Worst-case execution time analysis for many-core architectures with NoC

The optimal deployment of data streaming applications onto multi-/many-core platform providing real-time guarantees requires to solve the application partitioning and placement, buffer allocation and task mapping and scheduling optimisation problem using the tasks Worst-Case Execution Time (WCET). In turn, tasks WCET varies due to interferences that tasks experience when accessing shared resources that depend on the solutions of the optimisation problem. In this paper we propose a detailed interference-based method that first over-approximates the WCET and then tightens it by pruning out the interferences from tasks not overlapping in time and memory. We prove that derived bounds are safe. We have found that interferences on average amount to 10% of WCET, and were able to improve latency guarantee up to 34%.