Pipelining

• Von Neumann is a serial / sequential architecutre
  • While an instruction is being fetched, e.g. the ALU is idle
  • But each part of the Fetch-Decode-Execute cycle uses a different part of the CPU
• Pipelining is a way to make the Fetch-Decode-Execute cycle more efficient
  • For example, the architecture allows the next instruction to be fetched at the same time as the ALU is performing a calculation
  • Pipelining is sometimes divided into an instruction pipeline and an arithmetic pipeline