inside the computer

2-1 Data Storage: A Bit About the Bit

INTERNET BKIDGE A computer can have seemingly limitless capabilities. It's an entertainment center with hundreds of interactive games. It's a virtual school or university providing interactive instruction and testing in everything from anthropology to zoology. It's a painter's canvas. It's a video telephone. It's a CD player. It's a home or office library giving ready access to the complete works of Shakespeare or interactive versions of corporate procedures manuals. It's a television. It's the biggest marketplace in the world. It's a medical instrument capable of monitoring vital signs. It's the family photo album. It's a print shop. It's a wind tunnel that can test experimental airplane designs. It's a voting booth. It's a calendar with a to-do list. It's a recorder. It's an alarm clock that can remind you to pick up the kids. It's an encyclopedia. It can perform thousands of specialty functions that require specialized skills: preparing taxes, drafting legal documents, counseling suicidal patients, and much more.
In all of these applications, the computer deals with everything as electronic signals. Electronic signals come in two flavors—analog and digital. Analog signals are continuous wave forms in which variations in frequency and amplitude can be used to represent information from sound and numerical data. The sound of our voice is carried by analog signals when we talk on the telephone. With digital signals, everything is described in two states: the circuit as either on or off. Generally, the on state is expressed or represented by the number 1 and the off state by the number 0. Computers are digital and, therefore, require digital data. To make the most effective use of computers and automation, most everything in the world of electronics and communication is going digital.

A by-product of the computer revolution is a. trend toward "going digital" whenever possible. For example, the movement in the recording industry has been away from analog recording on records and toward digital recording on CD. One reason for this move is that analog signals cannot be reproduced exactly. If you have ever duplicated an analog audiotape, you know that the copy is never as good as the source tape. In contrast, digital CDs can be copied onto other CDs over and over without deterioration. When CDs are duplicated, each is an exact copy of the original.
So how do you go digital? You simply need to digitize your material. To digitize means to convert data, analog signals, images, and so on into the discrete format (Is and Os) that can be interpreted by computers. For example, Figure 2-1 shows how music can be digitized. Once digitized, the music recording, data, image, shape, and so on can be manipulated by the computer. For example, old recordings of artists from Enrico Caruso to the Beatles have been digitized and then digitally reconstructed on computers to eliminate unwanted distortion and static. Some of these reconstructed CDs are actually better than the originals!

It's amazing, but the seemingly endless potential of computers is based on the two digital states—on and off. The electronic nature of the computer makes it possible to combine these two electronic states to represent letters, numbers, colors, sounds, images, shapes, and much more—even odors. An "on" or off electronic state is represented by a bit, short for ftnary digit. In the binary numbering system (base 2), the on-bit is a 1 and the off-bit is a 0.
Vacuum tubes, transistors, and integrated circuits characterize the generations of computers. Each of these technologies enables computers to distinguish between on and off and, therefore, to use binary logic. Physically, these states are achieved in a variety of ways.
• In RAM, the two electronic states often are represented by the presence or absence of an electrical charge.
• In disk storage, the two states are made possible by the magnetic arrangement of the surface coating on magnetic tapes and disks.

The recording industry has gone digital. To create a master CD, analog signals are converted to digital signals that can be manipulated by a computer and written to a master CD. The master is duplicated and the copies are sold through retail channels.
• In CDs and CD-ROMs, digital data are stored permanently as microscopic pits.
• In fiber optic cable, binary data flows through as pulses of light.
Bits may be fine for computers, but human beings are more comfortable with letters and decimal numbers (the base-10 numerals 0 through 9). We like to see colors and hear sounds. Therefore, the letters, decimal numbers, colors, and sounds that we input into a computer system while doing word processing, graphics, and other applications must be translated into Is and Os for processing and storage. The computer translates the bits back into letters, decimal numbers, colors, and sounds for output on monitors, printers, speakers, and so on.

ASCII (American Standard Code for /nformation interchange—pronounced "AS-key") is the most popular encoding system for PCs and data communication. In ASCII, alphanumeric characters are encoded into a bit configuration on input so that the computer can interpret them. This coding equates a unique series of on-bits and off-bits with a specific character. Just as the words mother and father are arbitrary English-language character strings that refer to our parents, 1000010 is an arbitrary ASCII code that refers to the letter B. When you tap the letter B on a PC keyboard, the B is transmitted to the processor as a coded string of binary digits (for example, 1000010 in ASCII), as shown in Figure 2-3. The characters are decoded on output so we can interpret them. The combination of bits used to represent a character is called a byte (pronounced "bite").

