Computational Complexity of Allow Rule Ordering and its Greedy Algorithm

Packet classification is used to determine the behavior of incoming packets in network devices according defined rules. As it achieved using a linear search on rule list, large number rules will lead longer communication latency. To solve this, problem finding order minimizing latency has been studied. Misherghi et al. and Harada have proposed that relaxes policy-based constraints. In this paper, we show Relaxed Optimal Rule Ordering (RORO) for allowlist NP-hard, by reducing from RORO general list NP-hard. We also propose heuristic algorithm based greedy method an allowlist. Furthermore, demonstrate effectiveness our ClassBench, which benchmark packet algorithms.