PART ONE
Requirements

P

art One defines the needs for information communications in the business environment. It discusses the ways in which various forms of information are used and the need for interconnection and networking facilities.

ROAD MAP FOR PART ONE
Chapter 2 Business Information
The requirements for data communications and networking facilities in an organization are driven by the nature and volume of information that is handled. Chapter 2 provides an overview of the four basic categories of information used in any organization: audio, data, image, and video. The chapter discusses some of the salient characteristics of each type and looks at their networking implications.

Chapter 3 Distributed Data Processing
Chapter 3 describes the nature and role of distributed data processing (DDP) in an organization. Virtually all business information systems organizations are distributed, and the networking and communications requirements are driven by the way in which data and applications are distributed.

ISBN 0-558-69515-9

31
Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

CHAPTER

BUSINESS INFORMATION
2.1 2.2 2.3 Audio Networking Implications Data Networking Implications Image Image Representation Image and Document Formats Networking Implications Video Digital Video Networking Implications Performance Measures Response Time Throughput Summary Recommended Reading and Web Sites Key Terms, Review Questions, and Problems

2.4

2.5

2.6 2.7 2.8

ISBN 0-558-69515-9

32
Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.1 / AUDIO

33

Chapter Objectives After reading this chapter, you should be able to ♦ Distinguish between digital and analog information sources. ♦ Characterize business information types into one of four categories: audio, data, image, and video. ♦ Estimate quantitatively the communication resources required by the four types of information sources. ♦ Explain why system response time is a critical factor in user productivity. It is important to understand how information communication relates to business requirements. A first step in this understanding is to examine the various forms of business information. There is a wide variety of applications, each with its own information characteristics. For the analysis and design of information networks, however, the kinds of information usually can be categorized as requiring one of a small number of services: audio, data, image, and video. Our examination covers the following topics: • How the impact of information sources on communications systems is measured • The nature of the four major forms of business information: audio, data, image, and video • The types of business services that relate to each of these forms of information • An introductory look at the implications of these services from the point of view of the communications requirements that they generate Information sources can produce information in digital or analog form. Digital information is represented as a sequence of discrete symbols from a finite “alphabet.” Examples are text, numerical data, and binary data. For digital communication, the information rate and the capacity of a digital channel are measured in bits per second (bps). Analog information is a continuous signal (for example, a voltage) that can take on a continuum of values. An example is the electrical signal coming out of a microphone when someone speaks into it. In this case, the analog electrical signal represents the continuous acoustic changes in air pressure that make up sound. For analog communication, information rate and channel capacity are measured in hertz (Hz) of bandwidth (1 Hz = 1 cycle per second). Virtually any communication signal can be expressed as a combination of pure oscillations of various frequencies. The bandwidth measures the limits of these frequencies. The higher the frequencies allowed, the more accurately a complex signal can be represented.

2.1 AUDIO
ISBN 0-558-69515-9

The audio service supports applications based on sound, usually of the human voice. The primary application using audio service is telephone communication. Other applications include telemarketing, voice mail, audio teleconferencing, and entertainment
Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

34

CHAPTER 2 / BUSINESS INFORMATION
Compact disc FM quality Wideband speech Telephone quality 10 20 50 200 3400 7000 15k 20k

Figure 2.1 Signal Frequency and Bandwidth (Hz)

radio. The quality of sound is characterized mainly by the bandwidth used (Figure 2.1). Voice on a telephone is limited to about 3400 Hz of bandwidth, which is of moderate quality.Voice of teleconference quality requires about 7000 Hz of bandwidth. For highfidelity sound of reasonable quality, about 15,000 Hz (approximately the range of the human ear) is needed. For compact discs, 20,000 Hz is supported for each of two channels for stereo. Audio information can also be represented digitally. The details are given in Chapter 16. We give an abbreviated discussion here. To get a good representation of sound in digital format, we need to sample its amplitude at a rate (samples per second, or smp/s) equal to at least twice the maximum frequency (in Hz) of the analog signal. For voice of telephone quality, one usually samples at a rate of 8000 smp/s. For high-quality sound on compact discs, 44,100 smp/s is the rate used on each channel. After sampling, the signal amplitudes must be put in digital form, a process referred to as quantization. Eight bits per sample are usually used for telephone voice and 16 bits per sample for each channel for stereophonic compact disc. In the first case, 256 levels of amplitude can be distinguished, and in the second, 65,536 levels. Thus, without compression, digital voice requires 8 b/smp × 8,000 smp/s = 64,000 bps. In the case of CDs, a straightforward multiplication of the foregoing parameters leads to a data rate of about 1.41 Mbps for both channels. A CD is usually rated at a capacity of about 600 megabytes (MB). This leads to an audio capacity of about 1 hour of stereo sound. Typical telephone conversations have an average length in the range of 1 to 5 min. For ordinary voice telephone communication, information in either direction is transmitted less than half the time; otherwise, the two parties would be talking at once [SRIR88].

Networking Implications
The requirements just discussed suggest the need for both a powerful and flexible intralocation facility plus access to a variety of outside telephone services. Outside services are provided by public telephone networks, including the local telephone company and long-distance carriers such as AT&T or a national PTT (postal, telegraph, and telephone) authority. In addition, various private networking facilities and leased line arrangements are possible. All of these are discussed in Chapter 12.

