
for short, to provide the high levels of performance demanded by engineers.
That is, workstations run computationally intensive scientific, engineering, and
financial applications. Workstations are typically based on RISC (reduced instruction
set computing) architecture and provide both very high-speed calculations and
high-resolution graphic displays. These computers have found widespread acceptance
within the scientific community and, more recently, within the business community.
Workstation applications include electronic and mechanical design, medical imaging,
EXAMPLE
74 Chapter 3 Computer Hardware
scientific visualization, 3-D animation, and video editing. By the second half of the
1990s, many workstation features were commonplace in PCs, blurring the distinction
between workstations and personal computers.
M i c r o c o m p u t e r s
Microcomputers (also called micros, personal computers, or PCs) are the smallest and
least expensive category of general-purpose computers. They can be subdivided into
four classifications based on their size: desktops, thin clients, notebooks and laptops,
and mobile devices.
Desktop PCs. The desktop personal computer has become the dominant method of
accessing workgroup and enterprisewide applications. It is the typical, familiar microcomputer
system that has become a standard tool for business, and, increasingly, the
home. It is usually modular in design, with separate but connected monitor, keyboard,
and CPU. In general, modern microcomputers have between 64 megabytes and 512
megabytes of primary storage, one 3.5-inch floppy drive, a CD-ROM (or DVD) drive,
and up to 100 gigabytes or more of secondary storage.
Most desktop systems currently use Intel 32-bit technology (but are moving to 64-
bit technology), running some version of Windows. The exception is the Apple Macintosh,
which runs Mac OS (operating system) on a PowerPC processor. Apple offers
two desktop Macintosh systems, the high-performance Power Mac G4 series and the
entry-level iMac series.
Thin-client systems. Thin-client systems are desktop computer systems that do not
offer the full functionality of a PC. Compared to a PC, thin clients are less complex,
particularly because they lack locally installed software, and thus are easier and less
expensive to operate and support than PCs. The benefits of thin clients include fast
application deployment, centralized management, lower cost of ownership, and easier
installation, management, maintenance, and support. Disadvantages include user resistance
and the need to upgrade servers and buy additional server applications and licenses.
One type of thin client is the terminal, allowing the user to only access an
application running on a server.
Another type of thin client is a network computer, which is a system that provides
access to Internet-based applications via a Web browser and can download software,
usually in the form of Java applets. With PC vendors lowering their systems costs and
simplifying maintenance, NCs remain niche products. However, vendors continue to
manufacture thin-client systems for use at retail stations, kiosks, and other
sites that require access to corporate repositories but little desktop functionality.
Industry experts predicted that the PC would give way to network
computers, but as IT’s About Business Box 3.2 (on page 76) shows, that has
not been the case. Table 3.3 (on page 76) compares the classes of computers
discussed so far.
Laptop and notebook computers. As computers become much smaller and
vastly more powerful, they become portable, and new ways of using them
open up. Laptop and notebook computers are small, easily transportable,
lightweight microcomputers that fit easily into a briefcase. They are designed
for maximum convenience and transportability, allowing access to processing
power and data outside an office environment. Manager’s Checklist 3.1 (on page 77)
compares the trade-offs between desktop and portable PCs.
Section 3.4 Computer Hierarchy 75
A notebook computer.
Mobile Devices. Emerging platforms for computing and communications include
such mobile devices as handheld computers, often called personal digital assistants
(PDAs) or handheld personal computers, and mobile phone handsets with new wireless
and Internet access capabilities formerly associated with PDAs. Other emerging
platforms (game consoles and cable set-top boxes) are consumer electronics devices
that are expanding into computing and telecommunications. Mobile devices are be-
76 Chapter 3 Computer Hardware
A bb oo uu tt B uu ss ii nn ee ss ss
Box 3.2: Predictions of the death of PCs were exaggerated
Conventional wisdom says that the personal computer is
a $1,000 commodity with too much processing power,
too much memory, too much storage, and an unhealthy
dependence on Windows and a Web browser. But in a
survey, three-quarters of IT executives said that the PC
will remain their main desktop computer for the next
five years.
The GartnerGroup (a marketing research group
that often studies IT trends) predicted that 20 percent of
the desktop market would be thin clients by the end of
2002, a figure that in June 2001 actually was about 1 percent.
Instead, the differences between PCs and thin
clients are blurring. Analysts stated that thin clients
could save up to 39 percent on PC total cost of ownership
(TCO), but that has not happened. IT executives
say they do not measure TCO simply as PC versus thin
client. They note that the cost of thin clients is not so low
that it is worth reengineering, retraining, and running a
mixed environment. In fact, the executives said that
TCO should stand for “thin clients are oversimplified.”
Another problem is that if a company moves from
PCs to thin clients, the new environment will require
complex storage systems and higher-end servers. Also, if
an application runs in only one place (i.e., the network),
then the network must always be up and running. One
executive said that his firm could just buy PCs, or it
could spend twice as much on the communications network
in order to keep the client thin.
Source: CIO Magazine (June 1, 2001).
Questions
1. What are the advantages and disadvantages of network
PCs?
2. What is the biggest reason for staying with desktop
PCs?
