HGraph: Parallel and Distributed Tool for Large-Scale Graph Processing

Graph are ubiquitous because the fields of application are varied. Well-known examples are social networks, biological networks and path-finding in road networks. Real-world graphs processing is very challenging because of 4V characteristics related to big data. They are huge to process them on single-node and the time complexity is exponential. Unfortunately, due to the lack of research, only a few systems are able to ensure the storage and quick processing of large-scale graphs. In this paper, we propose HGraph, a parallel and distributed tool which handles large-scale graphs. HGraph is build on top of Hadoop and Spark frameworks. The proposed tool provides high scalability and is adapted to easily implement algorithms for various graph problems. Experimental tests performed on real-world graphs showed that HGraph is reliable and achieves significant gain time over the state of the art of graph processing systems.