Computer Architecture and Organization

• Computer architecture encompasses the design aspects of a computer system
• Computer architecture focuses on how the hardware and software interact to create a functional computing platform
• Key attributes of computer architecture include instruction set architecture, memory organization, and I/O structure

Computer Organization

• Computer organization details the hardware components and their interconnections within a computer system
• It specifies how the different parts of the system (CPU, memory, I/O) work together to implement the architectural specifications
• Organizational aspects include control signals, interfaces, memory technology, and addressing schemes
• An example of an organizational decision is whether to use a hardwired or microprogrammed control unit

Parallel Processing

• Parallel processing is a method of computation where multiple instructions are executed simultaneously
• The aim of parallel processing is to speed up computation by dividing a problem into smaller, independent parts that can be solved concurrently
• Parallel processing can be implemented at various levels, including instruction-level, thread-level, and process-level
• Common architectures for parallel processing include:
  • Multi-core processors: multiple CPUs on a single chip
  • Symmetric Multiprocessors (SMP): multiple processors sharing a common memory
  • Distributed systems: multiple computers connected via a network

Instruction Set Architecture (ISA)

• Instruction Set Architecture (ISA) defines the set of instructions that a processor can execute
• The ISA acts as an interface between the hardware and software
• Key components of an ISA include:
  • Instruction formats: the layout of bits in an instruction
  • Addressing modes: how operands are located (e.g., direct, indirect, register)
  • Instruction types: categories of instructions (e.g., arithmetic, logical, control)
  • Registers: storage locations within the CPU
• Types of ISAs:
  • Complex Instruction Set Computing (CISC): characterized by a large number of complex instructions
  • Reduced Instruction Set Computing (RISC): uses a smaller set of simpler instructions, typically executing in one clock cycle

Memory Hierarchy

• Memory hierarchy is a structure that uses multiple levels of memory storage to provide a cost-effective balance between speed and capacity
• The hierarchy typically includes:
  • Registers: fastest, smallest memory directly within the CPU
  • Cache memory: fast, relatively small memory used to store frequently accessed data
  • Main memory (RAM): larger, slower memory used to store currently running programs and data
  • Secondary Storage (Hard drives, SSDs): slowest, largest storage for long-term data storage
• Key principles of memory hierarchy:
  • Locality of reference: programs tend to access memory locations that are near recently accessed locations (temporal and spatial locality)
  • Caching: storing frequently used data in faster memory to reduce access time
• Memory hierarchy design considerations:
  • Cache size: larger caches can store more data but are more expensive
  • Block size: the amount of data transferred between memory levels
  • Replacement policy: algorithm for deciding which data to evict from the cache when it is full (e.g., LRU, FIFO)
  • Write policy: how and when data is written back to main memory (e.g., write-through, write-back)

CPU Design

• CPU (Central Processing Unit) design involves the architectural and organizational aspects of the processor
• Key components of CPU design:
  • Arithmetic Logic Unit (ALU): performs arithmetic and logical operations
  • Control Unit: fetches instructions from memory, decodes them, and controls the execution of instructions
  • Registers: store data and addresses for use during processing
  • Cache memory: small, fast memory for storing frequently accessed data
• CPU performance factors:
  • Clock speed: the rate at which the CPU executes instructions (higher is generally faster)
  • Instruction Pipeline: overlapping the execution of multiple instructions to improve throughput
  • Number of cores: multiple processing units on a single chip can execute multiple threads concurrently
  • Cache size and organization: larger and more efficient caches reduce memory access time
• CPU design considerations:
  • Power consumption: balancing performance with energy efficiency
  • Heat dissipation: managing the heat generated by the CPU
  • Complexity: balancing functionality with design and manufacturing complexity
• Instruction execution cycle:
  • Fetch: retrieve the next instruction from memory
  • Decode: determine the operation to be performed and the operands to be used
  • Execute: perform the operation specified by the instruction
  • Memory (if needed): access memory to read or write data
  • Write-back: store the result of the execution