Driven by technological advances and new business models, we have seen a renewed interest in LEO satellite constellations. The deployment of large-scale LEO satellite networks is becoming a reality. The network topology changes periodically, as satellites orbit the Earth. This imposes a great challenge to network monitoring. Meanwhile, as a new network monitoring method, In-band Network Telemetry (INT) can provide per-hop granular telemetry metadata, needed to tackle the mobile nature of LEO satellite constellations. Given this, we apply INT to LEO satellite networks for real-time fine-grained monitoring. We propose a path planning solution to identify the paths for network-wide telemetry and the paths for disseminating the telemetry data to the ground facilities. By taking advantage of the predictable satellite trajectories and topology variations, the path planning solution is designed to achieve network-wide coverage and minimize telemetry overhead. We take the LEO48 constellation as an example to visually show the detailed paths of the monitoring scheme. We conduct experiments on different sizes of networks to evaluate the original path planning algorithm and the improved balanced algorithm in this paper, demonstrating the timeliness and balance of the telemetry solution.

Considering the high costs of satellite manufacturing and launch, as well as the complexity of in-orbit debugging, pre-launch emulation on the ground will significantly reduce the development costs of low Earth orbit (LEO) satellite networks. LEO satellite network emulation faces challenges in emulating mega-scale constellations in a lightweight and scalable manner, as well as efficiently handling the frequent link on/off switching for both inter-satellite networks and terrestrial access networks. Existing simulation/emulation tools, such as NS-3, Mininet, QualNet, fall short in addressing these issues effectively. In this work, we propose a lightweight satellite emulation testbed based on Docker containers and the VLAN protocol. In the data plane, our testbed uses Docker containers to emulate satellites/terminals, and uses VETH-pairs and bridges to emulate inter-satellite networks and terrestrial access networks. Furthermore, these virtual network elements can horizontally scale across multiple servers for mega-scale constellation emulation. In the control plane, the real-time constellation topology changes are efficiently emulated through the configuration of VLAN segmentation according to satellite movement patterns. Evaluation shows the testbed's low resource occupancy and high efficiency, with 100 nodes consuming only 1000MB memory and 100 links switching in less than 2s.

In-band Network Telemetry (INT) empowers high-resolution network monitoring by collecting hop-by-hop device-internal states through the data plane without frequently disturbing the control plane. To achieve network-wide monitoring, a high-level orchestration is made to provision multiple monitoring paths to cover the entire network. The path number and path overlapping are kept minimum to maximally reduce the telemetry overhead. However, in production deployment, except for the telemetry overhead, the telemetry timeliness is equally important for fine-grained monitoring, which creates new requirements of balanced monitoring path planning. Given the INT probes from multiple paths are collected to the central controller for analysis, the late arrival of even one probe will delay the analysis process and affect the monitoring timeliness. To address the problem, we propose INT-balance, a novel path planning algorithm for balanced INT path generation. In INT-balance, we first break the original network graph into multiple path segments at the odd vertices. Then, we iteratively splice the two shortest path segments with the joint endpoints to form a longer path segment until the path segment number reaches half the number of the odd vertices. INT-balance generates the minimum number of INT paths with well-balanced path lengths, covering every edge of the network graph without any path overlapping. Evaluation on a network of 100 switches shows that the path length variance of INT-balance is 67% less than that of INT-path, while the algorithm execution time is increased only by 0.012s.

Software-based virtual switches are indispensable in multi-tenant cloud networks. They either work as bridges between virtual machines and the underlying networks, or act as the virtual network function carriers for flexible service chaining and orchestration. Therefore, high-accuracy monitoring of virtual switches is significant for ease of data center network management. The recently proposed In-band Network Telemetry, which relies on the protocol-independent switch architecture (PISA), can achieve the monitoring requirements. However, not all the virtual switches with production quality are P4-based or built under the PISA. In this work, we provide the design and implementation of label-based INT and probe-based INT on top of OVS and VPP, the two mainstream software-based virtual switches with non-PISA architecture. Extensive evaluation shows that our implementation has low performance overhead in terms of forwarding latency, packet loss ratio and CPU consumption. Under 100Mbps traffic pressure, the CPU overhead of INT on OVS and VPP are less than 0.1% and 0.5%, and the switch latency are added by less than$4 \mu\mathrm{s}$and$3\mu\mathrm{s}$, respectively.