‘s
Table 3.3 Comparing Computers (Desktop and Larger)
Type Processor Speed Amount of RAM P h y s i c a l S i z e Common Role/Use
Supercomputer 60 billion to 3 trillion 8,000 MB Like a small Scientific calculation, complex
FLOPs car system modeling, and simulation
Mainframe 500–4,500 256–4,096 MB Like a Enterprisewide systems, corporate
MIPS refrigerator database management
Midrange Computers
Minicomputer 250–1,000 MIPS 256–2,048 MB Like a file Department-level or small company;
cabinet dedicated to a particular system
(e.g., e-mail)
Server 100–500 MIPS 256–1,024 MB Fits on Supports computer networks;
desktop e-commerce
Workstation 50–250 MIPS 128–1,024 MB Fits on Engineering/CAD
desktop software development
Microcomputer 10–100 MIPS 64–512 MB Fits on Personal/workgroup
desktop productivity, communication
A personal digital assistant
(PDA).
MIS
coming more popular and more capable of augmenting, or even substituting for, desktop
and notebook computers.
Table 3.4 (on page 78) describes the various types of mobile devices. In general,
mobile devices have the following characteristics:
• They cost much less than PCs.
• Their operating systems are simpler than those on a desktop PC.
• They provide good performance at specific tasks but do not replace the full functions
of a PC.
• They provide both computer and/or communications features.
• They offer a Web portal that is viewable on a screen.
The following example describes an application of PDAs in the U.S. Navy.
PDAs in the U.S. Navy. The U.S. Navy recently realized that its aircraft carrier
flight-grading system was not effective. Officers spent all day recording flight evaluations
in spiral notebooks, then sat for up to two hours each night reentering that data
into a computer.
Using application development software and Palm handheld devices, the Navy
created a flight-recording program, PASS, for devices running the Palm operating system.
The results were immediate. PASS is now in official use by more than 50 landingsignal
officers—pilots who grade flight landings—on two of the Navy’s 12 aircraft
carriers.
The application is simple to use. Using custom menus and the Palm’s built-in
handwriting recognition software, officers input information on each flight, including
the plane ID number, pilot name, which wire helped catch the plane, a grade for the
EXAMPLE
Section 3.4 Computer Hierarchy 77
Impractical for mobile computing Designed for mobile computing
Lower cost Higher cost
Easily expanded Difficult to expand
Comfortable ergonomics Uncomfortable ergonomics (small
keyboard, often with inconvenient
placement of function keys)*
Easy-to-use mouse or other Awkward pointing devices (some
pointing device allow traditional mouse to be
connected)
High-resolution/brightness Lower resolution, less bright*
monitor
High RAM and hard-drive Somewhat less RAM and hard-drive
capacity capacity
Easy serviceability More difficult to service/repair
Can utilize all current PC chips Some models cannot use some chips,
due to cooling problems
* Most portable PCs can be used with a conventional desktop monitor and keyboard when connected
to them through a docking station (a popular option).
Desktop Personal Computer Portable Personal Computer Manager’s Checklist 3.1
Desktop or Portable PC?
The Tradeoffs
landing, and additional evaluation comments. After completing notes on a day’s
worth of flights, the pilot synchronizes his PDA with a desktop computer. The PASS
software flags any evaluations where data are missing or the comments do not mesh
with the flight grade, then it enters the evaluations into the computer’s database. Data
are backed up on Zip disks and periodically sent to the Pacific Fleet’s central data
repository in San Diego.
The program has freed up more than 100 man-hours a month on each of the ships.
Those extra hours are time the officers can spend doing more meaningful training.
PASS has also increased the accuracy of flight evaluations. Thanks to the Palm’s ability
to time-stamp records, landing-signal officers can precisely record landing intervals.
(With flights coming in every 45 seconds at peak times, precision is crucial.)
PASS’s shortcut keys for entering comments take the guesswork out of deciphering
notebook scribbles. ●
C o m p u t i n g D e v i c e s
As technology has improved, ever-smaller computing/communication devices have
become possible. Technology such as wearable computing/communication devices (à
la Star Trek)—which for generations seemed like science fiction—has now become reality.
This section briefly looks at some of these new computing devices.
Wearable computing. Wearable computers are designed to be worn and used on the
body. This new technology has so far been aimed primarily at niche markets in industry
rather than at consumers. Industrial applications of wearable computing include
systems for factory automation, warehouse management, and performance support,
such as viewing technical manuals and diagrams while building or repairing something.
The technology is already widely used in diverse industries such as freight deliv-
78 Chapter 3 Computer Hardware
Table 3.4 Mobile Devices and Their Uses
Device Description and Use
Handheld companions Devices with a core functionality of accessing and managing
data; designed as supplements to notebooks or PCs
PC companions Devices primarily used for personal information management
(PIM), e-mail, and light data-creation capabilities
Personal companions Devices primarily used for PIM activities and data-viewing
activities
Classic PDAs Handheld units designed for PIM and vertical data collection.
Smart phones Emerging mobile phones with added PDA, PIM, data, e-mail
or messaging creation/service capabilities
Vertical application Devices with a core functionality of data access, management,
devices creation, and collection; designed for use in vertical markets*
Pen tablets Business devices with pen input and tablet form for gathering
data in the field or in a mobile situation
Pen notepads Pen-based for vertical data collection applications
Keypad handhelds Business devices with an alphanumeric keypad used in
specialized data-collection applications
*Vertical markets refer to specific industries, such as manufacturing, finance, healthcare, etc.
ery, aerospace, securities trading, and law enforcement. Governments have been examining
such devices for military uses.
Embedded computers are placed inside other products to add features and capabilities.
For example, the average mid-sized automobile has more than 3,000 embedded
computers that monitor every function from braking to engine
performance to seat controls with memory.
