Recently, large-scale and mega low earth orbit (LEO) constellations have become the primary development trends in satellite networks. Considering the complex topology and limited processing capabilities of LEO satellites, traditional distributed routing algorithms with significant flooding overhead are not suitable for the era of mega constellations. In this context, centralized routing technology based on Software Defined Network (SDN) architecture has been widely applied. However, centralized routing technology exhibits a slower response to inter-satellite link (ISL) changes due to longer informing path, which induce that longer time is required to restore traffic in the event of ISL terminal or satellite /hardware faults. To this end, this paper proposes a lightweight Geographic-based Auxiliary Routing Scheme (GARS), adapted to Walker Delta constellation, as a traffic fast restoration strategy for SDN-based centralized routing scheme. This scheme has an ultralow computation complexity ofO(1), enabling the rapid calculation and establishment of temporary paths to restore service during the SDN controller in ground station rerouting. In addition, an optimized geometric routing algorithm with anti-loop mechanism is proposed to enhance the stability and reliability of geometric routing algorithm. Simulation results show that (1) compared with traditional centralized routing system, the routing scheme proposed in this paper shortens interruption time of traffic by at least 65.5% and reduces mean packet loss ratio by at least 67.2% without increasing link load. (2) Compared to other traditional geometric routing algorithms without advanced anti-loop mechanism, GARS with the optimized anti-loop mechanism also effectively ensures the continuity of service, reducing packet loss rate and interruption time by at most 96%.

Optical inter-satellite link (ISL) with tens/hundreds Gbps is applied into LEO constellation to overcome the insufficient bandwidth of microwave ISL. Moreover, compared with GEO/MEO constellation, LEO constellation suffers from frequent topology switching because of the lowest orbital-altitude. Topology switching between adjacent topology snapshots is generally done through ISL handover (remove or establish ISL). However, service interruption caused by ISL handover within topology switching is a vital issue to address. After ISL handover occurs, isolated topology and slow route convergence (derived from long ISL delay and forwarding the updated message among hundreds/thousands of satellites) may cause severe service interruption. But considering the fact that time-varying topology is predictable and all the ISL handover could be pre-scheduled for topology management purpose, service interruption problem is investigated from the aspect of ISL handover and routing strategy in this paper. Owing to long establishing time of optical ISL, grouped optical ISL handover scheme is raised to decrease route convergence time during topology switching. Furthermore, predictive-update routing scheme based on Open Shortest Path First (OSPF-PUR) is proposed to minimize route convergence time for each ISL handover. For predictable ISL handover, in OSPF-PUR scheme, forwarding table is refreshed according to locally stored ISL handover information but without flooding. Simulation results illustrate that grouped optical ISL handover scheme reduces route convergence time within topology switching while topology connectivity is guaranteed. The route convergence time of ISL handover is effectively decreased by OSPF-PUR scheme. Besides, for OSPF-PUR scheme, packet loss rate remains unchanged while topology switching frequency increases.

Switching the traffic in the optical domain has been considerably investigated as a future-proof solution to overcome the intrinsic bandwidth bottleneck of electrical switches in data center networks (DCNs). However, due to the lack of fast and scalable optical switch control mechanism, the lack of optical buffers for contention resolution, and the complicated implementation of fast clock and data recovery (CDR), the practical deployment of fast optical switches in data centers (DCs) remains a big challenge. In this work, we develop and experimentally demonstrate for the first time a flow-controlled and clock-distributed optical switch and control system, implementing 43.4 ns optical switch configuration time, less than 3.0E-10 packet loss rate resulting from the packet contention, and 3.1 ns fast CDR time. Experimental results confirm that zero buffer overflow caused packet loss and lower than$3~ \boldsymbol {\mu }\text{s}$server-to-server latency are achieved for network deploying a smaller electrical buffer of 8192 bytes at a traffic load of 0.5. Real servers running the Transmission Control Protocol (TCP) traffic generating and monitoring tools are exploited in this switch and control system as well, validating its capability of running practical DCNs services and applications with full TCP bandwidth.