The 7-bit ASCII code can represent up to 128 characters (27). Although the English language has considerably fewer than 128 printable characters, the extra bit configurations are needed to represent additional common and not-so-common special characters (such as - [hyphen]; @ [at]; I [a broken vertical bar]; and [tilde]) and to signal a variety of activities to the computer (such as ringing a bell or telling the computer to accept a piece of datum).
ASCII is a 7-bit code, but the PC byte is 8 bits. There are 256 (28) possible bit configurations in an 8-bit byte. Hardware and software vendors accept the 128 standard ASCII codes and use the extra 128 bit configurations to represent control characters or noncharacter images to complement their hardware or software product. For example, the IBM-PC version of extended ASCII contains the characters of many foreign languages (such as a [umlaut] and e [acute]) and a wide variety of graphic images that can be combined on a text screen to produce larger images (for example, the box around a window on a display screen).
Microsoft Windows uses the 8-bit ANSI encoding system (developed by the American National Standards /nstitute) to enable the sharing of text between Windows applications. As in IBM, extended ASCII, the first 128 ANSI codes are the same as the ASCII codes, but the next 128 are defined to meet the specific needs of Windows applications.

A consortium of heavyweight computer industry companies, including IBM, Microsoft, and Sun Microsystems, is sponsoring the development of Unicode, a uniform 16-bit encoding system. Unicode will enable computers and applications to talk to one another more easily and will accommodate most languages of the world (including Hebrew, Japanese if, and upper- and lower-case Greek i|i). ASCII, with 128 (27) character codes, is sufficient for the English language but falls far short of the Japanese language requirements. Unicode's 16-bit code allows for 65,536 characters (216). The consortium is proposing that Unicode be adopted as a standard for information interchange throughout the global computer community. Universal acceptance of the Unicode standard would help facilitate international communication in all areas, from monetary transfers between banks to e-mail. With Unicode as a standard, software could be created more easily to work with a wider base of languages.
Unicode, like any advancement in computer technology, presents conversion problems.
The 16-bit Unicode demands more memory than do traditional 8-bit codes. An A in Unicode takes twice the RAM and disk space of an ASCII A. Currently, 8-bit encoding systems provide the foundation for most software packages and existing databases. Should Unicode be adopted as a standard, programs would have to be revised to work with Unicode, and existing data would need to be converted to Unicode. Information processing needs have simply outgrown the 8-bit standard, so conversion to Unicode or some other standard is inevitable. This conversion, however, will be time-consuming and expensive.Creating text for a video display calls for a technology called "character generation." An 8-bit encoding system, with its 256 unique bit configurations, is more than adequate to represent all of the alphanumeric characters used in the English language. The Chinese, however, need a 16-bit encoding system, like Unicode, to represent their 13,000 characters. .