Active badges can be worn as ID cards by employees who wish to stay
in touch at all times while moving around the corporate premises. The clipon
badge contains a microprocessor that transmits its (and its wearer’s) location
to the building’s sensors, which send it to a computer. When
someone wants to contact the badge wearer, the phone closest to the person
is identified automatically. When badge wearers enter their offices,
their badge identifies them and logs them on to their personal computers.
Memory buttons are nickel-sized devices that store a small database
relating to whatever it is attached to. These devices are analogous to a
bar code, but with far greater informational content and a content that is subject to
change. The U.S. Postal Service is placing memory buttons in residential mailboxes to
track and improve collection and delivery schedules.
An even smaller form of computer is the smart card. Similar in size and thickness
to ordinary plastic credit cards, smart cards contain a small processor and memory
that allow these “computers” to be used in everyday activities such as personal identification
and banking.
Uses for smart cards are appearing rapidly. People are using them as checkbooks;
a bank ATM (automated teller machine) can “deposit money” into the card’s memory
for “withdrawal” at retail stores. Many states and private health maintenance organizations
are issuing smart health cards that contain the owner’s complete health
history, emergency data, and health insurance policy data. Smart cards are being used
to transport data between computers, replacing floppy disks. Adding a small transmitter
to a smart card can allow businesses to locate any employee and automatically
route phone calls to the nearest telephone.
Input technologies allow people and other technologies to put data into a computer.
We begin with human data-entry devices.
H u m a n D a t a - E n t r y D e v i c e s
Human data-entry devices allow people to communicate with the computer. Some of
these devices are very common, such as the keyboard and the mouse. Others, such as
the touch screen, stylus, trackball, joystick, and microphone, are used for somewhat
more specialized purposes.
B e f o r e y o u g o o n . . .
1. Describe the computer hierarchy from the largest to the smallest computers.
2. What type of desktop PC has the least amount of processing power?
3. Give examples of the uses of supercomputers and handheld computers.
Section 3.5 Input Technologies 79
Active badge worn by
employees.
3.5 INPUT TECHNOLOGIES
Keyboards. Keyboards are the most common input device. The keyboard is designed
like a typewriter but with many additional function keys. Most computer users
utilize keyboards regularly. However, excessive use of keyboards can lead to repetitive
stress injuries like carpal tunnel syndrome. This type of injury is thought to be
caused by improper placement of the hands and wrists when typing at the computer
keyboard. As a result, a new generation of keyboards has been designed to encourage
the proper hand and wrist positions by splitting and angling the keypad and by incorporating
large wrist rests.
A more radical keyboard redesign is the DataHand keyboard from DataHand
Systems of Phoenix, Arizona. The DataHand keyboard consists of two unattached
pads, and rather than a conventional array of keys, the device has touch-sensitive receptacles
(or finger wells) for the fingers and thumbs. Each finger well allows five different
commands, which are actuated by touching one of the sides or the bottom of
the finger wells. Complex commands can be programmed so that a single flick of the
finger can be used to enter frequently used sequences of commands or chunks of data.
The DataHand Web site has an excellent demonstration of the company’s ergonomic
keyboard (see datahand.com).
Mice and trackballs. A mouse is a handheld device used to point a cursor at a desired
place on the screen, such as an icon, a cell in a table, an item in a menu, or any
other object. Once the arrow is placed on an object, the user clicks a button on the
mouse, instructing the computer to take some action. The use of the mouse reduces
the need to type in information or use one of the function keys.
A variant of the mouse is the trackball, which is often used in graphic design. The
user holds an object much like a mouse, but rather than moving the entire device to
move the cursor (as with a mouse), he or she rotates a ball that is built into the top of
the device. Portable computers have some other mouselike technologies, such as the
glide-and-tap pad, used in lieu of a mouse. Many portables also allow a conventional
mouse to be plugged in when desired.
Another variant of the mouse, the optical mouse, replaces the ball, rollers, and
wheels of the mechanical mouse with a light, lens, and a camera chip. It replicates the
action of a ball and rollers by taking photographs of the surface it passes over, and
comparing each successive image to determine where it is going.
The pen mouse resembles an automobile stick shift in a gear box. Moving the
pen and pushing buttons on it perform the same functions of moving the cursor on
the screen as a conventional pointing device. But the pen mouse base stays immobile
on the desk. With a pen mouse, the forearm rests on the desk, saving wear and
tension. Because the mouse is not lifted or moved, the fingers, not the arm, do the
work.
Other human data-entry devices. Touch screens are a technology that divides a computer
screen into different areas. Users simply touch the desired area (often buttons
or squares) to trigger an action. These are common in computers built into self-service
kiosks such as ATM machines and even bridal registries.
A stylus is a pen-style device that allows the user either to touch parts of a predetermined
menu of options (as with a wearable computer, discussed above) or to handwrite
information into the computer (as with some PDAs). (See the photo of the PDA
and stylus on page 76.) The technology may respond to pressure of the stylus, or the
stylus can be a type of light pen that emits light that is sensed by the computer.
A joy stick is used primarily at workstations that display dynamic graphics. It is
also used to play video games. The joy stick moves and positions the cursor at the desired
place on the screen.
80 Chapter 3 Computer Hardware
The “split” keyboard has
improved ergonomics.
A microphone is becoming a popular data-input device as voice-recognition software
improves and people can use microphones to dictate to the computer. These are
also critical technologies for people who are physically challenged and cannot use the
more common input devices.
S o u r c e D a t a A u t o m a t i o n
The object of source data automation is to input data with minimal human intervention.
These technologies speed up data collection, reduce errors, and gather
data at the source of a transaction or other event. Below are the common types.
Cash-transaction devices. Various input devices are common in association
with cash transactions. The most common are ATMs and POS terminals.
