Hardware transports such as RoCE deliver high performance with minimal host CPU, but are best suited to special-purpose deployments that limit their use, e.g., backend networks or Ethernet with Priority Flow Control (PFC). We introduce Falcon, the first hardware transport that supports multiple Upper Layer Protocols (ULPs) and heterogeneous application workloads in general-purpose Ethernet datacenter environments (with losses and without special switch support). Key design elements include: delay-based congestion control with multipath load balancing; a layered design with a simple request-response transaction interface for multi-ULP support; hardware-based retransmissions and error-handling for scalability; and a programmable engine for flexibility. The first Falcon hardware implementation delivers a peak performance of 200 Gbps, 120 Mops/sec, with near-optimal operation completion times that are up to 8× lower than CX-7 RoCE under network congestion, and up to 65% higher goodput under lossy conditions.

Dataplane languages like P4 enable flexible and efficient packet-processing using domain-specific primitives such as programmable parsers and match-action tables. Unfortunately, P4 programs tend to be monolithic and tightly coupled to the hardware architecture, which makes it hard to write programs in a portable and modular way---e.g., by composing reusable libraries of standard protocols.

This paper presents McNetKAT, a scalable tool for verifying probabilistic network programs. McNetKAT is based on a new semantics for the guarded and history-free fragment of Probabilistic NetKAT in terms of finite-state, absorbing Markov chains. This view allows the semantics of all programs to be computed exactly, enabling construction of an automatic verification tool. Domain-specific optimizations and a parallelizing backend enable McNetKAT to analyze networks with thousands of nodes, automatically reasoning about general properties such as probabilistic program equivalence and refinement, as well as networking properties such as resilience to failures. We evaluate McNetKAT's scalability using real-world topologies, compare its performance against state-of-the-art tools, and develop an extended case study on a recently proposed data center network design.

Network virtualization stacks are the linchpins of public clouds. A key goal is to provide performance isolation so that workloads on one Virtual Machine (VM) do not adversely impact the network experience of another VM. Using data from a major public cloud provider, we systematically characterize how performance isolation can break in current virtualization stacks and find a fundamental tradeoff between isolation and resource multiplexing for efficiency. In order to provide predictable performance, we propose a new system called PicNIC that shares resources efficiently in the common case while rapidly reacting to ensure isolation. PicNIC builds on three constructs to quickly detect isolation breakdown and to enforce it when necessary: CPU-fair weighted fair queues at receivers, receiver-driven congestion control for backpressure, and sender-side admission control with shaping. Based on an extensive evaluation, we show that this combination ensures isolation for VMs at sub-millisecond timescales with negligible overhead.

ProbNetKAT is a probabilistic extension of NetKAT with a denotational semantics based on Markov kernels. The language is expressive enough to generate continuous distributions, which raises the question of how to compute effectively in the language. This paper gives an new characterization of ProbNetKAT’s semantics using domain theory, which provides the foundation needed to build a practical implementation. We show how to use the semantics to approximate the behavior of arbitrary ProbNetKAT programs using distributions with finite support. We develop a prototype implementation and show how to use it to solve a variety of problems including characterizing the expected congestion induced by different routing schemes and reasoning probabilistically about reachability in a network.

As businesses move their critical IT operations to multi-tenant cloud data centers, it is becoming increasingly important to provide network performance guarantees to individual tenants. Due to the impact of network congestion on the performance of many common cloud applications, recent work has focused on enabling network reservation for individual tenants. Current network reservation methods, however, do not gracefully degrade in the presence of network over subscriptions that may frequently occur in a cloud environment. In this context, for a shared data center network, we introduce Network Satisfaction Ratio (NSR) as a measure of the satisfaction derived by a tenant from a given network reservation. NSR is defined as the ratio of the actual reserved bandwidth to the desired bandwidth of the tenant. Based on NSR, we present a novel network reservation mechanism that can admit time-varying tenant requests and can fairly distribute any degradation in the NSR among the tenants in presence of network over subscription. We evaluate the proposed method using both synthetic network traffic trace and representative data center traffic trace generated by running a reduced data center job trace in a small test bed. The evaluation shows that our method adapts to changes in network reservations, and it provides significant and fair improvement in NSR when the data center network is oversubscribed.