An Efficient Branch-and-Bound Routing Optimization Method for Optical NoCs

Silicon photonics-based optical networks-on-chip (ONoCs) are emerging as a power-efficient on-chip communication architecture for the next generation of chip multiprocessors. However, the thermal sensitivity of photonic devices presents power consumption challenges. Existing routing schemes optimized for optical power loss tend to avoid passing through high-temperature nodes, which in turn leads to contention at low-temperature nodes. It remains a crucial challenge to develop an adaptive routing algorithm that strikes a balance between the power consumption optimization and performance optimization. In this work, we first propose an efficient branch-and-bound routing (BBR) optimization method for ONoCs. To the best of our knowledge, it is the first time that the branch-and-bound (BB) method is adopted to solve the routing optimization problem in ONoCs. We further developed three variants of the BBR optimization method: 1) BBTR; 2) BBCR; and 3) 3BOR. Among them, 3BOR employs a bi-objective bounding function to optimize both optical power loss and network performance, while enhancing algorithmic efficiency. Experimental results demonstrate that, compared to the state-of-the-art heuristic contention-aware thermal-reliable routing algorithm, 3BOR reduces the thermal-induced optical power loss by 15.6% while reducing the algorithm running time by 82.2%.