XGDBench: A benchmarking platform for graph stores in exascale clouds

Abstract

In recent years many online graph database service providers have started their operations in public clouds due to the growing demand for graph data storage and analysis. While this trend continues to rise there is little support available for benchmarking graph database systems in cloud environments. In this paper we introduce XGDBench which is a graph database benchmarking platform for cloud computing systems. XGDBench has been designed to operate in current cloud service infrastructures as well as on future exascale clouds. We extend the famous Yahoo! Cloud Serving Benchmark to the domain of graph database benchmarking. The benchmark platform is written in X10 which is a PGAS language intended for programming future exascale systems. We describe the architecture of the XGDBench focusing on its importance for exascale clouds. We did a cluster analysis to compare the realistic nature of XGDBench data generator and saw that the community structures of synthetic graphs produced by XGDBench outperforms RMAT. We also conducted performance evaluation of four famous graph data stores AllegroGraph, Fuseki, Neo4j, an OrientDB using XGDBench on Tsubame 2.0 HPC cloud environment. © 2012 IEEE.