Von Neumann Architecture
The first computers had fixed programs, and changing a program meant physically rewiring or redesigning the machine. This made repurposing a computer difficult, expensive and time-consuming.
In 1945 the mathematician John von Neumann described the design for a stored program computer, where the programs themselves could be stored in memory, not just data.
What was a major limitation of the first computers when changing programs?
Stored Program Computer
A stored-program computer (aka Von Nuemann architecture) is a type of computer that stores program instructions in its memory alongside data. This design allows the computer to fetch, decode, and execute instructions sequentially, making it more flexible and capable of running different programs without needing hardware changes.
Key Features
Memory Storage of Instructions â Programs and data are stored in the same memory.
Sequential Execution â The CPU fetches and executes instructions one by one.
Reprogrammability â A computer can run different programs without modifying its hardware.
Components of the Von Neumann Architecture
The Von Neumann architecture consists of several key components, each with its own specific function. These components are:
- Central Processing Unit (CPU)
- Memory (RAM)
- Control Unit (CU)
- Arithmetic Logic Unit (ALU)
- Input/Output (I/O) System
- Registers
- Data Bus
- Address Bus
- Program Counter (PC)
Central Processing Unit (CPU)
The CPU is the brain of the computer and performs all the processing tasks. It executes instructions stored in memory, performs arithmetic and logical operations, and controls the flow of data within the system.
Memory (RAM)
Memory is used to store both data and instructions that the CPU needs to access during program execution. In the Von Neumann architecture, memory is a shared resource, meaning that data and instructions are stored in the same addressable memory space. This is often referred to as the "stored program concept."
Which component of Von Neumann Architecture is responsible for storing instructions and data?
Control Unit (CU)
The control unit is responsible for fetching instructions from memory, decoding them, and then executing them. It manages the operation of the CPU, including the sequencing of instructions and controlling data flow.
What is the main purpose of the control unit in Von Neumann Architecture?
Arithmetic Logic Unit (ALU)
The ALU is responsible for performing arithmetic and logical operations, such as addition, subtraction, multiplication, division, and comparisons. It carries out these operations based on the instructions provided by the CPU.
Which component of Von Neumann Architecture is responsible for performing calculations and logical operations?
Input/Output (I/O) System
The I/O system facilitates communication between the computer and external devices, such as keyboards, monitors, storage devices, and network connections. It manages the input and output of data between the CPU and these peripherals.
Registers
Registers are small, high-speed storage locations within the CPU that are used for temporary data storage and manipulation. They store operands, intermediate results, and memory addresses, enabling faster access than main memory.
System Busses
The system busses are critical components in computer architecture that facilitates communication among the major components of a computer system, such as the CPU, memory, and I/O devices.
The system bus typically consists of three main types of buses:
- Data Bus
- Address Bus
- Control Bus
What is the primary function of the system bus in a computer?
Data Bus
The data bus is a bi-directional communication pathway that allows data to be transferred between the CPU, memory, and input/output devices. The data bus can carry both instructions or data, depending on the instruction currently being executed.
It carries binary data (0s and 1s) between these components.
Which bus is responsible for carrying actual data between the processor and memory?
Address Bus
The address bus is uni-directional communication pathway, but it is used specifically for transmitting memory addresses from the CPU to the memory.
It enables the CPU to specify the location in memory where data or instructions are to be read from or written to.
Addressable Memory
Every memory location has a unique binary address, which the CPU sends along the address bus. The width of the address bus, the number of lines n, decides how many different addresses exist.
- Each line carries one bit, so n lines give 2n possible addresses.
- Adding one more line doubles the addressable memory.
- Total addressable memory = 2n à the size of one location (often one byte).
A CPU has a 16-bit address bus. How many memory locations can it address?
Addressable Memory Worked Examples
To find how much memory can be addressed, work out 2n, then multiply by the size of one location.
- 20-bit address bus, 1 byte per location: 220 = 1,048,576 bytes = 1 MiB.
- 32-bit address bus, 1 byte per location: 232 bytes = 4 GiB.
- 24-bit address bus, 2 bytes per location: 224 Ã 2 bytes = 32 MiB.
Leaving the answer as an expression, such as 216 Ã 2 bytes, still shows the method clearly.
Control Bus
The control bus sends signals which control the flow of data along the data and address busses. These control signals include commands for reading from or writing to memory, initiating fetch and execution cycles, and managing the flow of data and instructions within the computer.
Whilst most of the control signals flow one way (e.g. clock/read-write signals) such signals can be sent the other way (e.g. device interrupt signals). This means that the control bus is bi-directional.
The control bus is bidirectional, because it carries signals both from the CPU to other components and from components back to the CPU.
CPU â memory/I/O devices:
Signals like READ, WRITE, or MEMORY ENABLE tell devices what operation to perform.
Memory/I/O â CPU:
Signals like READY or ACKNOWLEDGE tell the CPU that the device has completed the operation and is ready for the next instruction.
What is the primary role of the control bus in a computer system?
Program Counter (PC)
The program counter is a special register that keeps track of the address of the next instruction to be fetched from memory. It is incremented as instructions are fetched and executed, allowing the CPU to follow the sequential order of a program.
Benefits and Drawbacks of Von Neumann Architecture
Storing programs and data in one shared memory makes a computer flexible, but it also limits its speed.
Benefits
- A new program can be loaded without changing the hardware.
- One memory and one set of buses keep the design simpler and cheaper.
- Memory can be shared freely between programs and data.
Drawbacks
- Instructions and data share one bus, so they cannot be fetched at the same time.
- A program error could overwrite instructions held in memory.
Which of these is a benefit of von Neumann architecture?
The Von Neumann Bottleneck
The von Neumann bottleneck is the limit on performance caused by instructions and data sharing one bus to memory. The CPU can process data much faster than memory can supply it, so it often has to wait.
- Only one item, an instruction or a piece of data, can travel at a time.
- A faster processor makes the waiting worse, not better.
- Cache memory, wider buses and pipelining reduce the effect by keeping the CPU supplied.
Von Neumann Architecture
Von Neumann Architecture
Complete! Ready to test your knowledge?
Introduction to Von Neumann Architecture
- Von Neumann Architecture
- Stored Program Computer
- Components of the Von Neumann Architecture
- Central Processing Unit (CPU)
- Memory (RAM)
- Control Unit (CU)
- Arithmetic Logic Unit (ALU)
- Input/Output (I/O) System
- Registers
- System Busses
- Data Bus
- Address Bus
- Addressable Memory
- Addressable Memory Worked Examples
- Control Bus
- Program Counter (PC)
- Benefits and Drawbacks of Von Neumann Architecture
- The Von Neumann Bottleneck
Topic Tests
Videos
- Von Neumann Architecture
- Von Neumann Architecture