Detailed performance and waiting-time predictability analysis of scheduling options in on-demand video streaming

  • Authors:

  • Mohammad A. Alsmirat;Nabil J. Sarhan

  • Affiliations:

  • Electrical and Computer Engineering Department, Media Research Lab, Wayne State University, Detroit, MI;Electrical and Computer Engineering Department, Media Research Lab, Wayne State University, Detroit, MI

  • Venue:
  • Journal on Image and Video Processing - Special issue on selected papers from multimedia modeling conference 2009
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The number of on-demand video streams that can be supported concurrently is highly constrained by the stringent requirements of real-time playback and high transfer rates. To address this problem, stream merging techniques utilize the multicast facility to increase resource sharing. The achieved resource sharing depends greatly on how the waiting requests are scheduled for service. We investigate the effectiveness of the recently proposed cost-based scheduling in detail and analyze opportunities for further tunings and enhancements. In particular, we analyze alternative ways to compute the delivery cost. In addition, we propose a new scheduling policy, called Predictive Cost-Based Scheduling(PCS), which applies a prediction algorithm to predict future scheduling decisions and then uses the prediction results to potentially alter its current scheduling decisions. Moreover, we propose an enhancement technique, called Adaptive Regular Stream Triggering(ART), which significantly enhances stream merging behavior by selectively delaying the initiation of full-length video streams. We analyze the effectiveness of the proposed strategies in terms of their performance effectiveness and impacts on waiting-time predictability through extensive simulation. The results show that significant performance benefits as well as better waiting-time predictability can be attained.