Automated teller machines (ATMs) are interactive input/output devices
that enable people to make bank transactions from remote locations. ATMs
utilize touch screen input as well as magnetic card readers.
Point-of-sale (POS) terminals are computerized cash registers that also
often incorporate touch screen technology and bar-code scanners (described
below). These devices allow the input of numerous data such as item sold, price,
method of payment, name or Zip code of the buyer, and so on. Some inputs are
automated; others may be entered by the operator.
Optical scanners. Bar-code scanners, ubiquitous in retail stores, scan the black-andwhite
bar code lines typically printed on labels on merchandise. In addition, bar-code
scanners are very popular for tracking inventory and shipping.
An optical mark reader is a special scanner for detecting the presence of pencil
marks on a predetermined grid, such as multiple-choice test answer sheets. Similarly,
magnetic ink character readers (MICRs) are used chiefly in the banking industry. Information
is printed on checks in magnetic ink that can be read by the MICR technology,
thus helping to automate and greatly increase the efficiency of the check-handling
process.
Optical character recognition (OCR) software is used in conjunction with a scanner
to convert text into digital form for input into the computer. Although the scanner
can digitize any graphic, the OCR software can recognize the individual characters, so
that they can be manipulated. As a practical example, the scanner by itself could “take
a picture” of this page of text and convert it into digital information that the computer
could store as a picture of the text. But you would not be able to decompose the “picture”
file into individual words that could be further modified (manipulated by a wordprocessing
program, for example). The OCR technology enables this last part.
OCR-equipped scanning technologies are very useful when printed documents not
only must be preserved but also would benefit from any manipulations or modifications.
OCR technologies would enable you to scan data, process them with the OCR
software, and then put them into a database, spreadsheet, or word-processing format.
As noted in the earlier section on handheld computers, OCR software is usually
incorporated in stylus-input devices. Although quite sophisticated, OCR programs require
training in order to be able to recognize handwriting. Even then, their accuracy
in interpreting handwritten characters is less than when they are used to interpret
typed text.
Other source data automation devices. Voice-recognition systems are used in conjunction
with microphones to input speech to computers. Voice-recognition software
(VRS) attempts to identify spoken words and translate them into digital text. Like
Section 3.5 Input Technologies 81
A POS terminal.
A comparison of an LCD
(left) with an OLED display
(right).
OCR software used for handwriting recognition, VRS requires training to become accustomed
to the user’s voice and accent. These systems also leave much to be desired
in terms of accuracy of word recognition, though the technology continues to improve.
Sensors are extremely common technologies embedded in other technologies.
They collect data directly from the environment and input them into a computer system.
Examples might include your car’s airbag activation sensor or fuel mixture/pollution
control sensor, inventory control sensors in retail stores, and the myriad types of
sensors built into a modern aircraft.
Cameras can now operate digitally, capturing images and converting them into
digital files. There are digital still-image cameras, and there are now many types of
digital motion-picture cameras. Many computer enthusiasts and practical business
people find it useful to attach small digital cameras to their personal computers. When
linked to the Internet, and using special software such as Microsoft’s NetMeeting,
such a system can be used to conduct desktop videoconferencing.
The output generated by a computer can be transmitted to the user via several devices
and media. The presentation of information is extremely important in encouraging
users to embrace computers. Below is a discussion of common types of output
technologies.
M o n i t o r s
Monitors are the video screens used with most computers that display input as well as
output. Like television sets, monitors come in a variety of sizes and color/resolution
quality. And like television sets, the common desktop monitor uses cathode ray tube
(CRT) technology to shoot beams of electrons to the screen. The electrons
illuminate tiny points on the screen known as pixels. The more
pixels on the screen, the better the resolution. That is, the less space between
pixels—that is, the finer the dot pitch—the better the resolution.
Here are some other useful facts about monitors:
• Portable computers use a flat screen that uses liquid crystal display
(LCD) technology, not CRT.
• LCDs use less power than CRT monitors but cost six to eight times
what an equivalent CRT does.
• LCD monitors may be passive matrix, which have somewhat less display
speed and brightness compared to active matrix monitors, which
function somewhat differently (and cost significantly more).
Organic light-emitting diodes. Organic light-emitting diodes (OLEDs) provide displays
that are brighter, thinner, lighter, and faster than liquid crystal displays (LCDs).
B e f o r e y o u g o o n . . .
1. Distinguish between human data input devices and source data automation.
2. Describe the relationship between OCR technology and scanner technology.
82 Chapter 3 Computer Hardware
3.6 OUTPUT TECHNOLOGIES
LCDs, invented in 1963, have become the standard display for everything from
watches to laptop computers. However, LCD screens are hard to make and expensive.
Compared to LCDs, OLEDs take less power to run, offer higher contrast, look
equally bright from all angles, handle video, and are cheaper to manufacture.
OLEDs do face technical obstacles with color. If you leave OLEDs on for a
month or so, the color becomes very nonuniform. However, OLEDs are probably
good enough right now for cell phones, which are typically used for 200 hours per year
and would likely be replaced before the colors start to fade. But such performance is
not adequate for handheld or laptop displays, for which several thousand hours of life
are required.
Retinal scanning displays. As people increasingly
use mobile devices, many are frustrated
with the interfaces. The interfaces are
too small, too slow, and too awkward to
process information effectively. As a result,
Web sites become unusable, e-mails are constrained,
and graphics are eliminated. One solution
does away with screens altogether. A
firm named Microvision (mvis.com) projects
an image, pixel by pixel, directly onto a
viewer’s retina. This technology, called retinal
scanning displays (RSDs), is used in a variety
of work situations, including medicine, air
traffic control, and controls of industrial machines.
