Unit 1.1 - Systems Architecture
No description yet. This curated deck is ready for daily study.
Card preview (24)
Purpose Of The CPU
Continuously perform the FDE cycle (Fetch, Decode, Execute) Fetch Data & Instructions form Memory, Decodes and executes instructions
CU (Control Unit)
Controls the way data flows around the CPU
PC (Program Counter)
Stores the address of the data / instruction to be processed NEXT.
MAR (Memory Address Register)
Stores the address of the data / instruction to be processed (Fetch/read)
MDR (Memory Data Register)
Stores the data / instruction to be processed (fetch/read).
ALU (Arithmetic Logic Unit)
Performs Arithmetic Calculations & Logical Operations
ACC (Accumulator)
Stores the result of calculations by the ALU
IR (Instruction Register)
Stores instruction currently being Processed
Registers
Temporary store of data on the CPU (While the CPU performs the FDE Cycle it will shift data in and out of the registers such as the MAR, MDR, PC, ACC, IR)
Purpose of Cache
Stores frequently used Instructions & Data More Cache = More instructions & data stored less FDE Cycles used accessing main memory so programs run faster.
Clock Speed
The number of FDE cycles the CPU can perform per second Faster Clock speed = Faster FDE Cycles so more FDE cycles in a given time.
CPU Cores
Multiple CPUs sharing the workload More Cores = More FDE cycles can be completed in a given time. Some tasks /software cannot be split/run across multiple cores.
An Embedded System
A computer system that is built into a product that performs a dedicated function.
Embedded Systems - Features
• Smaller than a PC. • No unnecessary features. • Usually automated. • Software stored in ROM needs firmware update to change.
Embedded Systems - Examples
Microwave, Washing Machine, Dishwasher, Watch, Sat Nav
Instruction Set
Set of commands written in machine language that the CPU understands
Opcode
Part of the instruction that tells the CPU What to do
Operand
Part of the instruction that is the Value or Memory location Where to do it
Von Neumann Architecture
Instructions are fetched, decoded and executed one at a time. Instructions and data are held together in the same memory space.
Fetch (FDE Cycle)
The CU sends the contents of the PC to the MAR The CU then sends READ command on the Control Bus and Address of the data needed on the Address Bus. The contents of Main Memory are sent via the data bus into the MDR. The CU copies the data from the MDR to the IR The PC is incremented so that it points to the next instruction to be executed.
Decode (FDE Cycle)
The CU decodes the instruction using an Instruction Set The instruction is separated into the Opcode & Operand
Execute (FDE Cycle)
The function of the instruction is performed. Move data to and from memory. If the instruction involves arithmetic or logic, the ALU is utilized. The ALU may utilize the ACC if it needs to store data to perform the calculation
BUS
Allow data to travel between the components, memory & other hardware.
Data Bus
Used to send the data / instructions needed from/to memory. Bi-Directional (both ways).