RAM TECHNOLOGY: SDRAM AND RDRAM

RAM, a read-and-write memory, enables data to be both read and written to memory. RAM (primary storage) is electronic circuitry with no moving parts. Electrically charged points in the RAM chips represent the bits (Is and Os) that comprise the data and other information stored in RAM. With no mechanical movement, data can be accessed from RAM at electronic speeds. Over the past two decades, researchers have given us a succession of RAM technologies, each designed to keep pace with ever-faster processors. Existing PCs have FPM RAM, EDO RAM, SRAM, DRAM, BEDO RAM, and other types of memory. However, most new PCs are being equipped with synchronous dynamic RAM (SDRAM). SDRAM is able to synchronize itself with the processor, enabling data transfer at more than twice the speed of previous RAM technologies. With the next generation of processors, we'll probably move to Rambus DRAM (RDRAM), which is six times faster than SDRAM.
A state-of-the-art SDRAM memory chip, smaller than a postage stamp, can store about 128,000,000 bits, or more than 12,000,000 characters of data! Physically, memory chips are installed on single in-line memory modules, or SIMMs, and on the newer dual in-line memory modules, or DIMMS. SIMMS are less expensive, but have only a 32-bit data path to the processor, whereas DIMMs have a 64-bit data path.
There is one major problem with RAM storage: It is volatile memory. That is, when the electrical current is turned off or interrupted, the data are lost. Because RAM is volatile, it provides the processor only with temporary storage for programs and data. Several nonvolatile memory technologies have emerged, but none has exhibited the qualities necessary for widespread use as primary storage (RAM). Although much slower than RAM, nonvolatile memory is superior to SDRAM for use in certain computers because it is highly reliable, it is not susceptible to environmental fluctuations, and it can operate on battery power for a considerable length of time. For example, it is well suited for use with industrial robots.
READIN', 'RITIN', AND RAM CRAM All programs and data must be transferred to RAM from an input device (such as a keyboard) or from secondary storage (such as a disk) before programs can be executed and data can be processed. Therefore, RAM space is always at a premium. Once a program is no longer in use, the storage space it occupied is assigned to another program awaiting execution. PC users attempting to run too many programs at the same time face RAM cram, a situation in which there is not enough memory to run the programs. Programs and data are loaded to RAM from secondary storage because the time required to access a program instruction or piece of datum from RAM is significantly less than from secondary storage. Thousands of instructions or pieces of data can be accessed from RAM in the time it would take to access a single piece of datum from disk storage. RAM is essentially a high-speed holding area for data and programs. In fact, nothing really happens in a computer system until the program instructions and data are moved from RAM to the processor.

Interaction between Computer System Components
During processing, instructions and data are passed between the various types of internal memories, the processor's control unit and arithmetic and logic unit, the coprocessor, and the peripheral devices over the common electrical bus. A system clock paces the speed of operation within the processor and ensures that everything takes place in timed intervals. Refer back to this figure as you read about these components throughout the remainder of this chapter.
illustrates how all input/output (I/O) is "read to" or "written from" RAM. Programs and data must be "loaded," or moved, to RAM from secondary storage for processing. This is a nondestructive read process; that is, the program and data that are read reside in both RAM (temporarily) and secondary storage (permanently) .
The data in RAM are manipulated by the processor according to program instructions. A program instruction or a piece of datum is stored in a specific RAM location called an address. RAM is analogous to the rows of boxes you see in post offices. Just as each P.O. box has a number, each byte in RAM has an address. Addresses permit program instructions and data to be located, accessed, and processed. The content of each address changes frequently as different programs are executed and new data are processed.

To facilitate an even faster transfer of instructions and data to the processor, computers are designed with cache memory (see Figure 2-5). Cache memory is used by computer designers to increase computer system throughput. Throughput refers to the rate at which work can be performed by a computer system. Like RAM, cache is a high-speed holding area for program instructions and data. However, cache memory uses internal storage technologies, such as SDRAM, that are much faster (and much more expensive) than conventional RAM. With only a fraction of the capacity of RAM, cache memory holds only those instructions and data that are likely to be needed next by the processor. Cache memory is effective because,in a typical session, the same data or instructions are accessed over and over. The processor first checks cache memory for needed data and instructions, thereby reducing the number of accesses to the slower SDRAM.

A special type of internal memory, called readonly memory (ROM), cannot be altered by the user (see Figure 2-5). The contents of ROM (rhymes with "mom"), a nonvolatile technology, are "hard-wired" (designed into the logic of the memory chip) by the manufacturer and can be "read only." When you turn on a microcomputer system, a program in ROM automatically readies the computer system for use and produces the initial display-screen prompt.
A variation of ROM is programmable read-only memory (PROM). PROM is ROM into which you, the user, can load read-only programs and data. Generally, once a program is loaded to PROM, it is seldom, if ever, changed. ROM and PROM are used in a variety of capacities within a computer system.
Flash memory is a type of PROM that can be altered easily by the user. Flash memory is a feature of many new processors, I/O devices, and storage devices. The logic capabilities of these devices can be upgraded by simply downloading new software from a vendor-supplied disk to flash memory. Upgrades to early processors and peripheral devices required the user to replace the old circuit board or chip with a new one. The emergence of flash memory has eliminated this time-consuming and costly method of upgrade. Look for nonvolatile flash memory to play an increasing role in computer technology as its improvements continue to close the gap between the speed and flexibility of CMOS RAM.