RSDs can also be used in dangerous
situations, for example, giving firefighters in a
smoke-filled building a floor plan.
P r i n t e r s
Printers come in a variety of styles for varying purposes. The three main types are impact
printers, nonimpact printers, and plotters.
Impact printers. Impact printers work like typewriters, using some kind of striking
action. A raised metal character strikes an inked ribbon that makes a printed impression
of the character on the paper. These devices cannot produce high-resolution
graphics, and they are relatively slow, noisy, and subject to mechanical failure. Although
inexpensive, they are becoming less popular.
Nonimpact printers. Nonimpact printers come in two main styles. Laser printers are
higher-speed, high-quality devices that use laser beams to write information on photosensitive
drums, whole pages at a time; then the paper passes over the drum and picks
up the image with toner (similar to ink). Laser printers produce very-high-resolution
text and graphics, making them suitable for a broad range of printing needs from simple
text to desktop publishing. Inkjet printers work differently, by shooting fine
streams of colored ink onto the paper. These are less expensive than laser printers,
but offer somewhat less resolution quality.
Plotters. Plotters are printing devices that use computer-directed pens for creating
high-quality images. They are used in complex, low-volume situations, for example,
creating maps and architectural drawings. Some plotters are quite large, suited for
producing correspondingly large graphics.
Section 3.6 Output Technologies 83
A retinal scanning display
(RSD) device.
V o i c e O u t p u t
Voice output is now possible via sophisticated synthesizer software that can be installed
in most personal computers. A voice output system constructs the sonic equivalent
of textual words, which can then be played through speakers. Other types of
software can manage spoken communication in different ways. For example, one can
purchase programs that integrate telephone voice mail with the computer, so that the
computer can record and make limited responses to incoming calls.
M u l t i f u n c t i o n D e v i c e s
Multifunction devices combine a variety of technologies and are particularly appropriate
for home offices. The technologies include fax, printer, scanner, copy machine,
and answering machine. Depending on how much one wishes to invest and one’s
needs, any combination can be found in a single cost-effective machine.
M u l t i m e d i a
Multimedia output is the computer-based integration of text, sound, still images, animation,
and digitized motion video. It merges the capabilities of computers with televisions,
VCRs, CD players, DVD players, video and audio recording equipment, and
music and gaming technologies. Multimedia usually represents a collection of various
input and output technologies, a system unto itself, as shown in Figure 3.10. Later in
the book we will discuss the business uses of multimedia technology, but for the moment,
consider these useful facts:
84 Chapter 3 Computer Hardware
Figure 3.10 Multimedia
authoring system with a
great variety of input
sources and output displays.
[Source: Based on
illustration in Reseller
Management (November
1993). From the 11/93 VAR
Workbook Series by John
McCormick and Tom Fare,
Multimedia Today Supplement:
VAR Workbook
Series, pp. 4–5, 7.]
TOSHIBA
11:20
SOUND INPUT & OUTPUT
INPUT DEVICES
GRAPHICS AND VIDEO INPUT
Audiocassette player
MIDI synthesizer
CD player
Microphone
VCR for playback
on videocassette
Stereo speakers
for sound output Scanner Videodisc player
CD-ROM
DVD
VCR
Camcorder
Digital camera
To CD-ROM drive for
beta or custom discs
Pen/digitizer Track ball Mouse
Keyboard
MPC Level 2 PC
or Multimedia
Mac w/authoring
software
Touch screen
Bernoulli or
Syquest drive
• High-quality multimedia processing requires the most powerful and sophisticated
microprocessors available. Firms like Intel produce generations of chips especially
designed for multimedia processing.
• Because of the variety of devices that can make up a multimedia system, standards
such as the Multimedia Personal Computer (MPC) Council certification are important
in ensuring that the devices are compliant and compatible.
• Extensive memory capacity—both primary and secondary storage—is essential for
multimedia processing, particularly with video. Video typically requires using compression
techniques to reduce the amount of storage needed. Even with compression
techniques, those who work extensively with video processing often must
augment their secondary storage with devices like writeable CD drives or external
hard drives.
IT’s About Business Box 3.3 discusses a multimedia application.
The majority of this chapter has explained how hardware is designed and how it
works. But it is what the hardware enables, how it is advancing, and how rapidly it is
advancing that are the more complex and important issues for most businesspeople.
B e f o r e y o u g o o n . . .
1. What are the differences between various types of monitors?
2. What are the main types of printers? How do they work?
3. Describe the concept of multimedia, and give an example of a multimedia system.
Section 3.7 Strategic Hardware Issues 85
A bb oo uu tt B uu ss ii nn ee ss ss
Box 3.3: Advertising real estate with multimedia condopronto.com
CondoPronto.com allows travelers to research and book
online for short-term rental and vacation properties. Site
visitors can see a detailed preview of each property, and
obtain information on availability and cost. Before this
Web site opened, there had not been any free, centralized
one-stop for condo owners and travelers. CondoPronto
offers a free “showcase” page to owners, which includes:
one to three photos of the property, text description,
reservation calendar, and online booking services. Condo-
Pronto also offers a multitiered selection of services, and
owners can pay for upgrades in their ads, such as fullmotion-
video virtual walkthroughs. Visitors to the site can
also see a moving 3-D panorama of properties.
