C1 Structure & Function Of The Processor And Types Of Processors Flashcards
What is a control unit?
Coordinate other components activity
What is the ALU?
The problem solving part of the CPU
What is a register?
Block of memory that stores instructions
What are buses?
Series of connectors transfer signals between internal components
What is MAR?
Holds the address in memory and tell where to fetch/store data from/to
What is a hard drive?
Type of data storage device used in laptops and desktop computers
What is a memory unit?
Stores currently running programs and data
How do the number of cores affecting the CPU performance?
- each core process different instructions at the same time with own FDE cycle
- however software may not take full advantage of all the processors available
How does the clockspeed effect the CPU performance?
- regular on/off signal syncs the operations of processor components
- actions each take a fixed number of cycles to complete
How does cache memory affect the CPU performance?
- level 1 cache split into instruction cache and data cache so that they can be fetched simultaneously
- more cache = more likely data won’t be fetched from RAM
- already loaded into super fast cache memory where they can be retrieved quickly
What is pipelining?
One instruction is being fetched while the next one is being decoded and the other executed
What are the limitations to pipelining?
You can’t predict which instruction will be fetched next
What is branch prediction?
A digital circuit try to guess which way branch will go before known definitively
- essential part of modern CPU
- hardware: BPU
- BPU predicts which branch will be chosen with high confidence given current encoded instruction
- where significant improvement comes from from in latest processors
What is Von Neumann Architecture?
Instructions and data stored in common main memory and transferred using single shared bus
What are the advantages to Von Neumann?
- simplified design of CPU
- data from memory and devices accessed in same way
- all generation purpose computers are VN principle
What are the uses for Von Neumann?
- PC, laptops and servers
- data and instructions share the same memory and use the same word length
- one bus = bottleneck
What is Harvard Architecture?
Separate buses for data and instructions so the data and instructions are no longer competing for the same bus
Uses of Harvard architecture?
- different word length use for data and instructions
- specialised embedded systems and digital signal processing
- control unit with 2 buses = complex and costly
What is in the fetch phase in the FDE cycle?
- Address of next instruction copy from PC ago MAR
- Instruction in MAR copy to the MDR
- PC contents increment by 1
- MDR content copy to CIR
What is involved in the decode stage of FDE cycle?
- Instructions in CIR decoded
- Split into opcode and operand which can determine the instruction type and additional data fetched from memory
- Pass to ACC
What happens in the execute phase of the FDE cycle?
- Instruction is executed and the results held in ACC or stored in memory
What is involved in the systems bus?
The system of the data, control and address bus
What is data bus?
Transport data between components
What is the control bus?
Carries the commands
What is the address bus?
Only goes from memory to CPU and transmit memory address specify where data sent or retrieved from
What is RISC processors?
- simple instructions
- fewer instructions
- fewer addressing modes
- only load and store instructions can access memory
- processor hardware more simple
What is CISC processors?
- complex instructions ( normally made of simple instructions)
- large range of instructions
- many addressing modes
- many instructions access memory
- processor more complex
What are parallel systems?
- 2 processors work together to perform single task
- task split into small sub tasks
- any task can be processed by any of the processors
- decreased time takes to execute programs, software specifically written to take advantage of multi core system
What is MISD?
- multiple processors
- different set instructions on some set of data
What is SIMD?
- multiple processors follow same set of instructions
- process different data concurrently
What is MIMD?
- multiple processors
- process instructions independently of others
- most common is desktop computers
What are co-processors?
Like parallel processing but the additional processor responsible for carrying out specific task
What are the advantages to multicore systems?
- more jobs done in short amount of time
- task shared means the workload is reduced in individual processors
What are the disadvantages to multicore systems?
- difficult to write programs for multi core systems
- not all tasks can be split across multiple processors
What are GPUs?
-Type of processor
- perform complex calculations to render graphics
- located on a separate graphics card
- run faster
- cost efficient on CPUS
What are the uses of GPUS?
- medical and scientific research
- oil and gas exploration
- financial modelling
What is Little Man Computer?
Simple architecture designed to help understand concept of machine code and instruction sets
