Optimal Assignment of Real-Time Systems into Multi-context Dynamically Reconfigurable Processors

In this paper, we focus on the problem of implementing a periodic concurrent system with timing constraints into multi-context dynamically reconfigurable processors (DRP). A concurrent system has multiple tasks that can be executed in parallel. Moreover, some tasks in a specific set of processes might be required to synchronize each other. We propose a method for assigning tasks into a multi-context DRP such that timing constraints of the system are satisfied and the size of the program area required on each context for implementing the given system is minimized. We formulate the problem as an ILP problem and propose a heuristic algorithm for solving the ILP problem efficiently. Experimental results and a case study using ubiquitous sensor devices are given.