Statistical quality of service guarantee for temporal consistency of real-time data objects

In this paper, we study the problem of temporal consistency maintenance where a certain degree of temporal inconsistency is tolerable. We propose a suite of statistical more-less (SML) approaches to tradeoff of quality of service (QoS) of temporal consistency against the number of supported transactions. We begin with a base-line algorithm, SML-BA, which provides the requested QoS of temporal consistency. We then propose SML with optimization (SML-OPT) to further improve the QoS by better utilizing the excessive CPU capacity. Finally, we enhance SML-OPT with a slack reclaiming scheme (SML-SR). The reclaimed slacks are used to process jobs whose required computation time is larger than the guaranteed computation time. Simulation experiments are conducted to compare the performance of these schemes (SML-BA, SML-OPT and SML-SR) together with the deterministic more-less and half-half schemes. Our results show that the SML schemes are effective in trading off the schedulability of transactions and the QoS guaranteed. Moreover, SML-SR performs best and offers a significant QoS improvement over SML-BA and SML-OPT.