Revising the classic computing paradigm and its technological implementations

TL;DR

This paper critiques the longstanding von Neumann computing paradigm, highlighting its outdated assumptions, and proposes an extended model to address modern technological limitations and improve computing efficiency.

Contribution

It revises the classic von Neumann paradigm by extending it to account for non-negligible transfer times, aligning theory with current technological realities.

Findings

Identifies performance issues caused by outdated assumptions

Highlights the importance of transfer time in modern computing

Proposes a new procedure for non-negligible transfer times

Abstract

Today's computing is told to be based on the classic paradigm, proposed by von Neumann, a three-quarter century ago. However, that paradigm was justified (for the timing relations of) vacuum tubes only. The technological development invalidated the classic paradigm (but not the model!) and led to catastrophic performance losses in computing systems, from operating gate level to large networks, including the neuromorphic ones. The paper reviews the critical points of the classic paradigm and scrutinizes the confusion made around it. It discusses some of the consequences of improper technological implementation, from the shared media to the parallelized operation. The model is perfect, but it is applied outside of its range of validity. The paradigm is extended by providing the "procedure" that enables computing science to work with cases where the transfer time is not negligible apart from processing time.