Allocation of Multirate Systems on Multiprocessors with Memory Hierarchy Modeling and Optimization

The paper presents a task allocation scheme for system-level synthesis of multirate real-time tasks on multiprocessors with memory hierarchies. The allocation algorithm is the first to take into account the effect of memory hierarchies at task level and optimizes for it. The algorithm is based on a task-level model of hierarchical memories first proposed in our previous work [1]. Caches are essential for modern RISC embedded cores to obtain sustained high performance. However, caches have received limited use in priority-driven preemptive real-time systems due to the unpredictability of caches -- average-case improvements are of no use in systems with hard deadlines. Our task-level model of performance in the presence of memory hierarchies provides an efficient means to bound the guaranteed memory performance of tasks running in a multi-rate, multitasking environment. Our system synthesis algorithm uses cache partitioning and reservation techniques to guarantee cache hits for some tas..