NSU Research Contributions

Title : Big Graph Analytics

Authors : Rahman Ahsanur, Motahar Tamanna

Abstract : A number of computational frameworks/platforms have been proposed in the literature to tackle the challenges of big graph mining. These frameworks can broadly be classified into different categories based on whether they are distributed or are single-machine systems and whether they are designed for static or dynamic graphs. This chapter discusses these categories and subcategories of big graph frameworks and compares them. It highlights their relative advantages and disadvantages and discusses which use cases are best suited for each type of framework. The chapter also reviews some popular big graph platforms and discusses how to implement some important graph algorithms in these frameworks. Distributed systems are obvious choices for analyzing big graphs due to the difficulty in processing/storing them in a single machine. Vertex-centric frameworks like Pregel try to achieve extreme parallelism, even among the processes running on the vertices within the same worker.

Journal :	Volume :	Year : 2018	Issue :
Pages : 97-130	City : Boca Raton, Florida, USA	Edition : 1	Editors : Ahmed Mohiuddin, Pathan Al-Sakib Khan
Publisher : CRC Press	ISBN : 978-1138500815	Book : Data Analytics: Concepts, Techniques and Applications	Chapter : 5
Proceeding Title :	Institution :	Issuer :	Number :