Authors: Kok-Leong Ong; Andrzej Goscinski; Yuzhang Han; Peter Brezany; Zahir Tari; Lei Yan
Addresses: La Trobe Business School, La Trobe University, Australia ' School of Information Technology, Deakin University, Australia ' Department of Computer Science, Duke University, Durham, North Carolina, USA ' Research Group Scientific Computing, Faculty of Computer Science, University of Vienna, Nordbergstrasse 15/C/315, Vienna, Austria ' School of Computer Science, RMIT University, Australia ' Spatial Information Integration and its Applications Laboratory, Peking University, Beijing, China
Abstract: Stream-oriented applications account for one of the major types of today's computing practices. They deal with high-speed data streams. When it comes to stream-oriented distributed systems, some challenges arise: for one thing, processing capability of a single compute node might become the bottleneck of the entire system. For another, transmitting streams among physically dispersed nodes could incur both communication overhead and instability of a distributed application. In this paper, we discuss the potential of using cloud technology to respond to various challenges faced by stream-oriented distributed systems. We introduce our idea of a special cloud, the scientific streaming cloud (SSC), which supports stream management in distributed environments. We also present our design of a language and an algorithm which are used on the SSC to describe and optimise stream-related tasks, respectively. Furthermore, we demonstrate our prototype implementation and a real-world application of the SSC.
Keywords: cloud computing; distributed systems; cloud application management; data streams; data stream management.
International Journal of High Performance Computing and Networking, 2016 Vol.9 No.3, pp.171 - 189
Available online: 13 Apr 2016 *Full-text access for editors Access for subscribers Purchase this article Comment on this article