CondoPronto’s IT infrastructure includes IBM’s
RS/6000 server running IBM’s DB/2 relational database
and IBM’s WebSphere Commerce Suite. It took only a
couple of months to get the site up and running, and visitors
are providing positive feedback. Condo owners are
saving time and money that they would otherwise have
spent on brokers or advertisements. At the same time,
owners are reaching a worldwide audience.
Source: ibm.com and condopronto.com.
Questions
1. What advantages does multimedia offer condo
owners on CondoPronto?
2. What other multimedia functions can Condo-
Pronto offer?
‘s
3.7 STRATEGIC HARDWARE ISSUES
MKT
In many industries, exploiting computer hardware is a key to competitive advantage.
Successful hardware exploitation comes from thoughtful consideration of the following
issues.
P r o d u c t i v i t y
Hardware technology can affect both personal and organizational productivity. Businesses
need to assess whether employees’ personal productivity is likely to increase as
microprocessor power and speed increase. Perhaps your PC now takes 1/10th of a second
to call up a program. If a new generation of chip can get your PC to call it up in
1/100th of a second, does your productivity increase by tenfold? If so, then an investment
in a more powerful microprocessor might produce a competitive advantage.
Similarly, as primary storage capacity increases, what advantages come your way?
A trend in software is to make each new version more complex. Consider, for example,
the differences in Microsoft Office 2000. This software suite (discussed in Chapter
4) has so many more instructions that it cannot run well on machines with less than
128 MB RAM. To take advantage of the newer software, you need to upgrade machines.
You also need to invest considerable time to understand whether the new innovations
will help you, and then you must master them. The learning curve that
comes with new machines and software typically comes with a cost to your productivity,
at least in the short term. And perhaps by the time you master a new generation of
technology, it will be obsolete. Multiply this decision by the number of employees
who will use the new software, and you have an issue of organizational productivity to
solve.
At the same time, the cost of computers is decreasing while the power is increasing.
Is the workforce prepared to take advantage of these more powerful machines?
How would your business measure the anticipated increases in productivity? You
would need to be able to measure or somehow quantify the changes in organizational
productivity in order to make a reasoned cost–benefit decision.
C h a n g i n g W o r k S t y l e s
Advances in miniaturization of microprocessors and memory devices are ushering in
ever-smaller computing and communication devices that can assist employees in
achieving a productive, nontraditional work style. This is particularly true for employees
who work largely out of the office. Whether at home or on the road, employees
can stay connected to the home office and keep their efforts coordinated with organizational
goals via the cellular telephone, modem (discussed in Chapter 6), and
portable computers of one style or another. All of these devices are enabled by advances
in these technologies. The issue the organization must consider is whether
these new work styles will benefit employees and the firm as a whole. In particular,
does the firm know how to manage these new work styles?
N e w P r o d u c t s a n d S e r v i c e s
Because the cost of computing power continues to decline à la Moore’s Law, organizations
may find that supercomputers are affordable and justifiable. With a supercomputer,
business organizations can tackle increasingly sophisticated problems, from
forecasting to product development to advanced market research. Similarly, advances
in miniaturization of microcontrollers, microprocessors, and memory devices can also
drive the development of new products and services for your firm. Is the organization
ready and able to take advantage of these advances? What new products and services
would advances in hardware make possible for the business?
86 Chapter 3 Computer Hardware
I m p r o v e d C o m m u n i c a t i o n s
Multimedia is often thought of as the basis for an entertainment system, with limited
use in the business world. This is short-sighted thinking. Increasingly, organizations
recognize that multimedia capability is an important aspect of knowledge management
and communication (as IT’s About Business Box 3.3 showed). When integrated
with a firm’s network and/or the Internet, multimedia technology makes possible incredibly
rich communication and knowledge sharing throughout the organization, as
well as with the rest of the world. Many commercial Web sites feature multimedia,
making video, audio, graphic, and textual information available to all who visit. Multimedia
presentations are now the standard for excellence in the business world, and
anyone who has to sell a product, service, or idea benefits from exploiting this technology.
Is your organization ready to do so? What multimedia applications might provide
a competitive advantage for your organization?
B e f o r e y o u g o o n . . .
1. How would you explain the role of various types of computer hardware in personal
productivity? In organizational productivity?
2. What are the upsides and downsides that accompany advances in microprocessor
design?
What’s In IT For Me? 87
WHHAATT’’SS IIN FFORR MEE ??
FOR ALL BUSINESS MAJORS AND NONBUSINESS MAJORS
There are practically no professional jobs in business today that do not require computer
literacy and skills for personal productivity. And there are no industries that do
not use computer technology for one form of competitive advantage or another.
Clearly, the design of computer hardware has profound impacts for businesspeople.
It is also clear that personal and organizational success can an understanding
of hardware design and a commitment to knowing where it is going and what opportunities
and challenges innovations will bring. Because these innovations can occur so
rapidly, hardware decisions at the individual level and at the organizational level are
difficult.
At the individual level, most people who have a home or office computer system
and want to upgrade it, or people contemplating their first computer purchase, are
faced with the decision of when to buy as much as what to buy and at what cost.
At the organizational level, these same issues plague IS professionals, but they
are more complex and more costly. Most organizations have many different computer
systems in place at the same time. Innovations may come to different classes of computers
at different times or rates, and managers must decide when old hardware
legacy systems still have a productive role in the IS architecture, or when they should
be replaced.
IS management at the corporate level is one of the most challenging careers
today, due in no small part to the constant innovation in computer hardware. That
may not be your career objective, but an appreciation of that area is beneficial. After
all, the people who keep you equipped with the right computing hardware, as you can
now see, are very important allies in your success.
