Article

Von Neumann and Harvard Architectures

Wednesday, 10 December 2025

System bus is a generic term for any buses.

Von Neumann

Covered in Structure and Function of the CPU.

  • Shared memory space for instructions and data
  • Instructions and data are stored in the same format
  • A single control unit or processor follows a linear fetch, decode, execute cycle
  • One instruction at a time
  • Registers are used as fast access to instructions and data

Von Neumann Overview

Harvard

  • Instructions and data are stored in separate memory units
  • Each has its own bus
  • Reading and writing data can be done at the same time as fetching an instruction
  • Used by RISC processors

Harvard Overview

Contemporary

Both Von Neumann and Harvard are examples of older CPU architectures, having originated in the 1940s. More modern machine architectures are known collectively as “contemporary” architectures.

Specifics are not considered explicitly in the exam specification, it is likely to be asked open questions about how they differ from a pure Von Neumann architecture.

SIMD (Single Instruction Multiple Data)

Parallel processing is where a processor carries out a single instruction on multiple data items at the same time. This is often used by graphics processing units (GPUs).

MIMD (Multiple Instruction Multiple Data)

A version of SIMD where multiple instructions are carried out on multiple data items across several cores.