ISBN 0-558-69515-9

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.1 / AUDIO

35

Access line to central office PBX Customer premises Telephone company central office (a) Private branch exchange

Customer premises Telephone company central office (b) Centrex

Figure 2.2

Business Telephone Configurations

ISBN 0-558-69515-9

The most effective way of managing voice requirements is to tie all of the phones at a given site into a single system. There are two main alternatives for this: the private branch exchange and Centrex. The private branch exchange (PBX) is an on-premise switching facility, owned or leased by an organization, that interconnects the telephones within the facility and provides access to the public telephone system (Figure 2.2a). Typically, a telephone user on the premises dials a three- or four-digit number to call another subscriber on the premises and dials one digit (usually 8 or 9) to get a dial tone for an outside line, which allows the caller to dial a number in the same fashion as a residential user. Centrex is a telephone company offering that provides the same sort of service as a PBX but performs the switching function in equipment located in the telephone company’s central office as opposed to the customer’s premises (Figure 2.2b). All telephone lines are routed from the customer site to the central switch. The user can still make local calls with a short extension number, giving the appearance of an onpremise switch. Either a PBX or Centrex facility can support a wide variety of voice-related services. Both voice mail and audio teleconferencing can be supported by either approach. A dramatic change in the way in which audio transmission is supported has been underway for some years. This is the increasing reliance on packet transmission using the Internet Protocol (IP), over both the Internet and private IP-based networks. In general, this type of transmission is referred to as voice over IP (VoIP). Both IP-based PBX and IP Centrex offerings are now common.

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

36

CHAPTER 2 / BUSINESS INFORMATION

2.2 DATA
Data consist of information that can be represented by a finite alphabet of symbols, such as the digits 0 through 9 or the symbols represented on a computer keyboard. Common examples of data include text and numerical information. Symbols are often represented in computers or for transmission by groups of 8 bits (octets or bytes). A familiar example of digital data is text or character strings. While textual data are most convenient for human beings, they cannot, in character form, be easily stored or transmitted by data processing and communications systems. Such systems are designed for binary data. Thus a number of codes have been devised by which characters are represented by a sequence of bits. Perhaps the earliest common example of this is the Morse code. Today, the most commonly used text code is the International Reference Alphabet (IRA). 1 Each character in this code is represented by a unique 7-bit pattern; thus 128 different characters can be represented. This is a larger number than is necessary, and some of the patterns represent invisible control characters. IRA-encoded characters are almost always stored and transmitted using 8 bits per character. The eighth bit is a parity bit used for error detection. This bit is set such that the total number of binary 1s in each octet is always odd (odd parity) or always even (even parity). Thus a transmission error that changes a single bit, or any odd number of bits, can be detected. Text, numerical data, and other types of data are typically organized into a database. This topic is explored in Chapter 3. To get some practice in using the concepts introduced so far, let us estimate approximately how many bits are required to transmit a page of text. Commonly, a letter of the alphabet or a typographical symbol is represented by a byte, or 8 bits. Let us consider an 8.5-by-11-in. sheet, with a 1-in. margin on all sides. This leaves a 6.5-by-9 in. message space. A double-spaced page ordinarily has 3 lines to the inch, or 27 lines for the page. In a common typeface, there are 10 characters per inch, or 65 characters per line. This gives us a total 8 × 27 × 65 = 14,040 bits. This overstates the situation because contiguous spaces at the ends of lines are not ordinarily included, and some pages are not full. As a round number, 10,000 bits per page is probably a fair estimate. For a PC communicating with a server over a telephone line using a relatively slow modem, a typical channel capacity is 56,000 bps. Thus, it would take about 0.18 s to transmit a page. This is by no means the whole story. For example, English text is very redundant. That is, the same information can be sent by using many fewer bits. In the experiment just described, we used a standard compression routine to reduce the file to less than 40% of its size, or 4098 bits per page. Another feature that characterizes

ISBN 0-558-69515-9

IRA is defined in ITU-T Recommendation T.50 and was formerly known as International Alphabet Number 5 (IA5). The U.S. national version of IRA is referred to as the American Standard Code for Information Interchange (ASCII). A description and table of the IRA code is contained in Appendix D.

1

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.3 / IMAGE

37

many data-oriented information sources is the response time required, discussed later in this chapter.

Networking Implications
The networking requirements for supporting data applications in an organization vary widely. We begin a consideration of these requirements in Chapter 3.

2.3 IMAGE
The image service supports the communication of individual pictures, charts, or drawings. Image-based applications include facsimile, computer-aided design (CAD), publishing, and medical imaging. As an example of the types of demands that can be placed by imaging systems, consider medical image transmission requirements. Table 2.1 summarizes the communication impact of various medical image types [DWYE92]. As well as giving the bits per image and the number of images per exam, the table gives the transmission time per exam for three standard digital transmission rates: DS-0 = 56 kbps, DS-1 = 1.544 Mbps, and DS-3 = 44.736 Mbps. Again, compression can be used. If we allow some barely perceivable loss of information, we can use “lossy” compression, which might reduce the data by factors

