The implementation of 5G and the future deployment of 6G necessitate the utilization of optical networks that possess substantial capacity and exhibit minimal latency. The dynamic arrival and departure of connection requests in optical networks result in particular central links experiencing more traffic and congestion than non-central links. The occurrence of congested links leads to service blocking despite the availability of resources within the network, restricting the efficient utilization of network resources. The available algorithms in the literature that aim to balance load among network links offer a trade-off between blocking performance and algorithmic complexity, thus increasing service provisioning time. This work proposes a dynamic routing-based congestion-aware routing, modulation, core, and spectrum assignment (RMCSA) algorithm for space division multiplexing elastic optical networks (SDM-EONs). The algorithm finds alternative candidate paths based on real-time link occupancy metrics to minimize blocking due to link congestion under dynamic traffic scenarios. As a result, the algorithm reduces the formation of congestion hotspots in the network owing to link-betweenness centrality. We have performed extensive simulations using two realistic network topologies to compare the performance of the proposed algorithm with relevant RMCSA algorithms available in the literature. The simulation results verify the superior performance of our proposed algorithm compared to the benchmark Yen's K-shortest paths and K-Disjoint shortest paths RMCSA algorithms in connection blocking ratio and spectrum utilization efficiency. To expedite the route-finding process, we present a novel caching strategy that allows the proposed algorithm to demonstrate a much-reduced service delay time compared to the recently developed adaptive link weight-based load-balancing RMCSA algorithm.

翻译：5G的实施以及未来6G的部署，要求光网络具备大容量和低时延特性。光网络中连接请求的动态到达与离开，导致某些中心链路比非中心链路承载更多流量并出现拥塞。拥塞链路的出现会导致网络内部虽有可用资源，但业务仍被阻塞，从而限制了网络资源的高效利用。现有文献中旨在平衡网络链路间负载的算法，在阻塞性能与算法复杂度之间存在权衡，这增加了业务供给时间。本文针对空分复用弹性光网络（SDM-EON），提出一种基于动态路由的、拥塞感知的路由、调制、纤芯与频谱分配（RMCSA）算法。该算法基于实时链路占用度量寻找替代候选路径，以最小化动态流量场景下因链路拥塞导致的阻塞。因此，该算法减少了因链路介数中心性而在网络中形成的拥塞热点。我们使用两种实际网络拓扑进行了大量仿真，将所提算法与文献中相关的RMCSA算法进行性能比较。仿真结果验证了，在连接阻塞率和频谱利用效率方面，我们提出的算法优于基准的Yen's K最短路径和K不相交最短路径RMCSA算法。为了加速路由发现过程，我们提出了一种新颖的缓存策略，使得所提算法与近期发展的基于自适应链路权重的负载均衡RMCSA算法相比，能够展现出显著降低的服务时延。