< architecture , computability > A computer architecture conceived by mathematician John von Neumann , which forms the core of nearly every computer system in use today (regardless of size). In contrast to a Turing machine , a von Neumann machine has a random-access memory (RAM) which means that each successive operation can read or write any memory location, independent of the location accessed by the previous operation.
A von Neumann machine also has a central processing unit (CPU) with one or more registers that hold data that are being operated on. The CPU has a set of built-in operations (its instruction set ) that is far richer than with the Turing machine, e.g. adding two binary integers , or branching to another part of a program if the binary integer in some register is equal to zero ( conditional branch ).
The CPU can interpret the contents of memory either as instructions or as data according to the fetch-execute cycle .
Von Neumann considered parallel computers but recognized the problems of construction and hence settled for a sequential system. For this reason, parallel computers are sometimes referred to as non-von Neumann architectures.
A von Neumann machine can compute the same class of functions as a universal Turing machine .
[Reference? Was von Neumann's design, unlike Turing's, originally intended for physical implementation?]