Table 2.1 Transfer Time for Digital Radiology Images Mbytes per Image
0.52

Image Type
Computerized tomography (CT) Magnetic resonance imagery (MRI) Digital angiography Digital fluorography Ultrasound Nuclear medicine Computerized radiography
ISBN 0-558-69515-9

Images per Exam
30

DS-0 Time/Exam (seconds)
2247

DS-1 Time/Exam (seconds)
81

DS-3 Time/Exam (seconds)
3

0.13

50

928

34

1

1 1 0.26 0.016 8 8

20 15 36 26 4 4

2857 2142 1337 59 4571 4571

104 78 48 2 166 166

4 3 2 0.1 6 6

Digitized film DS-0 = 56 kbps DS-1 = 1.544 Mbps DS-3 = 44.736 Mbps

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

38

CHAPTER 2 / BUSINESS INFORMATION

of roughly 10 : 1 to 20 : 1. On the other hand, for medical imaging lossy compression usually is not acceptable. Using lossless compression, ratios for these applications run below 5 : 1.

Image Representation
There are a variety of techniques used to represent image information. These fall into two main categories: • Vector graphics: An image is represented as a collection of straight and curved line segments. Simple objects, such as rectangles and ovals, and more complex objects are defined by the grouping of line segments. • Raster graphics: An image is represented as a two-dimensional array of spots, called pixels.2 In the simplest form, each pixel is either black or white. This approach is used not only for computer image processing but also for facsimile. All of the figures in this book were prepared with a graphics package (Adobe Illustrator) that makes use of vector graphics. Vector graphics involves the use of binary codes to represent object type, size, and orientation. In all these cases, the image is represented and stored as a set of binary digits and can be transmitted using digital signals. Figure 2.3 shows a simple 10 × 10 representation of an image using raster graphics. This could be a facsimile or raster-scan computer graphics image. The 10 × 10 representation is easily converted to a 100-bit code for the image. In this example, each pixel is represented by a single bit that indicates black or white. A grayscale image is produced if each pixel is defined by more than one bit, representing shades of gray. Figure 2.4 shows the use of a 3-bit grayscale to produce eight shades of gray, ranging from white to black. Grayscale can also be used in vector graphics to define the grayscale of line segments or the interior of closed objects such as rectangles.
0000000000 0000000000 0001111000 0001001000 0001111000 0000001000 0000001000 0001111000 0000000000 0000000000 (a) Image (b) Binary code

Figure 2.3

A 100-Pixel Image and Its Binary code
ISBN 0-558-69515-9

2

A pixel, or picture element, is the smallest element of a digital image that can be assigned a gray level. Equivalently, a pixel is an individual dot in a dot-matrix representation of a picture.

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.3 / IMAGE
0 (white) 1 2 3 4 5 6 7 (black)

39

Figure 2.4

An Eight-Level Grayscale

Images can also be defined in color. There are a number of schemes in use for this purpose. One example is the RGB (red-green-blue) scheme, in which each pixel or image area is defined by three values, one for each of the three colors. The RGB scheme exploits the fact that a large percentage of the visible spectrum can be represented by mixing red, green, and blue in various proportions and intensities. The relative magnitude of each color value determines the actual color.

Image and Document Formats
The most widely used format for raster-scan images is referred to as JPEG. The Joint Photographic Experts Group (JPEG) is a collaborative standards-making effort between ISO and ITU-T. JPEG has developed a set of standards for the compression of raster-scan images, both grayscale and color. The JPEG standard is designed to be general purpose, meeting a variety of needs in such areas as desktop publishing, graphic arts, newspaper wire photo transmission, and medical imaging. JPEG is appropriate for high-quality images, including photographs. Another format that is often seen on the Web is the Graphics Interchange Format (GIF). This is an 8-bit color format that can display up to 256 colors and is generally useful for nonphotographic images with a fairly narrow range of color, such as a company logo. Table 2.2 compares these and other popular formats. There are also two popular document formats that are suitable for documents that include text and images. The Portable Document Format (PDF) is widely used on the Web, and PDF readers are available for virtually all operating systems.
Table 2.2 Comparison of Common Graphics File Formats Type
Graphics Interchange Format Joint Photographic Experts Group Portable Network Graphics Raw negative

File Extension
.gif

Compression Methods
Lempel-Ziv-Welch (LZW) algorithm Various lossy Lossless None

Principal Application/Usage
Flat-color graphics, animation Photographic images Replacement for GIF High-end digital camera Document imaging, scanning On-screen display

Originated By
Compuserve

.jpg .png Various

Joint Photographic Experts Group World Wide Web Consortium Individual equipment makers Adobe Systems, Inc. Microsoft Corp.

ISBN 0-558-69515-9

Tagged Image File Format Windows bit map

.tif .bmp

Various or none None

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

40

CHAPTER 2 / BUSINESS INFORMATION

Postscript is a page description language that is built into many desktop printers and virtually all high-end printing systems.

