An analysis of the graph processing landscape

TL;DR

This paper provides a comprehensive overview of the diverse computational approaches, system architectures, and paradigms used in graph processing, highlighting their capabilities and limitations across various system types.

Contribution

It offers a detailed classification and analysis of graph processing systems based on multiple dimensions, serving as a valuable resource for researchers and practitioners.

Findings

Classifies graph processing systems by paradigms, system types, and communication models.

Highlights the diversity and limitations of current graph processing solutions.

Provides insights into partitioning techniques and update capabilities.

Abstract

The value of graph-based big data can be unlocked by exploring the topology and metrics of the networks they represent, and the computational approaches to this exploration take on many forms. The use-case of performing global computations over a graph, it is first ingested into a graph processing system from one of many digital representations. Extracting information from graphs involves processing all their elements globally, and can be done with single-machine systems (with varying approaches to hardware usage), distributed systems (either homogeneous or heterogeneous groups of machines) and systems dedicated to high-performance computing (HPC). We provide an overview of different aspects of the graph processing landscape and describe classes of systems based on a set of dimensions we detail. The dimensions we detail encompass paradigms to express graph processing, different types of systems to use, coordination and communication models in distributed graph processing, partitioning techniques and different definitions related to the potential for a graph to be updated. This survey is aimed at both the experienced software engineer or researcher as well as the newcomer looking for an understanding of the landscape of solutions (and their limitations) for graph processing.