unit 2: hardware 22

unit 2: hardware 22

1. CPU

   • Von Neumann architecture
   • registers
     • Program Counter
       • stores the address of the next instruction to be fetched
     • Current Instruction Register
       • stores the instruction currently being executed
     • Accumulator
       • stores values currently being processed
     • Memory Address Register
       • stores the address to be read/written in main memory
     • Memory Data Register
       • stores the data to be read/written to main memory
   • ALU
     • performs mathematical operations
   • Control Unit / clock
     • decodes the program instruction in the CIR, selects machine resources and operations, and coordinates activation of those operations
   • Input & Output devices
     • used to allow the user to interface with the CPU
       • e.g. keyboard, mouse, speakers, monitor
   • fetch-execute cycle
     • 1) The memory address in the PC is copied to the MAR
       • 2) The instruction stored in the memory location contained in the MAR is transferred to the MDR
         • 3) The instruction now in the MDR is copied to the CIR
           • 4) The PC is incremented, ready for the next instruction
             • 5) The instruction in the CIR is decoded and executed by the CU, potentially making use of the ALU

2. speed of a CPU

   • Clock speed
     • the faster the clock speed, the more instructions can be fetched and executed in a second
       • more power is needed for higher clock speeds, produces more heat and affects battery life - excessive heat can damage a processor
         • clock speed is measured in Gigahertz (GHz)
   • Number of cores
     • A core is a complete processing unit, with its own ALU, Control Unit and Registers
       • Having multiple cores allow several instructions to be fetched and executed at the same time, resulting in a faster CPU - however instructions that must be executed in order are not effected
         • Multiple cores on a CPU increases the cost of manufacturing
   • Cache size
     • Cache memory is extremely fast memory, used to store frequently used instructions and data
       • more cache memory improves performance as more instructions can be stored, meaning fewer instructions need ti be fetched from RAM, which is slower
         • 3 levels of Cache:
           • L1 > 2-64KB > smallest and fastest
             • L2 > 256-512KB > if instructions are not present in L1, L2 us searched > some latency due to increased size
               • L3 > 1-8MB > instructions are searched for in here if not present in L1 and L2. Largest and slowest cache but still faster than RAM

Images and files in the map