Networking Implications
The various configurations by which image information is used and communicated do not fundamentally differ from the configurations used for text and numerical data. The key difference is in the volume of data. As was mentioned, a page of text contains about 10,000 bits of 8-bit character data. The bit image of a good-quality personal computer screen requires over 2 million bits (i.e., for the 640 × 480 × 256 video mode). A facsimile page with a resolution of 200 dots per inch (which is an adequate but not unnecessarily high resolution) would generate about 4 million bits for a simple black-and-white image and considerably more bits for grayscale or color images. Thus, for image information, a tremendous number of bits is needed for representation in the computer. The number of bits needed to represent an image can be reduced by the use of image compression techniques. In a typical document, whether it contains text or pictorial information, the black and white areas of the image tend to cluster. This property can be exploited to describe the patterns of black and white in a manner that is more concise than simply providing a listing of black and white values, one for each point in the image. Compression ratios (the ratio of the size of the uncompressed image, in bits, to the size of the compressed image) of from 8 to 16 are readily achievable. Even with compression, the number of bits to be transmitted for image information is large. As usual, there are two concerns: response time and throughput. In some cases, such as a CAD/CAM (computer-aided manufacturing) application, the user is interactively manipulating an image. If the user’s terminal is separated from the application by a communications facility, then the communications capacity must be substantial to give adequate response time. In other cases, such as facsimile, a delay of a few seconds or even a few minutes is usually of no consequence. However, the communications facility must still have a capacity great enough to keep up with the average rate of facsimile transmission. Otherwise, delays on the facility will grow over time as a backlog develops.

2.4 VIDEO
The video service carries sequences of pictures in time. In essence, video makes use of a sequence of raster-scan images. Here it is easier to characterize the data in terms of the viewer (destination) of the TV screen rather than the original scene (source) that is recorded by the TV camera.To produce a picture on the screen, an electron beam scans across the surface of the screen from left to right and top to bottom. For black-andwhite television, the amount of illumination produced (on a scale from black to white) at any point is proportional to the intensity of the beam as it passes that point. Thus at any instant in time the beam takes on an analog value of intensity to produce the desired brightness at that point on the screen. Further, as the beam scans, the analog value changes. Thus the video image can be thought of as a time-varying analog signal.

ISBN 0-558-69515-9

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.4 / VIDEO
Screen Scan line Horizontal retrace

41

C

A

C

A

Vertical retrace

D
(a) Even field only

B C A

D
(b) Odd field only

B

D
(c) Odd and even fields

B

Figure 2.5 Video Interlaced Scanning

Figure 2.5 depicts the scanning process. At the end of each scan line, the beam is swept rapidly back to the left (horizontal retrace). When the beam reaches the bottom, it is swept rapidly back to the top (vertical retrace). The beam is turned off (blanked out) during the retrace intervals. To achieve adequate resolution, the beam produces a total of 483 horizontal lines at a rate of 30 complete scans of the screen per second. Tests have shown that this rate will produce a sensation of flicker rather than smooth motion. To provide a flicker-free image without increasing the bandwidth requirement, a technique known as interlacing is used. As Figure 2.5 shows, the odd-numbered scan lines and the even numbered scan lines are scanned separately, with odd and even fields alternating on successive scans. The odd field is the scan from A to B and the even field is the scan from C to D. The beam reaches the middle of the screen’s lowest line after 241.5 lines. At this point, the beam is quickly repositioned at the top of the screen and recommences in the middle of the screen’s topmost visible line to produce an additional 241.5 lines interlaced with the original set. Thus the screen is refreshed 60 times per second rather than 30, and flicker is avoided.
ISBN 0-558-69515-9

Digital Video
The term digital video refers to the capture, manipulation, and storage of video in digital formats. If an analog video camera signal is digitized and then transmitted or stored in a digital format, either compressed or uncompressed, this could be

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

42

CHAPTER 2 / BUSINESS INFORMATION Table 2.3 Digital Television Formats Format
CIF CCIR HDTV

Spatio-Temporal Resolution
360 × 288 × 30 720 × 576 × 30 1280 × 720 × 60

Sampling Rate
3 MHz 12 MHz 60 MHz

considered digital video. However, the term is more typically applied to video content that is initially captured with a digital video device. Digital video cameras capture moving images digitally. In essence, this is done by taking a series of digital photographs, at a rate of at least 30 frames per second. Typically, the resolution is considerably less than that of a digital still camera, more in line with a typical PC screen.The low end of digital video cameras are Web cameras (Webcams). Their low resolution is tailored to match the needs of Webcasting and video messaging. Digital video camera use either the interlaced technique discussed previously or progressive scan, in which all the lines of each frame are drawn in sequence. Progressive scan is used for computer monitors and most HDTV (high-definition television) schemes.

Networking Implications
Applications based on video include instructional and entertainment television, teleconferencing, closed circuit TV, and multimedia. For example, a black-and-white TV signal for video conferencing might have a frame resolution of 360 by 280 pixels sent every 1/30 s with an intensity ranging from black through gray to white represented by 8 bits. This would correspond to a raw data rate, without compression, of about 25 Mbps. To add color, the bit rate might go up by 50%. Table 2.3 gives the sampling rate for three common types of video. The table gives only the rates for luminance, because color is treated differently in the three formats. At the extreme, uncompressed high-definition color television would require more than a gigabit per second to transmit. As with images, lossy compression can be used. Moreover, use can be made of the fact that video scenes in adjacent frames are usually very similar. Reasonable quality can be achieved using compression ratios from about 20 : 1 to 100 : 1. Increasingly important to the enterprise is the transmission of video over IP-based networks, including the Internet and private intranets. This type of transmission is known as video streaming or TVoIP (television over IP). TVoIP places substantial burdens on enterprise networks but brings substantial benefits. Thus, a key management concern is scaling up IP networks to effectively support video transmission while at the same time providing adequate quality of service to other business transmission requirements.

