Markov Decision Processes: Discrete Stochastic Dynamic Programming
Markov Decision Processes: Discrete Stochastic Dynamic Programming
QoS Control Strategies for High-Quality Video Processing
Real-Time Systems
Model-based decision framework for autonomous application migration
ASMTA'11 Proceedings of the 18th international conference on Analytical and stochastic modeling techniques and applications
A Robust Mechanism for Adaptive Scheduling of Multimedia Applications
ACM Transactions on Embedded Computing Systems (TECS)
Streaming video over HTTP with consistent quality
Proceedings of the 5th ACM Multimedia Systems Conference
| Hi-index | 0.00 |
Many media processing applications create a load that varies significantly over time. Hence, if such an application is assigned a lower processing-time budget than needed in its worst-case load situation, deadline misses are likely to occur. This problem can be dealt with by designing media processing applications in a scalable fashion. A scalable media processing application can run in multiple qualities, leading to correspondingly different resource demands. The problem we consider is to find an accompanying quality control strategy, which minimizes both the number of deadline misses and the number of quality changes, while maximizing the quality of processing. We present an initial approach to the above problem by modeling it as a Markov decision process (MDP). Our model is based on measuring relative progress at milestones. Solving the MDP results in a quality control strategy that can be applied during runtime with only little overhead. We evaluate our approach by means of a practical example, which concerns a scalable MPEG-2 decoder.