ACC
FIN
MKT
POM
HRM
1 Identify the major hardware components of a computer system.
Today’s computer systems have six major components: the central processing unit
(CPU), primary storage, secondary storage, input technologies, output technologies,
and communications technologies.
2 Describe the design and functioning of the central processing unit.
The CPU is made up of the arithmetic-logic unit that performs the calculations, the
registers that store minute amounts of data and instructions immediately before
and after processing, and the control unit that controls the flow of information on
the microprocessor chip.
3 Discuss the relationships between microprocessor component designs and performance.
Microprocessor designs aim to increase processing speed by minimizing the physical
distance that the data (as electrical impulses) must travel, and by increasing the
bus width, clock speed, word length, and number of transistors on the chip.
4 Describe the main types of primary and secondary storage.
There are four types of primary storage: registers, random access memory (RAM),
cache memory, and read-only memory (ROM). All are direct-access memory; only
ROM is nonvolatile. Secondary storage includes magnetic media (tapes, hard drives,
and diskettes) and optical media (CD-ROM, DVD, FMD-ROM, and optical
jukeboxes).
5 Distinguish between primary and secondary storage along the dimensions of speed,
cost, and capacity.
Primary storage has much less capacity than secondary storage, and is faster and
more expensive per byte stored. Primary storage is located much closer to the CPU
than is secondary storage. Sequential-access secondary storage media such as magnetic
tape is much slower and less expensive than direct-access media (e.g., hard
drives, optical media).
6 Define enterprise storage and describe the various types of enterprise storage.
An enterprise storage system is an independent, external system with intelligence
that includes two or more storage devices. There are three major types of enterprise
storage subsystems: redundant arrays of independent disks (RAIDs), storage
area networks (SANs), and network-attached storage (NAS). RAID links groups
of standard hard drives to a specialized microcontroller. SAN is an architecture for
building special, dedicated networks that allow access to storage devices by multiple
servers. A NAS device is a special-purpose server that provides file storage to
users who access the device over a network.
7 Describe the hierarchy of computers according to power and their respective roles.
Supercomputers are the most powerful, designed to handle the maximum computational
demands of science and the military. Mainframes are not as powerful as
supercomputers, but are powerful enough for use by large organizations for centralized
data processing and large databases. Minicomputers are smaller and less
powerful versions of mainframes, often devoted to handling specific subsystems.
Workstations are in between minicomputers and personal computers in speed, capacity,
and graphics capability. Desktop personal computers (PCs) are the most
common personal and business computers. Network computers have less computing
power and storage, relying on connection to a network for communication,
data, processing, and storage resources.
Laptop or notebook computers are small, easily transportable PCs. Palmtop
computers are handheld microcomputers, usually configured for specific applica-
SUMMARY
88 Chapter 3 Computer Hardware
tions and limited in the number of ways they can accept user input and provide output.
Wearable computers, worn on the user’s clothing, free their users’ movements.
Embedded computers are placed inside other products to add features and capabilities.
Employees may wear active badges as ID cards. Memory buttons are nickelsized
devices that store a small database relating to whatever it is attached to. Smart
cards contain a small processor, memory, and an input/output device that allows
them to be used in everyday activities such as personal identification and banking.
8 Differentiate the various types of input and output technologies and their uses.
Principal input technologies include the keyboard, mouse, trackball, touch screen,
stylus, joystick, ATM, POS terminal, bar-code scanner, optical mark reader, optical
character reader, handwriting and voice-recognition systems, sensor, microphone,
and camera. Common output technologies include the monitor, impact and
nonimpact printers, plotter, voice output, multifunction devices, and multimedia.
9 Describe what multimedia systems are and what technologies they use.
Multimedia computer systems integrate two or more types of media, such as text,
graphics, sound, voice, full-motion video, images, and animation. They use a variety
of input and output technologies, often including microphones, musical instruments,
digitizers, CD-ROM, magnetic tape, and speakers. Multimedia systems
typically require additional processing and storage capacity.
10 Discuss strategic issues that link hardware design and innovation to competitive
business strategy.
According to Moore’s Law, microprocessor capability increases ever more rapidly.
Miniaturization is also increasing. These advancements usher in new generations
of faster, more powerful, and more compact computers, as well as new generations
of microcontrollers. Organizations must continually appraise the issues of productivity
work styles, new products and services, and improved communications
against these new options. Adoption decisions are difficult because of heavy past,
current, and future investment.
Discussion Questions 89
1. What factors affect the speed of a microprocessor?
2. If you were the chief information officer (CIO) of a
firm, what factors would you consider when selecting
secondary storage media for your company’s
records (files)?
3. What applications can you think of for voice-recognition
systems?
4. Given that Moore’s Law has proven itself over the
past two decades, speculate on what chip capabili-
INTERACTIVE LEARNING SESSION
Go to the CD, access Chapter 3: Computer Hardware, and read the case presented. It
will describe a business problem that will require you to make a decision on buying
personal computers for your company. You will have to decide which variables are
the most important (e.g., type of processor, processor speed, amount of RAM, hard
drive capacity, etc.), and you must stay within your budget.
For additional resources, go to the book’s Web site for Chapter 3. There you will
find Web resources for the chapter, including additional material about hardware
technologies and systems; links to organizations, people, and technology; ``IT’s About
Business’’ company links; ``What’s in IT for Me?’’ links; and a self-testing Web quiz
for Chapter 3.
DISCUSSION QUESTIONS
6. How would you justify to your employer the added
cost of a multimedia system over that of a nonmultimedia-
capable PC?
7. Give some examples of how wearable computers
might help your company.