2.5 PERFORMANCE MEASURES
This section considers two key parameters related to performance requirements: response time and throughput.

ISBN 0-558-69515-9

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.5 / PERFORMANCE MEASURES

43

Response Time
Response time is the time it takes a system to react to a given input. In an interactive transaction, it may be defined as the time between the last keystroke by the user and the beginning of the display of a result by the computer. For different types of applications, a slightly different definition is needed. In general, it is the time it takes for the system to respond to a request to perform a particular task. Ideally, one would like the response time for any application to be short. However, it is almost invariably the case that shorter response time imposes greater cost. This cost comes from two sources: • Computer processing power: The faster the computer, the shorter the response time. Of course, increased processing power means increased cost. • Competing requirements: Providing rapid response time to some processes may penalize other processes. Thus the value of a given level of response time must be assessed versus the cost of achieving that response time. Table 2.4, based on [MART88], lists six general ranges of response times. Design difficulties are faced when a response time of less than 1 second is required.

Table 2.4 Response Time Ranges
Greater than 15 seconds This rules out conversational interaction. For certain types of applications, certain types of users may be content to sit at a terminal for more than 15 seconds waiting for the answer to a single simple inquiry. However, for a busy person, captivity for more than 15 seconds seems intolerable. If such delays will occur, the system should be designed so that the user can turn to other activities and request the response at some later time. Greater than 4 seconds These are generally too long for a conversation requiring the operator to retain information in short-term memory (the operator’s memory, not the computer’s). Such delays would be very inhibiting in problem-solving activity and frustrating in data entry activity. However, after a major closure, delays from 4 to 15 seconds can be tolerated. 2 to 4 seconds A delay longer than 2 seconds can be inhibiting to terminal operations demanding a high level of concentration. A wait of 2 to 4 seconds at a terminal can seem surprisingly long when the user is absorbed and emotionally committed to completing what he or she is doing. Again, a delay in this range may be acceptable after a minor closure has occurred. Less than 2 seconds When the terminal user has to remember information throughout several responses, the response time must be short. The more detailed the information remembered, the greater the need for responses of less than 2 seconds. For elaborate terminal activities, 2 seconds represents an important response-time limit. Subsecond response time Certain types of thought-intensive work, especially with graphics applications, require very short response times to maintain the user’s interest and attention for long periods of time.
ISBN 0-558-69515-9

Decisecond response time A response to pressing a key and seeing the character displayed on the screen or clicking a screen object with a mouse needs to be almost instantaneous—less than 0.1 second after the action. Interaction with a mouse requires extremely fast interaction if the designer is to avoid the use of alien syntax (one with commands, mnemonics, punctuation, etc.).

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

44

CHAPTER 2 / BUSINESS INFORMATION

That rapid response time is the key to productivity in interactive applications has been confirmed in a number of studies [SHNE84; THAD81; GUYN88; SEVC03]. These studies show that when a computer and a user interact at a pace that ensures that neither has to wait on the other, productivity increases significantly, the cost of the work done on the computer therefore drops, and quality tends to improve. It used to be widely accepted that a relatively slow response, up to 2 seconds, was acceptable for most interactive applications because the person was thinking about the next task. However, it now appears that productivity increases as rapid response times are achieved. The results reported on response time are based on an analysis of online transactions. A transaction consists of a user command from a terminal and the system’s reply. It is the fundamental unit of work for online system users. It can be divided into two time sequences: • User response time: The time span between the moment a user receives a complete reply to one command and enters the next command. People often refer to this as think time. • System response time: The time span between the moment the user enters a command and the moment a complete response is displayed on the terminal. As an example of the effect of reduced system response time, Figure 2.6 shows the results of a study carried out on engineers using a computer-aided design graphics program for the design of integrated circuit chips and boards [SMIT88]. Each transaction consists of a command by the engineer that alters in some way the graphic image being displayed on the screen. The results show that the rate of transactions increases

4000 3500 3000 2500 2000 1500 1000 500
ISBN 0-558-69515-9

Expert Average Novice

Transactions per user-hour

0 0.25

0.50

0.75 1.00 System response time (seconds)

1.25

1.50

Figure 2.6 Response Time Results for High-Function Graphics

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.5 / PERFORMANCE MEASURES
Full Fast Some Medium Little interaction Slow Changing TV channels on cable service Cross-USA telephone call connect time Other response times Point-of-sale credit card verification Making a 28.8-kbps modem connection Executing a trade on the NY Stock Exchange

45

Web user concentration 0

3

10

Time (seconds)

