The characteristics of contemporary processors, input, output and storage devices

Covers the A-Level content of OCR Chapter 1.1 (components of a computer and their uses)

Components of the Processor and Their Functions

• Arithmetic Logic Unit - all arithmetic and logical functions

• Control Unit - directs and co-ordinates all functions on the computer by sending out signals

• Registers - small memory cells operating at high speeds to temporarily store data

• Buses - parallel wires connecting two or more components

describe the:

• data bus

• address bus

• control bus

• bi-directional, data and instructions

• transmits memory addresses specifying where data needs to be sent to or from

• bi directional, transmits control signals to co-ordinate the other buses

what are the 5 types of registers and they’re functions

• PC - holds the address of the next instruction to be executed

• ACC - stores calculation results

• MAR - holds the address of a location to be read from or written to

• MDR - temporarily stores the data in the MAR

• CIR - holds the current instruction being executed

what are the factors affect CPU performance

• amount of cache (quick access memory stored on or near the ram) - reduces the distance the data has to travel, making it faster to access, typically recently or commonly used instructions

• number of cores - enables multiple FDE cycles to be carried out at once, making a machine faster

• clock speed - faster clock means more instructions can be carried out per second as it synchronises activity on the CPU, can be sped up via overclocking temporarily but can eventually slow down the CPU

what is pipelining

• the FDE takes place simultaneously, meaning that there is no ‘gaping’ in the process, stopping the CPU from being idle and wasting time

• instruction and arithmetic

what happens in the FDE cycle

• fetch - address from PC copied to mar, data copied to mdr, pc goes up by 1, mdr copied to cir

• decode - cir split into operand and opcode

• execute - the decoded instruction is executed

von Neumann vs harvard architecture

• von neumann

  • shared memory and data buses for both data and instructions causing bottleneck

  • used in most devices

  • simpler CU design

• harvard

  • separate buses for data and instructions, which are stored in different memories

  • used embedded systems

  • more expensive and complicated CU design

CISC vs RISC

• RISC (reduced instruction set computers)

  • simpler code that uses more lines and more RAM

  • more taxing for compiler t o translate

  • each instruction takes 1 cycle, so it can use pipelining

• CISC (complex instruction set computers)

  • more complex code that takes up fewer lines

  • uses assembly code built into the hardware

  • less common now, used in embedded systems

multi-core and parallel systems

• multicore cpu’s have multiple processors that can complete instructions

• parallel systems can do the same thing with one processor, but multi core is much more efficient

the GPU

a co-processor with thousands of cores working in parallel, used previously only for graphics, but now used for repetitive activities like machine learning and linear algebra

input devices

Peripheral devices that allow the user to communicate with a computer e.g. remote, game controller, mouse, keyboard

output devices

peripheral devices that make computer output readable/accessible for the user e.g. screens, vibrations etc

RAM and ROM

• Random Access Memory is used to store data and programs in the computer or device. it is volatile

• Read-Only Memory stores things like the bootstrap that is permanently required, so its non-volatile

Optical Storage

• read and written from pits and lands representing 1 and 0, drawn in by a laser

• the most portable and cheap, but very fragile and have low storage potential

• examples include dvds, cds and blu-ray

magnetic storage

• disks and platters with a read/write head, each portion of the disks gets magnetised to 1 or 0

• fragile with moving parts, but not too expensive, low cost per gb

• used in HDDs and magnetic tape (cloud storage is typically stored onto magnetic storage)

flash storage

• fast and compact, but most expensive per gigabyte

• the most widely used

• no moving parts, very durable

cloud storage

• cheap, can be accessed anywhere with a wifi connection

• information on the cloud is an abstraction of hundreds of SSDs and various other drives