Distributed Algorithms for Subgraph-Centric Graph Platforms

TL;DR

This paper introduces three novel subgraph-centric distributed algorithms for triangle counting, clustering, and minimum spanning forest, enhancing the efficiency and evaluation of large-scale graph processing on subgraph-centric platforms.

Contribution

It presents new subgraph-centric algorithms for key graph tasks, expanding the capabilities and evaluation scope of subgraph-centric graph processing models.

Findings

Enhanced performance of subgraph-centric algorithms over existing methods

Broader evaluation of graph processing algorithms and platforms

Improved scalability for large-scale graph analytics

Abstract

Graph analytics for large scale graphs has gained interest in recent years. Many graph algorithms have been designed for vertex-centric distributed graph processing frameworks to operate on large graphs with 100 M vertices and edges, using commodity clusters and Clouds. Subgraph-centric programming models have shown additional performance benefits than vertex-centric models. But direct mapping of vertex-centric and shared-memory algorithms to subgraph-centric frameworks are either not possible, or lead to inefficient algorithms. In this paper, we present three subgraph-centric distributed graph algorithms for triangle counting, clustering and minimum spanning forest, using variations of shared and vertex-centric models. These augment existing subgraph-centric algorithms that exist in literature, and allow a broader evaluation of these three classes of graph processing algorithms and platforms.