Understanding data flow graph for improving big data stream computing environments. (1st January 2014)

• Record Type: Journal Article
• Authors: Sun, Dawei; Fu, Ge; Liu, Xinran; Ren, Dongfeng
• Is Part Of: International journal of computing science and mathematics. Volume 5:Number 4(2014)
• Page Start: 394
• Page End: 404
• Publication Date: 2014-01-01
• Subjects: data flow graph -- big data -- stream computing -- high throughput -- high real time; Mathematics -- Periodicals; Computer science -- Periodicals; Mathematics -- Data processing -- Periodicals; 510.285
• DOI: 10.1504/IJCSM.2014.066446
• Languages: English
• ISSNs: 1752-5055

Abstract

High throughput issue is one of the major obstacles for opening up the new big data stream computing era. A high throughput big data stream computing system is need, optimising the data flow graph is an important way to offer a high throughput computing environment in big data stream computing system. In this paper, the definition of data flow graph in big data stream computing is given and the properties of data flow graph are systematically analysed by referring to the stream computing theories. A series of optimisation strategies for data flow graph in big data stream computing environments are put forward, all those strategies will greatly improve the structure of data flow graph, and provide a high throughput environment for big data stream computing.