8. What types of embedded computers can you think
of in your company? In your home?
90 Chapter 3 Computer Hardware
ties will be 10 years in the future. What might your
desktop PC be able to do?
5. If you were the chief information officer (CIO) of a
firm, how would you explain the workings, benefits,
and limitations of a network computer–based system
as opposed to using networked PCs (that is,
“thin” client vs. “fat” client)?
PROBLEM-SOLVING ACTIVITIES
1. Obtain back issues of Computerworld or other information
systems magazines. (Go back 5 or 10
years.) Note the cost and functionality (e.g., size of
RAM, hard drive capacity, chip speed) of computer
systems listed. Compare costs and functionality year
by year, and plot them on a graph.
2. Design multimedia systems for your personal use
and for your professional use. Give justifications of
the costs in terms of increased productivity and capability.
3. Describe what functions you would want in a PDA.
Give justifications of the cost in terms of increased
productivity and capability both for personal and
professional use.
4. What types of computing problems justify the investment
in a supercomputer for a private-sector firm?
INTERNET ACTIVITIES
1. Access the Web sites of the major hardware manufacturers,
for example, IBM (ibm.com), Sun
(sun.com), Apple (apple.com), Hewlett-Packard
(hp.com), and Silicon Graphics (sgi.com), and obtain
the latest information regarding hardware releases
for all platforms (supercomputer, mainframe,
workstation, personal computer, laptop). Prepare a
table comparing cost, speed, and capacity for each
product across manufacturers.
2. Access the Web sites of the major chip manufacturers,
for example, Intel (intel.com), Motorola (motorola.
com), and Advanced Micro Devices
(amd.com), and obtain the latest information regarding
new and planned chips. Compare performance
and costs across vendors.
3. Access Intel’s Web site (intel.com) and visit its museum
and the animated microprocessor page. Prepare
a presentation of each step in the machine
instruction cycle.
TEAM ACTIVITIES AND ROLE PLAYING
1. Visit your campus computer center. Note each different
type of computer in use and find out what
types of applications are run on each type.
2. Interview your campus CIO and find out on what
basis he or she decides to upgrade particular systems.
What is the CIO’s view of the dynamics of
technology advancement, costs of new technologies,
costs of in-place systems (sunk costs), and anticipated
gains in productivity?
REAL-WORLD CASE amica.com
I n s u r i n g G r o w t h a t A m i c a
The Business Problem Amica is a $3 billion insurance
company operating in 27 states. Facing increased competition
in the insurance industry, Amica wanted to
provide online services to its policyholders, giving them
improved access to the company without compromising
the company’s reputation for highly personalized customer
service.
Slowing market growth in the insurance industry
has led to aggressive price cutting, as insurers have
sought to increase their market share. With the consumer
market (the segment served by Amica) increasingly
willing to change providers for lower rates, the
need to increase customer loyalty has become more
acute.
Amica’s main strategy for growth has been geographic
expansion as well as a major advertising campaign
designed to raise Amica’s profile outside the
Northeast. Amica recognized the need to expand its
range of channels, and embraced the Internet as a new
distribution and communications channel.
The IT Solution In response to business pressures,
Amica developed a Web-based customer self-service
solution that delivers rich content to policyholders as
well as handles transactional services. Nonpolicyholders
can visit the Web site to view information designed
to support their insurance decisions as well as consumer
safety information. Policyholders can access detailed
billing and account history information, pay
premiums, and report claims online. Policyholders can
also obtain auto, homeowner, and liability insurance
quotes online.
Amica’s e-business solution uses multiple Windows
NT servers as its Web servers, linked to a number of
databases that reside on Amica’s mainframe computer.
The company feels that this IT infrastructure was essential
for the success and smooth operation of its Web
Virtual Company Assignment 91
site. The mainframe provides security, reliability, and
“24/7/365” availability. The Web servers provide flexibility,
scalability, and rapid response time to users accessing
the Web site.
The Results Amica experienced a 170 percent increase
in site requests and a 145 percent increase in site
visits during the Web site’s first month of operation. In
addition, through surveys, the company has discovered
that the Web site has increased customer satisfaction.
Amica has also increased new customer acquisition at
lower cost via its Web site.
Source: amica.com.
Questions
1. Explain how the mainframe delivers its advantages
and how the servers deliver their advantages.
2. Could Amica successfully get rid of its mainframe to
save money? Why or why not?
measurement that gives us a cost-benefit
analysis for acquiring new PC hardware.
One measurement is the total cost of ownership
model that investigates the total cost
of the computer equipment,’’ he says. You
have recently studied in depth about hardware
that is available. Your assignment will
be to recommend a TCO model for evaluating
hardware.
Assignment
1. Use a search engine to find information
about TCO. What is a TCO model?
2. What are the disadvantages to the TCO
model?
3. Describe how TCO could be used at EDS
to determine its total computer related
expenditures.
VIRTUAL COMPANY ASSIGNMENT wiley.com/college/turban
E x t r e m e D e s c e n t S n o w b o a r d s
Background The phone rings as you are
about to start your day. You answer the
phone, and Jacob March, the vice president
of information systems, asks if you can
come to his office. You hang up the phone,
grab your blazer, and head down the hall to
Jacob’s office.
Jacob stands up and offers his hand as
you walk into his spacious office. He offers
you a seat in front of his large mahogany
desk. After exchanging a few pleasantries
and some small talk, he asks if you are
ready for another assignment. You tell
Jacob excitedly that you are ready to accept
the challenge.
Jacob explains that some of the PCs
in the office are aging. ``We need some