Tractable schedulability analysis and resource allocation for real-time multimodal systems

Abstract

Real-time multimedia subsystems often require support for switching between different resource and application execution modes. To ensure that timing constraints are not violated during or after a subsystem mode change, real-time schedulability analysis is required. However, existing time-efficient multimode schedulability analysis techniques for application-only mode changes are not appropriate for subsystems that require changes in the resource execution behavior (e.g., processors with dynamic power modes). Furthermore, all existing multimode schedulability analysis that handles both resource and application mode changes is highly exponential and not scalable for subsystems with a moderate or large number of modes. As a result, the notion of resource optimality is still unaddressed for real-time multimodal systems. In this report, we first address the lack of tractable schedulability analysis for such subsystems by proposing a model for characterizing multiple resource and application modes and by deriving a sufficient schedulability test that has pseudo-polynomial time complexity. Finally, we propose an algorithm which leverages this pseudo-polynomial schedulability analysis to optimize the resource usages (e.g., to minimize peak-power load) of a multimodal real-time system. Simulation results show that our proposed algorithms for schedulability analysis and resource allocation, when compared with previously-proposed approaches, require significantly less time and are just as precise.