Cloud and AI/ML workloads are posing unprecedented new requirements on the wide-area network: it must combine strict availability, massive growth, and feature agility. It became increasingly clear that traditional WAN designs were ill-equipped to adapt to these requirements.

Private WANs are increasingly important to the operation of enterprises, telecoms, and cloud providers. For example, B4, Google's private software-defined WAN, is larger and growing faster than our connectivity to the public Internet. In this paper, we present the five-year evolution of B4. We describe the techniques we employed to incrementally move from offering best-effort content-copy services to carrier-grade availability, while concurrently scaling B4 to accommodate 100x more traffic. Our key challenge is balancing the tension introduced by hierarchy required for scalability, the partitioning required for availability, and the capacity asymmetry inherent to the construction and operation of any large-scale network. We discuss our approach to managing this tension: i) we design a custom hierarchical network topology for both horizontal and vertical software scaling, ii) we manage inherent capacity asymmetry in hierarchical topologies using a novel traffic engineering algorithm without packet encapsulation, and iii) we re-architect switch forwarding rules via two-stage matching/hashing to deal with asymmetric network failures at scale.

As live streaming networks grow in scale and complexity, they are becoming increasingly difficult to evaluate. Existing evaluation methods including lab/testbed testing, simulation, and theoretical modeling, lack either scale or realism. The industrial practice of gradually-rolling-out in a testing channel is lacking in controllability and protection when experimental algorithms fail, due to its passive approach. In this paper, we design a novel system called ShadowStream that introduces evaluation as a built-in capability in production Internet live streaming networks. ShadowStream introduces a simple, novel, transparent embedding of experimental live streaming algorithms to achieve safe evaluations of the algorithms during large-scale, real production live streaming, despite the possibility of large performance failures of the tested algorithms. ShadowStream also introduces transparent, scalable, distributed experiment orchestration to resolve the mismatch between desired viewer behaviors and actual production viewer behaviors, achieving experimental scenario controllability. We implement ShadowStream based on a major Internet live streaming network, build additional evaluation tools such as deterministic replay, and demonstrate the benefits of ShadowStream through extensive evaluations.

Physical layer techniques have come a long way and can achieve very close to Shannon capacity for point-to-pint links. It is apparent that, to further improve network capacity significantly, we have to resort to concurrent transmissions. Many concurrent transmission techniques (e.g., zero forcing, interference alignment and distributed MIMO) are proposed in which multiple senders jointly encode signals to multiple receivers so that interference is aligned and each receiver is able to decode its desired information. In this paper, we investigate the constraints and challenges of using interference alignment. Our main contribution is conducting the first systematic investigation on the key issue of identifying opportunities for interference alignment. We identify diverse, novel scenarios for using interference alignment. We show that identifying opportunities for interference alignment in the general case is computational challenging. However, we also present a promising, distributed algorithm for identifying a wide range of opportunities for interference alignment using a unifying framework based on the degree of freedom. Our second contribution is evaluating key practical implementation issues.

Network resiliency is crucial to IP network operations. Existing techniques to recover from one or a series of failures do not offer performance predictability and may cause serious congestion. In this paper, we propose Resilient Routing Reconfiguration (R3), a novel routing protection scheme that is (i) provably congestion-free under a large number of failure scenarios; (ii) efficient by having low router processing overhead and memory requirements; (iii) flexible in accommodating different performance requirements (e.g., handling realistic failure scenarios, prioritized traffic, and the trade-off between performance and resilience); and (iv) robust to both topology failures and traffic variations. We implement R3 on Linux using a simple extension of MPLS, called MPLS-ff. We then conduct extensive Emulab experiments and simulations using realistic network topologies and traffic demands. Our results show that R3 achieves near-optimal performance and is at least 50% better than the existing schemes under a wide range of failure scenarios.

We present muNet, a wireless mesh network design and implementation to harness the multiuser capacity of wireless channels. Traditionally, media access control is designed to schedule one transmission between one sender and one receiver without interference at any given time. However, this design is suboptimal in terms of achieving the multiuser capacity of multi-access wireless channels. In muNet, we implement effective physical layer techniques called superposition coding and successive interference cancellation to enable simultaneous unicast transmissions from a single transmitter to multiple receivers as well as from multiple transmitters to a single receiver. We design the first practical MAC protocol that leverages such a physical layer and exposes the multiuser capacity to upper layers. We also present a simple, effective routing protocol that increases simultaneous transmission opportunities for the MAC layer. A proof-of-concept muNet is implemented on the GNU radio platform. Measurements on the implementation shows that the throughput gains of muNet are significant (up to 93%).

A major barrier for the adoption of wireless mesh networks is severe limits on throughput. Many in-network packet mixing techniques at the network layer [1], [2], [3] as well as the physical layer [4], [5], [6] have been shown to substantially improve throughput. However, the optimal mixing algorithm that maximizes throughput is still unknown. We propose iPack, an algorithm for in-network generation of composite packets that integrates coding at two different layers of the protocol stack: XOR-based network coding and physical layer superposition coding. Using extensive simulations, we find that the throughput gain of the joint coding iPack algorithm is 30% more than the better performer of network coding and superposition coding in a wide range of scenarios, and automatically takes advantage of the best available coding opportunities. In a typical wireless mesh network when more traffic is between the clients and access points, the average throughput improvement of iPack, our joint optimization scheduler, can be 324%, while there can be little gain (less than 10%) if network coding alone is used. We also validate our results by implementing iPack on a small-scale testbed based on GNU Radio.

Configurations for today's IP networks are becoming increasingly complex. As a result, configuration management is becoming a major cost factor for network providers and configuration errors are becoming a major cause of network disruptions. In this paper, we present and evaluate the novel idea of shadow configurations. Shadow configurations allow configuration evaluation before deployment and thus can reduce potential network disruptions. We demonstrate using real implementation that shadow configurations can be implemented with low overhead.

A major barrier for the adoption of wireless mesh networks is severe limits on throughput. In this paper, we apply superposition coding to substantially improve network capacity of large, dense wireless mesh networks. Superposition coding is a physical layer technique that allows a transmitter to simultaneously send independent packets to multiple receivers. While superposition coding has been studied extensively by the physical layer community, we present the first design of practical and effective MAC protocols to take advantage of superposition coding in wireless mesh networks. Extensive evaluations show that superposition coding can be a practical method to increase the throughput of large, dense wireless mesh networks. Specifically, in a mesh network with 2 to 64 active receivers and one gateway, we show that our system can increase throughput up to 154%, with average gain ranging from 10% to 21%. When there are multiple gateways forming a mesh network, our system gains up to 98%, with average gain ranging from 24% to 46%. These results clearly demonstrate the potential benefits of our system. We also present results from an implementation of superposition coding using GNU Radio.