Modern AI interconnects face hard trade-offs: copper links are efficient but have a short reach, optical transceivers are long-reach but more costly and energy-hungry. Similar trade-offs occur in the switching domain: packet switching is flexible but power-hungry, while circuit switching is efficient but coarse-grained. Recently, slow-and-wide transceivers have closed the gap between copper and traditional optics by exposing many low-rate optical channels, but limit them to a single logical point-to-point link.

Prior work on LEO satellite networks focuses on regular constellations and grid-like topologies. Using existing Starlink satellite data, we uncover and systematically characterize the constellation's irregularities at multiple granularities, including shell distribution, orbital spacing, intra-orbit satellite placement, and hardware heterogeneity. To expose the actual impact of these irregularities, we try and superimpose grid-like topologies onto them, and study the impact of doing so. Our simulations show that forcing regular topologies on irregular constellations significantly affects routing and network performance, revealing a fundamental mismatch between idealized models and real deployments. Motivated by these findings, we outline future directions, including irregularity-aware topologies, systems and perspectives.

The size and communication patterns of modern ML training workloads place significant strain on datacenter fabrics. When bandwidth demand exceeds capacity, flows experience slowdowns and iteration time grows larger. Fine-grained load-balancing such as packet spraying cannot fully resolve this issue, yet it can shift the bottleneck from individual links to groups of links partitioning the network. In this work, we present Ampel: a system to schedule ML training flows at the network cuts. It leverages packet-spraying's ability to spread traffic evenly across all available paths to simplify the view of the topology into one only containing potential bottlenecks, and then bridges this new simplified model with past work on coflow scheduling. Our simulated experiments show that Ampel can reduce average training iteration time by up to 18% compared to state-of-the-art ML schedulers.

Private 5G networks will soon be ubiquitous across the future-generation smart wireless access infrastructures hosting a wide range of performance-critical applications. A high-performing User Plane Function (UPF) in the data plane is critical to achieving such stringent performance goals, as it governs fast packet processing and supports several key control-plane operations. Based on a private 5G prototype imple-mentation and analysis, it is imperative to perform dynamic resource management and orchestration at the UPF. This paper leverages Mobile Edge Cloud-Intelligent Agent (MEC-IA), a logically centralized entity that proactively distributes resources at UPF for various service types, significantly reducing the tail latency experienced by the user requests while maximizing resource utilization. Extending the MEC-IA functionality to MEC layers further incurs data plane latency reduction. Based on our extensive simulations, under skewed uRLLC traffic arrival, the MEC-IA assisted bestfit UPF-MEC scheme reduces the worst-case latency of UE requests by up to 77.8% w.r.t. baseline. Additionally, the system can increase uRLLC connectivity gain by 2.40× while obtaining 40% CapEx savings.

All-optical circuit-switching (OCS) technology is the key to design energy-efficient and high-performance datacenter network (DCN) architectures for the future. However, existing round-robin based OCS cores perform poorly under realistic workloads having high traffic skewness and high volume of inter-rack traffic. To address this issue, we propose a novel DCN architecture OSSV: a combination of OCS-based core (between ToR switches) and OCS-based reconfigurable edge (between servers and ToR switches). On one hand, the OCS core is traffic agnostic and realizes reconfigurably non-blocking ToR-level connectivity. On the other hand, OCS-based edge reconfigures itself to reshape the incoming traffic in order to jointly minimize traffic skewness and inter-rack traffic volume. Our novel optimization framework can obtain the right balance between these intertwined objectives. Our extensive simulations and testbed evaluation show that OSSV can achieve high performance under diverse DCN traffic while consuming low power and incurring low cost.

All-optical circuit-switched (OCS) core is the holy grail for the future generation datacenter architectures. However, such proposals consist of a common operational abstraction termed as round-robin circuit scheduling, which heavily suffers from a) high traffic skewness, and b) high volume of inter-rack traffic. To address this issue, we propose a novel architecture: round-robin OCS-core equipped with OCS-based reconfigurable edge for joint Skewness and Inter-rack traffic Volume (SV) minimization. Our architecture significantly improves the performance of all-optical cores, making it very close to a non-blocking network.

An optical circuit-switched network core has the potential to overcome the inherent challenges of a conventional electrical packet-switched core of today’s compute clusters. As optical circuit switches (OCS) directly handle the photon beams without any optical-electrical-optical (O/E/O) conversion and packet processing, OCS-based network cores have the following desirable properties: a) agnostic to data-rate, b) negligible/zero power consumption, c) no need of transceivers, d) negligible forwarding latency, and e) no need for frequent upgrade. Unfortunately, OCS can only provide point-to-point (unicast) circuits. They do not have built-in support for one-to-many (multicast) communication, yet multicast is fundamental to a plethora of data-intensive applications running on compute clusters nowadays. In this paper, we propose Shufflecast, a novel optical network architecture for next-generation compute clusters that can support high-performance multicast satisfying all the properties of an OCS-based network core. Shufflecast leverages small fanout, inexpensive, passive optical splitters to connect the Top-of-rack (ToR) switch ports, ensuring data-rate agnostic, low-power, physical-layer multicast. We thoroughly analyze Shufflecast’s highly scalable data plane, light-weight control plane, and graceful failure handling. Further, we implement a complete prototype of Shufflecast in our testbed and extensively evaluate the network. Shufflecast is more power-efficient than the state-of-the-art multicast mechanisms. Also, Shufflecast is more cost-efficient than a conventional packet-switched network. By adding Shufflecast alongside an OCS-based unicast network, an all-optical network core with the aforementioned desirable properties supporting both unicast and multicast can be realized.

Emerging distributed applications, such as microservices, machine learning, big data analysis, consist of both compute and network tasks. DAG-based abstraction primarily targets compute tasks and has no explicit network-level scheduling. In contrast, Coflow abstraction collectively schedules network flows among compute tasks but lacks the end-to-end view of the application DAG. Because of the dependencies and interactions between these two types of tasks, it is sub-optimal to only consider one of them. We argue that co-scheduling of both compute and network tasks can help applications towards the globally optimal end-to-end performance. However, none of the existing abstractions can provide fine-grained information for co-scheduling. We propose MXDAG, an abstraction to treat both compute and network tasks explicitly. It can capture the dependencies and interactions of both compute and network tasks leading to improved application performance.