Processors: ALU, CU, and the FDE Cycle

Processors: ALU, CU, Registers, and Buses

• A processor, the "brain" of a computer, performs calculations and manages data.
• The Arithmetic Logic Unit (ALU) performs arithmetic and logical operations.
• The Control Unit (CU) fetches instructions, decodes them, and controls the execution of operations.
• Registers are high-speed storage locations within the processor holding data.
  • Examples include general-purpose registers and special-purpose registers like the program counter and accumulator.
• Buses are communication pathways connecting different components.
  • Data buses transfer data, address buses specify memory locations, and control buses send control signals.
  • These buses enable flow of information between components.

Fetch-Decode-Execute (FDE) Cycle

• The FDE cycle is the fundamental process a processor uses to execute instructions.
• Fetch: The CU fetches the next instruction from memory.
  • Instruction's address is in the program counter.
• Decode: The instruction is broken down into smaller parts to understand what operations to perform.
• Execute: The ALU performs the necessary operations.
  • Data is retrieved from registers or memory, and the result is stored in a register or memory.

CPU Performance

• CPU performance is measured by various factors.
  • Clock speed (in GHz) which is the frequency at which the processor's internal clock ticks.
  • Number of cores (multi-core processors can process multiple instructions concurrently).
  • Cache memory (higher cache capacity can reduce read times from RAM).
  • Instruction set architecture (ISA): the instructions a CPU understands. Complex Instruction Set Computer (CISC) vs Reduced Instruction Set Computer (RISC).
  • Instruction pipeline depth (number of stages).
• Higher clock speed, more cores, and larger cache typically improve performance.
• CPU, memory, and I/O speed interact to impact overall system speed.

Pipelining

• Pipelining is a technique used to improve the performance of a processor.
• It overlaps stages of the instruction cycle which allows different instructions to be processed concurrently.
• A pipeline consists of several stages that execute in sequence.
  • An instruction might be in a different stage of execution than another instruction.
• Improved throughput, but latency may increase due to longer instruction flow length and potentially more complex circuitry.

Computer Architecture

• Architecture defines the logical structure of a computer system.
• Different architectures exist, each with various design choices.
• Underlying components and operations are determined by the architecture.
• Important design considerations include instruction complexity, parallelism and memory access.
• Examples:
  • Von Neumann architecture is the traditional way memory and instructions reside in the same address space.
  • Harvard architecture separates the memory spaces for instructions and data.
• Various factors affect architectural choices, including cost, performance, and design constraints.