NYNN: An in-memory distributed storage system for massive graph analysis

With the development of social networks, methodologies and approaches of computational intelligence are used for data mining and knowledge discovery regarding massive graphs generated by social networks. How to efficiently organize massive graphs in order to improve the performance of massive graph analysis is an important issue. The traditional graph data management systems are designed for general purpose but lack sufficient consideration on graph characteristics and access methods. As a result, the early systems are less suitable in scenarios of massive graph analysis. In order to solve the above problem, this paper proposes an in-memory organization system for graph data generated by social networks, and the system gives special consideration on update, random access and sparsity of massive graphs. Finally, experiments conducted on real-world social network data sets have shown that the proposed methods are superior to the industry´s advanced graph storage methods.