Computer architecture

A New Computational Model for Real Gains in Big Data Processing Power

Big data and high performance computing are seen by many as important tools that will be used to advance science. However, the computational power needed for this promise to materialize far exceeds what is currently available. This paper argues that the von Neumann computational model, the only model in everyday use, has inherent weaknesses that will prevent computers from achieving the envisaged performance levels. First, these weaknesses are explored and the properties of a computational model are identified that would be required to overcome these weaknesses.

Parallel Ordered-Access Machine Computational Model and Architecture

The article presents the new computational model which we name the parallel ordered-access machine because of its base – the parallel ordered-access memory. It also describes the computer architecture which implements proposed computational model and owing to this does not have such a limitation as the memory wall and provides parallel conflict-free memory access. The efficiency of the proposed ordered-access machine computational model is evaluated and an example of its implementation is presented.