30

Figure 2.7

Response Time Requirements

ISBN 0-558-69515-9

as system response time falls and rises dramatically once system response time falls below 1 second. What is happening is that as the system response time falls, so does the user response time. This has to do with the effects of short-term memory and human attention span. In terms of the types of computer-based information systems that we have been discussing, rapid response time is most critical for transaction processing systems. The output of management information systems and decision support systems is generally a report or the results of some modeling exercise. In these cases, rapid turnaround is not essential. For office automation applications, the need for rapid response time occurs when documents are being prepared or modified, but there is less urgency for things such as electronic mail and computer teleconferencing. The implication in terms of communications is this: If there is a communications facility between an interactive user and the application and a rapid response time is required, then the communications system must be designed so that its contribution to delay is compatible with that requirement. Thus, if a transaction processing application requires a response time of 1 s and the average time it takes the computer application to generate a response is 0.75 s, then the delay due to the communications facility must be no more than 0.25 s. Another area where response time has become critical is the use of the World Wide Web, either over the Internet or over a corporate intranet.3 The time it takes for a typical Web page to come up on the user’s screen varies greatly. Response times can be gauged based on the level of user involvement in the session; in particular, systems with very fast response times tend to command more user attention. As Figure 2.7 indicates, Web systems with a 3-second or better response time maintain a high level of user attention [SEVC96]. With a response time of between
Intranet is a term used to refer to the implementation of Internet technologies within a corporate organization, rather than for external connection to the global Internet; this topic is explored in Chapter 6.
3

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

46

CHAPTER 2 / BUSINESS INFORMATION

3 and 10 seconds, some user concentration is lost, and response times above 10 seconds discourage the user, who may simply abort the session. For an organization that maintains an Internet Web site, much of the response time is determined by forces beyond the organization’s control, such as the Internet throughput, Internet congestion, and the end user’s access speed. In such circumstances, the organization may consider keeping the image content of each page low and relying heavily on text, to promote rapid response time.

Throughput
The trend toward higher and higher transmission speed makes possible increased support for different services (e.g., broadband-based multimedia services) that once seemed too demanding for digital communications. To make effective use of these new capabilities, it is essential to have a sense of the demands each service puts on the storage and communications of integrated information systems. Services can be grouped into data, audio, image, and video, whose demands on information systems vary widely. Figure 2.8 gives an indication of the data rates required for various information types [TEGE95]. 4

Text browser Character-oriented data Text file server Black-and-white image Color image Facsimile Voice Freeze-frame video Low-end multimedia CD audio VCR-quality video Studio-quality video High-end multimedia High-definition video High-resolution multimedia document scanning 100 1K 10K 100K 1M Bits per second 10M 100M 1G Voice Image

Figure 2.8

Required Data Rates for Various Information Types
ISBN 0-558-69515-9

4

Note the use of a log scale for the x-axis. A basic review of log scales is in the math refresher document at the Computer Science Student Resource Site at WilliamStallings.com/StudentSupport.html.

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.5 / PERFORMANCE MEASURES

47

APPLICATION NOTE File Sizes Image file sizes are based on the amount of information on the page and the color content. High-resolution color images can take up a very large amount of storage space. Sound and video files can be larger still. Though audio files do not contain data or color they extend over a period of time. The characteristics that change audio file sizes are length, sampling rate, and bits per sample. The better the sound quality, the larger the file and the greater the bandwidth required to transmit the sounds. For example, CD quality sound takes up much more space than a basic recording of the same music. Video is essentially a series of color pictures over time and so these files can be immense. Frame rate describes the number of times per second an image is sent and this value can drastically change the file size or bandwidth required for transmission. The popularity of digital cameras has certainly increased and because they can take both pictures and record video, a little understanding of file sizes and types can help preserve hard drive and removable media storage space. To start, we must decide on the quality required. Higher quality reduces our options for reducing file size. Next, we will select the appropriate file type. Depending on the application, this can be an important choice. For example, using very high quality, large images and video on a Web page can have a negative impact on the performance. It is for this reason that so much thought goes in to optimizing content for the web. By adjusting the image size, color content or file type we can dramatically reduce the file size and the page loading time. As an example, the following is a table of various image file sizes based on the type of information. As we can see, changing the file type used and the color content can have a tremendous affect on the size of the file. It is no wonder that a digital camera taking high-resolution pictures can run out of memory so quickly. The downside of modifying an image or stream in order to reduce the size is that you usually sacrifice some amount of quality. Once the quality is gone it cannot be recovered without the original file. For applications like spreadsheets and word processing, there is usually little to be done because organizations standardize on the programs and operating systems for use. Fortunately, files of this type, with text only, do not take up much room. The network is often on the receiving end of this shift to digital images and video streams. As more information is sent via the network, links can become over utilized and response times can be slow. Voice over IP, streaming media, and video conferencing all contribute to
Information
640 × 480 pixel picture 640 × 480 pixel picture 640 × 480 pixel picture 640 × 480 pixel picture 640 × 480 pixel picture

File Type
24 bit bitmap 256 color (8 bit) bitmap 16 color (4 bit) bitmap GIF JPEG

Size
900 KB 300 KB 150 KB 58 KB 45 KB

ISBN 0-558-69515-9

(Continued)

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

48

CHAPTER 2 / BUSINESS INFORMATION

performance problems. It is for this reason that we must now design for these changes and network administrators are under pressure to supply sufficient network storage and the bandwidth. However, an administrator is not able to change the information once it is sent. By optimizing the data before it is transmitted, we can often alleviate network utilization and storage problems before they even start.

2.6 SUMMARY
Business communications systems and networks must deal with a variety of information types, which can be conveniently categorized as voice, data, image, and video. Each type presents its own requirements in terms of throughput, response time, and demands on the networking facility.

