Computer Architecture

Architecture of a Computer

Components of a computer:

Chassis: The chassis is the frame, constructed of metal and high impact plastic that houses all the internal components of the computer. When all the parts are assembled, the chassis is covered by a plastic case coated on the inside with conductive silver paint to reduce radio frequency emissions.

Motherboard: The motherboard is the main circuit board of a computer. The motherboard holds the computer’s CPU, some memory and most of its control circuitry.

Memory: Primary memory consists of chips mounted on their own small circuit boards that plug into special sockets on the motherboard or they can be soldered directly on the motherboard.

Power Supply: The power supply contains a transformer, which regulates the voltage level required for the computer.

Expansion Slots: An expansion slot is an internal connector, which allows a person to add additional circuits to the motherboard. The circuit is called an adapter. Such boards permit the computer to be connected to auxiliary peripheral
equipment.

Device Controllers: It is a set of chips or a circuit board that operates a piece of equipment such as a disk drive, display, keyboard, mouse or printer. Now a days, the hard disk controllers are on a removable expansion board so that they can be easily upgraded with a faster or higher capacity hard disk.

Disk Drives: A disk drive is a piece of equipment that can read and write data on magnetic disks. Most microcomputers have two kinds of disk drives. A floppy disk drive works with floppy disks, which are inexpensive flexible magnetic disks. The standard size of a floppy disk is 3 ½ inches, which has a capacity of 1.44 megabytes of programs and data. The computer also has a hard disk drive, which in our computers has a storage capacity of 4 gigabytes.

Other Components: Other components present in the computer may be sound cards, speakers,
battery pack, fan, a floppy disk connector etc.

Central Processing Unit

The computer primary deals with zeros and ones only. That means that it works on a binary number system which correspond to the on and off stage of an electronic switch.

Further, textual information in a computer is processed by associating each character with a number, which in turn is understood by the computer as per the binary system. The most prevalent code is AMERICAN STANDARD CODE FOR INFORMATION INTERCHNAGE or ASCII. In this system the no. 65 through 90 represent the capital letters from A to Z. The other characters are assigned other numbers. EXTENDED BINARY CODED DECIMAL INTERCHANGE CODE or EBIDC is the other computer-coding scheme. It has 8 bits per character. E.g. 'A' is represented by the number 193 (11000001 binary).

The actual processing of data takes place in the central processing unit. Ina microcomputer, the CPU consists of a single chip. In larger computers, the CPU may be constructed from several different chips and circuits. The major conceptual components of the CPU are:

Arithmetic and Logic Unit: ALU is a part of the CPU where calculations and logical operations are carried out. Calculations performed by the ALU involve the four elementary arithmetic operations of addition, subtraction, multiplication and division. Programs that employ these basic functions perform the other complex calculations. Logic operations performed by ALU consist of comparing or combining two numbers in some manner. This allows the computer to evaluate situations and take alternate courses of action. 1 symbolizes the “True” and 0 symbolizes The “False”.

Registers: The ALU can only manipulate one or two inputs to produce an output. The CPU’s registers are the special high-speed storage locations that temporarily hold the inputs and outputs for the ALU. In addition, registers also hold program instructions and data merely being transferred from one place to another. Registers are a part of CPU and not a part of primary storage. They are much faster, but also possess much less capacity than primary storage. Various types of registers are:

Accumulator: Accumulator holds the results of ALU computations and logic operations. Some CPU’s may have more than one accumulators.

Data Register: One or more of these registers is used to hold data just transferred to or from primary storage.

Address Register: These indicate the location of the data to be stored or retrieved from primary storage.

Program Counter: This register holds the primary storage address of the next instruction to be executed. It is sometimes considered to be part of the control unit.

Instruction Register: This register holds the information that is about to be executed.

Status register: This register contains individual bits, called flags that signal the results of logic operations or indicate other computer conditions. The arrangement of a CPU’s registers in relation to its ALU and control unit is often refereed to its architecture

An important aspect of a CPU is the size of its registers. The size of registers basically defines their word length. The sizes are also given in bits, most common sizes being 8, 16, 32, 64 and 128 bits.

