Von Neumann architecture and the stored program concept

Overview

Stored program concept, von Neumann bottleneck and modified Harvard architecture.

What you need to know

Answer each one before you reveal it.

Questions from this subtopic

Each one with the model answer underneath.

1. 1

   What is meant by the stored program concept?

   Model answer

   The stored program concept means instructions and data are both held in main memory. The processor fetches each instruction from memory in turn.

2. 2

   Which features define Von Neumann architecture?

   Model answer

   Von Neumann architecture uses one memory for instructions and data, one control unit, and instructions executed in sequence. A single shared bus carries both instructions and data.

3. 3

   What is the Von Neumann bottleneck?

   Model answer

   The Von Neumann bottleneck is the limit imposed by one shared bus between processor and memory. The processor sits idle while waiting for instructions or data to arrive.

4. 4

   How does Harvard architecture differ from Von Neumann architecture?

   Model answer

   Harvard architecture keeps instructions and data in separate memories with separate buses. An instruction and a data item can therefore be fetched at the same time.

5. 5

   Where is Harvard architecture typically used, and why?

   Model answer

   Harvard architecture is common in digital signal processors and microcontrollers. Predictable timing and simultaneous fetches matter more there than flexible use of memory.

6. 6

   What is a modified Harvard architecture?

   Model answer

   A modified Harvard architecture uses split instruction and data caches over a single main memory. The processor gains parallel fetching without needing two physically separate memories.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.