2.7 RECOMMENDED READING AND WEB SITES
Three books that provide expanded coverage of the topics explored in this chapter are [LI04], [RAO02], and [STEI02].
LI04 Li, Z., and Drew, M. Fundamentals of Multimedia. Upper Saddle River, NJ: Prentice Hall, 2004. RAO02 Rao, K.; Bojkovic, Z.; and Milovanovic, D. Multimedia Communication Systems: Techniques, Standards, and Networks. Upper Saddle River, NJ: Prentice Hall, 2002. STEI02 Steinmetz, R., and Nahrstedt, K. Multimedia Fundamentals, Volume 1: Media Coding and Content Processing. Upper Saddle River, NJ: Prentice Hall, 2002.

Recommended Web sites:
• Multimedia Communications Research Laboratory: At Bell Labs. Good source of leading-edge research information.

2.8 KEY TERMS, REVIEW QUESTIONS, AND PROBLEMS
Key Terms analog audio ASCII byte Centrex digital data GIF grayscale image interlacing JPEG IRA PDF pixel Postscript Private branch exchange (PBX) quantization raster graphics response time vector graphics video voice

ISBN 0-558-69515-9

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.8 / KEY TERMS, REVIEW QUESTIONS, AND PROBLEMS

49

Review Questions
2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 What are the two different interpretations of the prefixes kilo, mega, and giga? Define a context in which each interpretation is used. What is the bandwidth of telephone voice? The process that takes advantage of redundancy to reduce the number of bits sent for a given piece of data is called what? What is the difference between Centrex and PBX? What is the difference between a printable character and a control character? Explain the basic principles of vector graphics and raster graphics. List two common image formats. List two common document formats. Describe the process used to prevent flicker in a video screen. Define response time. What is considered an acceptable system response time for interactive applications and how does this response time relate to acceptable response times for Web sites?

Problems
2.1 How many CD-quality music channels can be transmitted simultaneously over a 10-Mbps Ethernet, assuming that no other traffic is carried on the same network and ignoring overhead? The compact disc (CD) was originally designed to hold audio data. The information on the CD is arranged in a specific format that divides the data into segments. Hardware design considerations in effect at the time the CD was developed dictated that each second of audio would span exactly 75 sectors on the CD. a. Using stereo audio at a standard high-quality audio CD sample rate, what is the maximum number of bytes that can be stored in a single audio CD sector? b. How many minutes of CD quality audio can a 700-MB CD hold? c. How much storage is required to hold 5 minutes of CD-quality audio? What types of media are available to hold this amount of data? A company’s telephone exchange digitizes telephone channels at 8000 smp/s, using 8 bits for quantization. This telephone exchange must transmit simultaneously 24 of these telephone channels over a communications link. a. What’s the required data rate? b. In order to provide answering-machine service, the telephone exchange can store 3-minute audio messages of the same quality as that of the telephone channels. How many megabytes of data storage space are needed to store each of these audio messages? How many bits will it take to represent the following sets of outcomes? a. The uppercase alphabet A, B, . . . , Z b. The digits 0, 1, . . . , 9 c. The seconds in a 24-hour day d. The people in the United States (about 300,000,000 of them) e. Population of the world (about 6 billion) IRA is a 7-bit code that allows 128 characters to be defined. In the 1970s, many newspapers received stories from the wire services in a 6-bit code called TTS. This code carried upper and lowercase characters as well as many special characters and formatting commands. The typical TTS character set allowed over 100 characters to be defined. How do you think this could be accomplished? Review the IRA code in the document at this book’s Web site. a. Indicate the 7-bit code for the following letters: D, d, H, h. b. Repeat part (a), but this time show the 8-bit code that includes an odd parity bit.

