Alleviating Datapath Conflicts and Design Centralization in Graph Analytics Acceleration

TL;DR

This paper introduces HiGraph, a graph analytics accelerator utilizing the MDP-network to reduce datapath conflicts and centralization, significantly improving throughput and scalability.

Contribution

It proposes the MDP-network as a general solution to datapath conflicts and centralization, and implements HiGraph to demonstrate practical benefits.

Findings

Up to 2.2x speedup over state-of-the-art accelerators

Average 1.5x speedup achieved

Enhanced scalability of the proposed solution

Abstract

Previous graph analytics accelerators have achieved great improvement on throughput by alleviating irregular off-chip memory accesses. However, on-chip side datapath conflicts and design centralization have become the critical issues hindering further throughput improvement. In this paper, a general solution, Multiple-stage Decentralized Propagation network (MDP-network), is proposed to address these issues, inspired by the key idea of trading latency for throughput. Besides, a novel High throughput Graph analytics accelerator, HiGraph, is proposed by deploying MDP-network to address each issue in practice. The experiment shows that compared with state-of-the-art accelerator, HiGraph achieves up to 2.2x speedup (1.5x on average) as well as better scalability.