Processor/memory Co-Scheduling using periodic resource server for real-time systems under peak temperature constraints

Memory-Aware Scheduling for Fixed Priority Hard Real-Time Computing Systems

“…While there are some works (e.g. [194]) that analytically capture temperature variations based on existing power traces, how to optimize the task allocations in face of real-time and temperature constraints remains a problem. To solve this problem, we adopt the periodic resource model in our approach to manage the CPU and memory resources.…”

“…For overhead, we choose a value of 0.5 ms which is able to account for CPU and memory wake ups, plus cache cold starts. Since a smaller server period offers a smaller SC's peak temperature [194], we choose the value of P test = 10 ms. is reduced because some tasks may have average power values that prevent them from being allocated alone to any SC. Also, it can be seen in Fig.…”