2.2

2.3

2.4

2.5

ISBN 0-558-69515-9

2.6

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

50

CHAPTER 2 / BUSINESS INFORMATION 2.7 2.8 In a document, what standard ASCII characters might fall into the category of invisible? A primary primitive (due to its atomic nature) data type available to most programming languages is the character data type.This data type has traditionally been represented internally to the computer system using the ASCII (7-bit) or Extended Binary-Coded Decimal Interchange Code, EBCDIC (8-bit) encoding schemes. Programming languages such as Java use the Unicode (16-bit) encoding scheme to represent primitive character data elements (www.unicode.org). Discuss the implications, beneficial and detrimental, inherent in the decision process regarding the use of these various encoding schemes. Base64 encoding allows arbitrary sequences of octets to be represented by printable characters. The encoding process represents 24-bit groups of input bits as strings of four encoded characters. The 24-bit groups are formed by concatenating three octets. These 24-bit groups are then treated as four concatenated 6-bit groups, each of which is translated to a character of the Base64 alphabet. The encoded output stream is represented by lines of no more than 76 printable characters, with line breaks being indicated by the “CR, LF” character sequence. How much will a file be expanded by encoding it with Base64? The text of the Encyclopaedia Britannica is about 44 million words. For a sample of about 2000 words, the average word length was 5.1 characters per word. a. Approximately how many characters are there in the encyclopedia? (Be sure to allow for spaces and punctuation between words.) b. How long would it take to transmit the text over a T-1 line at 1.544 Mbps? On a fiber optic link at 2.488 Gbps? c. Could the text fit on a 600-MB CD? A drawing in a 8.5-by-11-inch sheet is digitized by means of a 300 dpi (dots per inch) scanner. a. What is the visual resolution of the resulting image (number of dots in each dimension)? b. If 8 bits are used for the quantization of each pixel, how much data storage space is needed to store the image as raw data? When examining X-rays, radiologists often deal with four to six images at a time. For a faithful digital representation of an X-ray photograph, a pixel array of 2048 by 2048 is typically used with a grayscale of intensity for each pixel of 12 bits. As you would hope, radiologists do not look kindly on compression that degrades quality. a. How many levels of grayscale are represented by 12 bits? b. How many bits does it take to represent an X-ray based on these parameters? c. Suppose five X-rays have to be sent to another site over a T-1 line (1.544 Mbps). How long would it take, at best—ignoring overhead? d. Suppose now that we wish to build a communications system that will provide the five X-rays of part (c) upon demand; that is, that from the time the X-rays are requested we want them available within 2 s. What is a lowest channel rate that can support this demand? e. The next generation of displays for X-rays is planned for 4096 by 4096 pixels with a 12-bit grayscale.What does the answer to part d become when using this resolution? A multimedia version of a multivolume reference book is being prepared for storage on compact disc (CD-ROM). Each disc can store about 700 MB (megabytes). The input to each volume consists of 1000 pages of text typed 10 characters to the inch, 6 lines to the inch, on 8.5-by-11-in. paper with 1-in. margins on each side. Each volume also has about 100 pictures, which will be displayed in color at Super VGA resolution (1024 × 768 pixel, 8 b/pixel). Moreover, each volume is enhanced for the CD version with 30 min of audio of teleconferencing quality (16,000 smp/s, 6 b/smp). a. How many bits are there on a 700-megabyte CD (1 megabyte = 220 bytes)? b. Without compression and ignoring overhead, how many volumes can be put on one CD?

2.9

2.10

2.11

2.12

2.13

ISBN 0-558-69515-9

Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.

2.8 / KEY TERMS, REVIEW QUESTIONS, AND PROBLEMS

51

2.14

2.15

2.16

2.17

2.18

c. Suppose the material is to be transmitted over a T-1 facility (1.544 Mbps). How long will it take, exclusive of overhead, to transmit a volume? d. Suppose the text can be compressed at a 3 : 1 rate, the pictures at 10 : 1, and the audio at 2 : 1. How many volumes, exclusive of overhead, will fit on a CD? How long will it take to transmit a compressed volume on a T-1 channel? Commonly, medical digital radiology ultrasound studies consist of about 25 images extracted from a full-motion ultrasound examination. Each image consists of 512 by 512 pixels, each with 8 b of intensity information. a. How many bits are there in the 25 images? b. Ideally, however, doctors would like to use 512 × 512 × 8 bit frames at 30 fps (frames per second). Ignoring possible compression and overhead factors, what is the minimum channel capacity required to sustain this full-motion ultrasound? c. Suppose each full-motion study consists of 25 s of frames. How many such studies could fit on a 600-MB CD-ROM? An 800 × 600 image with 24-bit color depth needs to be stored on disc. Even though the image might contain 224 different colors, only 256 colors are actually present. This image could be encoded by means of a table (palette) of 256 24-bit elements and, for each pixel, an index of its RGB value in the table. This type of encoding is usually called Color Look-Up Table (CLUT) encoding. a. How many bytes are needed to store the image raw information? b. How many bytes are needed to store the image using CLUT encoding? c. What’s the compression ratio achieved by using this simple encoding method? A digital video camera provides an uncompressed output video stream with a resolution of 320 × 240 pixels, a frame rate of 30 fps, and 8 bits for quantization of each pixel. a. What’s the required bandwidth for the transmission of the uncompressed video stream? b. How much data storage space is needed to record two minutes of the video stream? An MPEG-encoded video stream with a resolution of 720 × 480 pixels and a frame rate of 30 fps is transmitted over a network. An old workstation is meant to receive and display the video stream. It takes 56 milliseconds for the workstation to decode each received frame and display it on the screen. Will this workstation display the video stream successfully? Why or why not? Refer to Table 2.4 for a list and description of general system response time categories. Assume you are providing five computer systems for video game developer and the budget for hardware (e.g., CPU clock speed, RAM, bus data rate, disc I/O data rate, etc.) is limited by the consumer. The base price of a system is $1.5K (providing basic >15 s response time for all application scenarios) and each increase in response time range adds $0.5K to the cost of the system hardware. Make a case for the classification of each of the systems running the application scenarios listed that will keep the customer’s total cost for the five systems below $15K. a. System 1—word-processing application and basic Internet access b. System 2—skateboard equipment checkout terminal (for gamer recreation) c. System 3—graphic design and basic Internet access d. System 4—programming and basic Internet access e. System 5—video game testing Note: There is no unique solution. In your answer, each lettered response should provide some rationalization as to why each category was selected, to show that you have given some thought to the impact on productivity that choices such as these can affect.

ISBN 0-558-69515-9 Business Data Communications, Sixth Edition, by William Stallings. Published by Prentice Hall. Copyright © 2009 by Pearson Education, Inc.