Memory Implementation

Bus

A bus is used to transmit information between components of a computer. Note that this bus is the same as the bus in USB (Universal Serial Bus).

Von Neumann Architecture

In a Von Neumann machine, Program and Data (Information) are stored in the same memory.

Harvard Machine

In a Harvard machine, separate memories exist for Program and Data.

Comparison

A Von Neumann machine has a single entry point to the memory, which is shared by Data and Instructions. This causes what is known as a Von Neumann bottleneck, that reduces speed. Meanwhile, Harvard memory has two buses, which allows for faster speed; but the fixed division can lead to potential wastage of memory

Von Neumann architecture allows us to treat instructions as data. This is very important for us, because it allows us to alter the code being run by our machine as if it were data. This allows computers to be (conveniently) programmable. This is why most general purpose computers use Von Neumann architecture.