Just as the processor is the nucleus of a computer system, the control unit is the nucleus of the processor. It has three primary functions:
1. To read and interpret program instructions
2. To direct the operation of internal processor components
3. To control the flow of programs and data in and out of RAM
During program execution, the first in a sequence of program instructions is moved from RAM to the control unit, where it is decoded and interpreted by the decoder. The control unit then directs other processor components to carry out the operations necessary to execute the instruction.
The control unit contains high-speed working storage areas called registers that can store no more than a few bytes (see Figure 2-5). Registers handle instructions and data at a speed about 10 times faster than that of cache memory and are used for a variety of processing functions. One register, called the instruction register, contains the instruction being executed. Other general-purpose registers store data needed for immediate processing. Registers also store status information. For example, the program register contains the RAM address of the next instruction to be executed. Registers facilitate the movement of data and instructions between RAM, the control unit, and the arithmetic and logic unit.

The arithmetic and logic unit performs all computations (addition, subtraction, multiplication, and division) and all logic operations (comparisons). The results are placed in a register called the accumulator. Examples of computations include the payroll deduction for social security, the day-end inventory, and the balance on a bank statement. A logic operation compares two pieces of data, either alphabetic or numeric. Based on the result of the comparison, the program "branches" to one of several alternative sets of program instructions. For example, in an inventory system each item in stock is compared to a reorder point at the end of each day. If the inventory level falls below the reorder point, a sequence of program instructions is executed that produces a purchase order.

Just as big cities have mass transit systems that move large numbers of people, the computer has a similar system that moves millions of bits a second. Both transit systems use buses, although the one in the computer doesn't have wheels. All electrical signals travel on a common electrical bus. The term bus was derived from its wheeled cousin because passengers on both buses (people and bits) can get off at any stop. In a computer, the bus stops are the control unit, the arithmetic and logic unit, internal memory (RAM, ROM, flash, and other types of internal memory), and the device controllers (small computers) that control the operation of the peripheral devices (see Figure 2-5).
The bus is the common pathway through which the processor sends/receives data and commands to/from primary and secondary storage and all I/O peripheral devices. Bits traveling between RAM, cache memory, and the processor hop on the address bus and the data bus. Source and destination addresses are sent over the address bus to identify a particular location in memory, then the data and instructions are transferred over the data bus to or from that location.

MORE IS NOT ALWAYS BETTER Most processors in mainframe computers and personal computers have a CISC (complex instruction set computer) design. A CISC computer's machine language offers programmers a wide variety of instructions from which to choose (add, multiply, compare, move data, and so on). CISC computers reflect the evolution of increasingly sophisticated machine languages. Computer designers, however, are rediscovering the beauty of simplicity. Computers designed around much smaller instruction sets can realize significantly increased throughput for certain applications, especially those that involve graphics (for example, computer-aided design). These computers have RISC (reduced instruction set computer) design. The RISC processor shifts much of the computational burden from the hardware to the software. Proponents of RISC design feel that the limitations of a reduced instruction set are easily offset by increased processing speed and the lower cost of RISC microprocessors.

In a single processor environment, the processor addresses the programming problem sequentially, from beginning to end. Today, designers are building computers that break a programming problem into pieces. Work on each of these pieces is then executed simultaneously in separate processors, all of which are part of the same computer. The concept of using multiple processors in the same computer is known as parallel processing. In parallel processing, one main processor examines the programming problem and determines what portions, if any, of the problem can be solved in pieces (see Figure 2-7). Those pieces that can be addressed separately are routed to other processors and solved. The individual pieces are then reassembled in the main processor for further computation, output, or storage. The net result of parallel processing is better throughput.
Computer designers are creating mainframes and supercomputers with thousands of integrated microprocessors. Parallel processing on such a large scale is referred to as massively parallel processing (MPP). These super-fast supercomputers have sufficient computing capacity to attack applications that have been beyond that of computers with traditional computer designs. For example, researchers can now simulate global warming with these computers.

