1.1 Systems architecture

https://ibaguette.com/cheatsheets/gcse/ComputerScience https://www.csnewbs.com/ocr-gcse copied all from this + a bit of smart revise terms work smarter not harder

Purpose of the CPU and computer

• computer is an electronic device that takes input, processes data and delivers output

• central processing unit

• to process data and instructions by constantly repeating the fetch-decode-execute cycle.

FDE cycle step by step

FETCH

1. Copy memory address from PC to MAR

2. Copy instruction stored in MAR address to MDR

3. Increment (increase) PC to point to the address of the next instruction, ready for next cycle

DECODE

1. Instruction in MDR decoded by CU

2. CU prepares for the next step (loading values into MAR/MDR)

EXECUTE

1. Instruction is performed + cycle repeats

2. E.g hault program, calculation operation

CPU components

• ALU (Arithmetic Logic Unit)

• CU (Control Unit)

• Cache

• Registers

ALU (Arithmetic Logic Unit)

• part of CPU where data is processed and manipulated

• Performs calculations and logical operations/comparisons

• Where decisions are made (e.g if x > 10)

• Handles bit shifting

CU (Control Unit)

• sends signals to control the flow of data around the CPU

• Control signals and timing signals are sent to ALU and other components like RAM

• decodes instructions as part of F-E cycle

Cache

• Memory inside the processor providing fast access to frequently used instructions and data

• used to temporarily store data - volatile

• split into different levels, 123

• bigger and slower to access than registers but much faster than RAM

Registers

• Small amounts of very high-speed memory in CPU

• designed for a very specific purpose

• used as a temporary storage space

• stores small amounts of data needed for processing

• Includes address of current instruction, next instruction to be executed, and results of calculations

Von Neumann architecture

• Data and instructions stored as binary in primary storage

• consists of CU, ALU, memory unit (typically RAM), inputs and outputs

• no way to know if binary held in memory is representing instructions or data by looking at it

• based on the stored-program concept

4 registers for processing - Von Neumann architecture

• PC

• MAR

• MDR

• Accumulator

Program counter (PC)

• Small memory location inside the CPU to assist with FDE cycle

• holds the memory address of the next instruction to be fetched from primary storage

memory address register

• Small memory location inside the CPU to assist with the FDE

• holds address of current instruction that is to be fetched from memory

memory data register

• Small memory location inside CPU to assist with the FDE

• holds data found at the address held in the MAR, or data which is to be transferred to primary storage

• Holds the data fetched from or to be written to the memory

accumulator

• used by ALU to hold the data being processed and the results of its calculations

Clock speed

• Measured in Hertz, e.g gigahertz - GHz

• number of cycles per second (3.2GHz = 3.2bill instructions/sec)

• measure of how quickly a CPU can process instructions

• faster clock speed = faster computer can perform FDE cycle = better performance as more instructions processed per sec

Overclocking

• Computer's clock speed is increased higher than the recommended rate (3.5)

• performs faster, but can lead to overheating and could damage machine

Underclocking

• Computer's clock speed is decreased lower than the recommended rate (3.5)

• perform slower but will increase lifespan of the machine

Cache size

• number of instructions/data that can be held in the CPU

• closer to CPU than RAM = provides data and instructions to the CPU at faster rate than RAM

• more cache memory = should have higher performance as repeatedly used instructions can be stored and accessed faster

• costly

• L1 is quickest, most expensive but has lowest capacity

• L2 is slower but holds more than L1

• L3 is slower but holds more than L2, cheapest

Core number

• Core - complete set of CPU components

• Each core is able to perform its own FDE cycle

• quad-core executes four instructions simultaneously

• More cores = higher performance as it can process more instructions at once

• BUT doubling the core doesn’t double the speed

What are the limitations of having more cores?

• If one core is waiting for another to finish processing, performance may not increase at all.

• Some software isn’t written to make use of multiple cores, so it won’t run any quicker on a multi-core computer

embedded system

• Computer system built into a larger machine to perform a specific pre-programmed task which is stored in ROM

• Uses combination of hard/software

• Adv

  • small

  • less power consumption and cheaper to make as their processors are not as powerful

  • more efficient at doing their task than general purpose computer