Von Neumann architecture

Von Neumann architecture is the name given to a design model for how a computer works. Most modern computers use this. It was created by mathematician John von Neumann in the 1940s.
Key ideas in Von Neumann architecture:
- Instructions and data are both stored in the same main memory (RAM).
- The CPU uses a single bus system to read or write either instructions or data.
- The computer processes instructions one at a time in the fetchexecute cycle.
There are 4 key registers that are used in Von Neumann architecture:

Register	Function
Memory Address Register (MAR)	The MAR holds the memory address of the data or instruction that is to be fetched or stored. For example, if the CPU needs to fetch an instruction from address 104, it will place 104 in the MAR
Memory Data Register (MDR)	The MDR temporarily holds the data that is being fetched or stored in memory. For example, If the CPU is reading from address 104, the data from that address goes into the MDR before it’s used.
Program Counter (PC)	The PC stores the address of the next instruction to be fetched. After the instruction is fetched, the PC is usually increased by 1 (so it points to the next instruction in memory). If a jump instruction is used (e.g. in loops), the PC might change to a completely different address.
Accumulator	The accumulator stores intermediate results of calculations carried out by the ALU. For example, if you’re adding several numbers together, the result of each calculation is stored here before moving on to the next. Without it, the CPU would have to keep writing results to main memory, which would be much slower.

Sign up to unlock the rest of this topic — plus every other topic, video, flashcard set and eBook guide in your SnapRevise+ subscription.

Cancel anytime — instant access the moment you sign up.