A stochastic approach for fine grain QoS control

We present a method for fine grain QoS control of multimedia applications. This method takes as input an application software composed of actions parameterized by quality levels. Our method allows the construction of a Quality Manager which computes adequate action quality levels, so as to meet QoS requirements (action deadlines are met and quality levels are maximal) for a given platform. We have developed a stochastic approach based on probability distribution functions for the execution time of the actions. Our method is parameterized according to the importance attributed to deadline misses. This means that the user is given the possibility to express how hard the realtime constraints are. Besides, given a value of the parameter, we can compute the expected deadline miss ratio for the controlled application. We present experimental results including the implementation of the method and benchmarks for an MPEG4 video encoder.

• Type: Conference
• Year: 2008
• Where: ESTIMEDIA
• Authors: Jacques Combaz, Loïc Strus
• Keywords: Action Quality Levels, ESTIMEDIA 2008, Fine Grain Qos, Multimedia, Quality Levels