GCSE Edexcel CS: Hardware (3.1)

RAM

Random Access Memory - temporary place to store material that works quickly. Volatile: erased when computer turns off. Directly accessible to CPU

What does the random in random access memory mean?

Any data chosen at random would be equally quickly accessible as any other data

Von Neumann Architecture

A processor where data and instructions are stored in the same memory and accessed via buses. It is a computer architecture that allows a machine to be programmable - to store programs instead of just doing one job

Registers

Registers are immediate access, small amounts of fast temporary memory within the processor

MDR vs MAR

MDR (Memory Data Register - holds data/program instruction when it is fetched from memory or data that is waiting to be written to memory)

MAR (Memory Address Register)

Problem and solution of bottlenecking in Von Neumann Architecture

Data and Program instructions are stored in the same memory (Harvard architecture stores in different locations). This results in bottlenecking, where the CPU is waiting for the data bus as it is busy carrying both program instructions and data. To solve this, you can use cache: a small amount of high-speed random-access memory (RAM) built directly within the processor. It is used to temporarily hold data and instructions that the processor is likely to reuse. This allows for faster processing as the processor does not have to wait for the data and instructions to be fetched from the RAM.

Address bus

Carries addresses from the processor to other components: uni directional.

Data bus

Carries data/instructions in both directions between the CPU and other devices: bi-directional.

Control bus

Carries control signals in both directions between the CPU and other devices: bi-directional.

What does more wires on buses mean

Number of wires is the bandwidth. Each wire represents a 1 or 0, so more wires can represent more memory addresses for the address bus

The size of the address bus determines how much memory the CPU can address directly. 64 bits can access 2^64 = 18.4 exabytes of memory, and 32 bits can access 2^32 = 4GB of memory.

What is the instruction set

Set of all operations that can be executed by a processor

Fetch-Decode-Execute Cycle definition

The process that the CPU uses to retrieve and execute instructions

Fetch

Control bus sends a read signal to the RAM (as well as carrying the clock pulses). The value in the program counter is placed in the Memory Address Register (MAR). A one-directional address bus carries the value of the MAR to the RAM and a data bus carries the value back to the CPU. The value is placed in the Memory Data Register (MDR), which is then placed in the instruction register as it is an instruction. Now the program counter (holds the address of the memory location storing the next instruction) is incremented (by the control unit) to fetch the next instruction.

Decode

The control unit decodes the value in the instruction register (a binary value) into opcode and operand. The first is the instruction and the second is data, or address value (e.g., load 6). The instructions are from a specific instruction set (set of all operations that can be executed by a processor)

Execute

Executes instruction

Example instruction #1:

The code says to LOAD 6, so the control units sets the Memory Address Register to address 6 and the CPU sends an address bus to the RAM for address 6. The RAM sends a data bus back and the value at address 6 is stored in the Memory Data Register (as it's not an instruction, you don't put it in the Instruction Register) and then copied into the Accumulator.

Example instruction #2:

SUB 4 would mean to subtract whatever is at location 4 from what's in the accumulator. This would require the use of the ALU. Any arithmetic/logical operations are performed by the ALU (Arithmetic Logic Unit), and the results are in a special purpose register called the accumulator.

Example instruction #3:

If you then wanted to transfer that result in the Accumulator back into RAM, the MAR would be set to the target address and MDR would be set to whatever value is in the Accumulator.

Clock

In one clock tick, FDE cycle happens once

Carried by control bus

For GCSE the clock is located on the CPU (in reality a system clock is on the motherboard, and the CPU clock multiplies up the system clock).

Clock speed

How many instructions the CPU can process each second, measured in Hz

Overclocking

Running the CPU at a speed faster than the manufacturer recommends (need to control heat produced or will cause damage)

Cores

Cores are processing units within the CPU which can carry out their own FDE cycles

Volatile memory

The entire contents of this type of memory are erased when you turn off the computer

Non-volatile memory

Memory that can retain the stored data even when not powered

BIOS & its function (including where it's stored)

Basic Input/Output System

When your computer switches on, the BIOS is the first to operate. It is fetched from ROM (small chip of read only memory - can be overwritten with a firmware update) to RAM, where the CPU can use it. The BIOS switches everything on and checks all hardware is working. It then instructs to load the Operating System from secondary storage to RAM and the CPU executes this to start the computer.

ALU - Arithmetic and Logic Unit

Does all mathematical calculations and makes all logical decisions

Control unit

The part of the central processing unit that controls the flow of information through the processor.

Magnetic Storage description

The HDD has a stack of metal platters that spin at high speed. The platter has a magnetic coating on both sides. The disc spins continuously and the drive head writes data by changing the surface to N & S. A change in polarity is a 1 and the same polarity is a 0. There are concentric tracks which are subdivided into sectors

E.g., magnetic tape, floppy disk

Magnetic storage advantages

Cheapest per gigabyte, therefore high capacity

Magnetic storage disadvantages

Not fast due to moving parts, but faster than optical

Will slow down and eventually break over time

Easily damaged

Optical Storage description

CDs, DVDs, Blu-rays (double layer blue ray 50GB) - small capacity, cheap, good for distribution

Lands and pits are burned into the disk

Photodiode detects reflections of surface, with a change between pit and land = 1 and everything else = 0.

Types:

- ROM

- R (write to once)

- RW (read or write)

Optical storage advantages

Only good for distribution - cheap, easy to produce, durable, lightweight, easy transport

Optical storage disadvantages

Slow

Small capacity

Can lose data when scratched

[Now becoming obsolete due to streaming services and cloud backups, so modern laptops often don't have optical drives]

Solid State storage description

NAND Flash memory cells trap charge (electrons) in pools which stays inside when the computer is off. To read, a small voltage is applied, and if the electron pool is empty, a 1 is read out of the transistor. Discharged = 1, charged = 0

Advantages of solid state storage

Very fast

Uses less electricity

Silent ∵ no moving parts

No need to defragment ∵ no moving parts

Small + compact

Disadvantages of solid state storage

Expensive

Less number of read/writes (because transistors break down, defragmentation = v. bad for SSDs)

Electrical charges leak away so data degrades faster

Virtual memory

Secondary storage is not volatile - can store OS and memory.

When RAM is full, a part of secondary storage is allocated to virtual memory, which functions as RAM. The CPU guesses which data isn't needed currently and puts it in virtual memory. If you then use that data, it takes time to load it into RAM and replace it with another piece of data. In this case, adding more RAM is beneficial.

Embedded System & examples

A computer system with a dedicated function, often part of a larger device. E.g.: insulin pump, burglar alarm, washing machine

The internet of things (IoT)

A network of physical objects that use embedded systems and wireless to collect and exchange data w/o human interaction. E.g., google home, thermostats, monitors & trackers

General-purpose computer

a computer that can run many different programs (e.g. a smartphone)

Disadvantages of embedded system

Lack flexibility

Harder to update software

Advantages of embedded system

More efficient at their task

Cheaper and smaller to build/design

Require less batteries & less processing power.

Embedded system's alternative to secondary storage

Embedded systems often don't use secondary storage - instructions needed to run them are stored in ROM (as firmware) and any user data is held in RAM.