Just as the brain sends and receives signals through the central nervous system, the *"" **"»"*il 4M.MAAH3 processor sends and receives electrical signals through its common electrical bus aword at a time. A word describes the number of bits that are handled as a unit within a particular computer system's bus or during internal processing.
Twenty years ago, a computer's word size applied both to transmissions through the electrical bus and to all internal processing. This is no longer the case. In some of today's computers, one word size defines pathways in the bus and another word size defines internal processing capacity. Internal processing involves the movement of data and commands between registers, the control unit, and the arithmetic and logic unit (see Figure 2-5). Many popular computers have 64-bit internal processing but only a 32-bit path through the bus. For certain input/output-oriented applications, a 64-bit computer with a 32-bit bus may not realize the throughput of a full 64-bit computer.
The word size of modern microcomputers is normally 64 bits (eight 8-bit bytes). Early PCs had word sizes of 8 bits (one byte) and 16 bits (two bytes). Workstations, mainframes, and supercomputers have 64-bit word sizes and up.

The PC's heart is its crystal oscillator and its heartbeat is the dock cycle. The crystal oscillator paces the execution of instructions within the processor. A micro's processor speed is rated by its frequency of oscillation, or the number of clock cycles per second. Most modern personal computers are rated between 300 and 600 megahertz, or MHz (millions of clock cycles). The elapsed time for one clock cycle is 1/frequency (1 divided by the frequency). For example, the time it takes to complete one cycle on a 400-MHz processor is 1/400,000,000, or 0.000000025 seconds, or
Many field sales representatives carry notebook PCs when they call on customers. This Merck Human Health-U.S. representative uses computer-based detailing to make a presentation on cholesterol reducer to a cardiologist. Computer data help her better target information to physicians. Her notebook has a 64-bit processor with a speed of 300 MHz and a RAM capacityof 1 28 MB.
The CRAY T90 supercomputer is the most powerful general-purpose computer ever built. General purpose computers are capable of handling a wide range of applications. The system is capable of crunching data at a peak rate of 64 GFLOPS (gigaflops). One GFLOP equals one billion floating point logic operations per second. Courtesy of E-Systems, Inc.
2.5 nanoseconds (2.5 billionths of a second). Normally several clock cycles are required to fetch, decode, and execute a single program instruction. The shorter the clock cycle, the faster the processor.
To properly evaluate the processing capability of a computer, you must consider both the processor speed and the word size. A 64-bit computer with a 450-MHz processor has more processing capability than does a 32-bit computer with a 450-MHz processor.

The capacity of RAM is stated in terms of the number of bytes it can store. Memory capacity for most computers is stated in terms of megabytes (MB). One megabyte equals 1,048,576 (220) bytes. Memory capacities of modern PCs range from 32 MB to 512 MB. Memory capacities of early PCs were measured in kilobytes (KB). One kilobyte is 1024 (210) bytes of storage.
Some high-end mainframes and supercomputers have more than 8000 MB of RAM. Their RAM capacities are stated as gigabytes (GB)—about one billion bytes. It's only a matter of time before we state RAM in terms of terabytes (TB)—about one trillion bytes. GB and TB are frequently used in reference to high-capacity secondary storage. Occasionally you will see memory capacities of individual chips stated in terms of kilobits (Kb) and megabits (Mb). Figure 2-9 should give you a feel for KBs, MBs, GBs, and TBs.

Inside The Computer

2-1 Data Storage: A Bit About the Bit

Going Digital

Binary Digits: On-Bits and Off-Bits

Going Digital with Compact Discs

Data in a Computer

Encoding Systems: Bits and Bytes

ASCII and ANSI

Unicode: 65,536 Possibilities

Analyzing a Computer System, Bit by Bit

RAM: Random-Access Storage

RAM TECHNOLOGY: SDRAM AND RDRAM

CACHE MEMORY

ROM, PROM, AND FLASH MEMORY

The Processor: Nerve Center

THE CONTROL UNIT

THE ARITHMETIC AND LOGIC UNIT

Buses: The Processor's Mass Transit System

Making the Processor Work

THE MACHINE LANGUAGE

THE MACHINE CYCLE: MAKING THE ROUNDS

Instruction time

Execution time

Processor Design: There Is a Choice

CISC AND RISC:

PARALLEL PROCESSING: COMPUTERS WORKING TOGETHER

Describing the Processor: Distinguishing Characteristics

Word Size: 16-, 32-, and 64-Lane Bitways

Processor Speed: Warp Speed

MEGAHERTZ: MHz

MIPS

FLOPS

RAM Capacity: Megachips