An energy-aware mapping algorithm for mesh-based network-on-chip architectures

摘要

Network-on-chip (NoC) mapping algorithms significantly affect NoC system performance in terms of communication cost and energy consumption. For a specific application represented by a task graph, this paper proposes an energy-aware mapping algorithm that searches for the mapping solution with best communication locality and therefore lowest energy consumption. During the search procedure, we employ a simulation-free, communication locality-based energy model to evaluate the quality of each candidate mapping. By iteratively updating the best candidate mapping using a greedy search heuristic, the search procedure converges to an mapping decision with optimal energy efficiency in the search space. Compared with the round-robin mapping strategy, the proposed method is capable of exploring energy-efficient mapping decision for various applications as well as network sizes.