A highly efficient cloud-based architecture for large-scale STB event processing: industry article

  • Authors:

  • Bin Cao;Jianwei Yin;Shuiguang Deng;Yueshen Xu;Youneng Xiao;Zhaohui Wu

  • Affiliations:

  • Zhejiang University, Hangzhou China;Zhejiang University, Hangzhou China;Zhejiang University, Hangzhou China;Zhejiang University, Hangzhou China;iPanel.TV Inc., N. Shenzhen China;Zhejiang University, Hangzhou China

  • Venue:
  • Proceedings of the 6th ACM International Conference on Distributed Event-Based Systems
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

With the popularization and ripeness of DTV (Digital television) and IPTV (Internet Protocol television), the STB (Set-Top Box) devices are playing more and more important roles in improving the interactive TV applications. Through STB devices, millions of users can publish/subscribe their interested events when they are participating in different applications. Furthermore, it is likely that a very large number of events would be published simultaneously and processing these events under critical timing constrains is often required by lots of applications. How to subscribe and process this large amount of events in real time is rather a challenge to both academia and industry. The combination of Cloud Computing and DDS (Data Distribution Service) middleware brings us a promising way to address this issue. Cloud computing can not only provide sufficient computing capacity but also reduce the processing cost in a great lot. Besides, DDS middleware supports real-time publish/subscribe communication model and is often used in military scenario. In this paper, in order to process such a large scale and high concurrency events efficiently and real-time, we propose a cloud-based architecture for STB event processing. Besides, we discuss the implementation details of reliability and load balancing. To demonstrate the effectiveness of our approach some experiments on the prototype are evaluated.