Uniting Control and Data Parallelism: Towards Scalable Memory-Driven Dynamic Graph Processing

TL;DR

This paper proposes a novel memory-driven, highly parallel computing system that unifies control and data parallelism to efficiently process irregular, dynamic graph workloads, using a message-driven programming model called diffusive computation.

Contribution

It introduces a new scalable, memory-centric computing architecture and a diffusive computation model that combine control and data parallelism for dynamic graph processing.

Findings

Demonstrates effectiveness with SSSP and Triangle Counting problems

Proposes a highly parallel non-von Neumann architecture

Introduces a message-driven programming model for irregular workloads

Abstract

Control parallelism and data parallelism is mostly reasoned and optimized as separate functions. Because of this, workloads that are irregular, fine-grain and dynamic such as dynamic graph processing become very hard to scale. An experimental research approach to computer architecture that synthesizes prior techniques of parallel computing along with new innovations is proposed in this paper. We establish the background and motivation of the research undertaking and provide a detailed description of the proposed omputing system that is highly parallel non-von Neumann, memory-centric and memory-driven. We also present a message-driven (or even-driven) programming model called "diffusive computation" and provide insights into its properties using SSSP and Triangle Counting problems as examples.