Control Unit

The control unit directs the operation of the CPU by interpreting program instructions; telling the ALU to carry out those instructions, and communicating with primary storage and input and out put device controllers. Based on program instructions it takes from primary storage, the control unit causes data to be transmitted back and forth between ALU, registers, primary storage, and input/ output devices, and it tells the ALU what operations are to be performed.

Primary Storage:

Primary storage, also called main memory, is where input, program instructions, intermediate calculations, and output are held just before or after the CPU processes them. Primary storage holds programs and data only temporarily.

Functions of primary storage:

One basic function of the primary storage is to hold the program that is being executed. The other basic function of the primary storage is to hold data that are required by the program being executed. In an extensive calculation, the results of carrying out the instructions using the input data are usually twofold: there are some intermediate results, which are used in subsequent calculations; and there are final results, which are to be presented as output. Primary storage thus includes working storage as well as space for input data and the final results.

Capacity of primary storage:

One feature of primary storage that distinguishes it from secondary storage is that its contents can be transmitted very rapidly to the CPU for processing. Also, its physical size is relatively small compared to that of secondary storage since it must be close to the CPU.

RANDOM ACCESS MEMORY (RAM)

An important characteristic of primary storage is that it takes the same amount of time to retrieve a byte from any location. Since any randomly selected location can be selected in the same amount of time, this feature is known as random access memory (RAM). In most computers, the majority of primary storage can hold its contents only as long as the computer is turned on. That is the RAM chips are volatile. Some types of RAM chips are:

DYNAMIC RANDOM ACCESS MEMORY (DRAM) chips: These chips must be periodically refreshed with additional boosts of electricity.

STATIC RANDOM ACCESS MEMORY (SRAM) chips: must be constantly supplied with a tiny electrical charge.
SRAM chips require simple support circuitry and work faster than DRAM chips and are also more expensive than them.

Other memory acronyms

ROM (Read Only Memory)

Although a ROM chip is a random access device, it is loaded only once, at the factory. After a ROM chip has been manufactured, the programs and data stored on it can only be retrieved. In other words, contents of a ROM chip can never be altered.

PROM (Programmable Read Only Memory)

The contents of a basic ROM are loaded by the manufacturer and are thereafter, essentially unalterable. In contrast, the user may initially load the contents of a PROM with a special tool called a PROM burner.

EPROM (Erasable Programmable Read Only Memory)

Since the programs and data stored in ROM chips are often complex, programming errors of some kind are possible, and infact, are quite likely to occur. The EPROM can only be erased by exposing the chip to a strong ultraviolet light for several minutes.

EEPROM (Electrically Erasable Programmable Read Only Memory)

Unlike an EPROM Chip, an EEPROM chip can be erased with an electrical charge while it is still in the computer. There is a limit, however, to the number of times an EEPROM can be erased.

The BUS

The bus is a set of wires and connectors that link the CPU to primary storage. It is also the pathway that data and information travel between memory and various input, output and secondary storage controllers. The bus gives a computer flexibility to expand memory, accommodate alternate input, output and secondary storage devices, and add interfaces to control and interact with other machines.

System Clock

Directly connected to the CPU and also tied into the control lines of the bus is an essential component of any computer known as the system clock. This consists of a chip containing a crystal that vibrates, or ticks, at a certain frequency, typically several million times per second. The speed of the system clock determines how fast a computer can execute instructions. The speed is measured in clock cycles per second.

How are programs executed?

Two major phases make up the machine cycle, during which a computer executes an instruction.

The Fetch Cycle

During this phase, the control unit fetches the next instruction from memory and decodes it.

The Execution Cycle

In this phase, any necessary data is obtained and the operation is actually completed by the ALU

The Program Counter and Branching

The normal sequence of a program gets successive instructions from consecutive primary storage addresses by simply incrementing the program counter after each fetch. Branching allows the computer to deviate from this sequence by loading new address into the program counter.