arXiv subjects

P. Brighten Godfrey

Publications and source records attributed to P. Brighten Godfrey.

At least 19 recordsLinked to original sources

arXiv2025

SafeTree: Expressive Tree Policies for Microservices

A microservice-based application is composed of multiple self-contained components called microservices, and controlling inter-service communication is important for enforcing safety properties. Presently, inter-service communication is configured using microservice deployment tools. However, such tools only support a limited class of single-hop policies, which can be overly permissive because they ignore the rich service tree structure of microservice calls. Policies that can express the service tree structure can offer development and security teams more fine-grained control over communication patterns. To this end, we design an expressive policy language to specify service tree structures, and we develop a visibly pushdown automata-based dynamic enforcement mechanism to enforce service tree policies. Our technique is non-invasive: it does not require any changes to service implementations, and does not require access to microservice code. To realize our method, we build a runtime monitor on top of a service mesh, an emerging network infrastructure layer that can control inter-service communication during deployment. In particular, we employ the programmable network traffic filtering capabilities of Istio, a popular service mesh implementation, to implement an online and distributed monitor. Our experiments show that our monitor can enforce rich safety properties while adding minimal latency overhead on the order of milliseconds.

cs.PL↗

arXiv2023

On a Foundation Model for Operating Systems

This paper lays down the research agenda for a domain-specific foundation model for operating systems (OSes). Our case for a foundation model revolves around the observations that several OS components such as CPU, memory, and network subsystems are interrelated and that OS traces offer the ideal dataset for a foundation model to grasp the intricacies of diverse OS components and their behavior in varying environments and workloads. We discuss a wide range of possibilities that then arise, from employing foundation models as policy agents to utilizing them as generators and predictors to assist traditional OS control algorithms. Our hope is that this paper spurs further research into OS foundation models and creating the next generation of operating systems for the evolving computing landscape.

cs.OS↗

arXiv2023

Kivi: Verification for Cluster Management

Modern cloud infrastructure is powered by cluster management systems such as Kubernetes and Docker Swarm. While these systems seek to minimize users' operational burden, the complex, dynamic, and non-deterministic nature of these systems makes them hard to reason about, potentially leading to failures ranging from performance degradation to outages. We present Kivi, the first system for verifying controllers and their configurations in cluster management systems. Kivi focuses on the popular system Kubernetes, and models its controllers and events into processes whereby their interleavings are exhaustively checked via model checking. Central to handling autoscaling and large-scale deployments is our design that seeks to find violations in a smaller and reduced topology. We also develop several model optimizations in Kivi to scale to large clusters. We show that Kivi is effective and accurate in finding issues in realistic and complex scenarios and showcase two new issues in Kubernetes controller source code.

cs.DC↗

arXiv2023

Flock: Accurate network fault localization at scale

Inferring the root cause of failures among thousands of components in a data center network is challenging, especially for "gray" failures that are not reported directly by switches. Faults can be localized through end-to-end measurements, but past localization schemes are either too slow for large-scale networks or sacrifice accuracy. We describe Flock, a network fault localization algorithm and system that achieves both high accuracy and speed at datacenter scale. Flock uses a probabilistic graphical model (PGM) to achieve high accuracy, coupled with new techniques to dramatically accelerate inference in discrete-valued Bayesian PGMs. Large-scale simulations and experiments in a hardware testbed show Flock speeds up inference by >10000x compared to past PGM methods, and improves accuracy over the best previous datacenter fault localization approaches, reducing inference error by 1.19-11x on the same input telemetry, and by 1.2-55x after incorporating passive telemetry. We also prove Flock's inference is optimal in restricted settings

cs.NI↗

arXiv2019

Plankton: Scalable network configuration verification through model checking

Network configuration verification enables operators to ensure that the network will behave as intended, prior to deployment of their configurations. Although techniques ranging from graph algorithms to SMT solvers have been proposed, scalable configuration verification with sufficient protocol support continues to be a challenge. In this paper, we show that by combining equivalence partitioning with explicit-state model checking, network configuration verification can be scaled significantly better than the state of the art, while still supporting a rich set of protocol features. We propose Plankton, which uses symbolic partitioning to manage large header spaces and efficient model checking to exhaustively explore protocol behavior. Thanks to a highly effective suite of optimizations including state hashing, partial order reduction, and policy-based pruning, Plankton successfully verifies policies in industrial-scale networks quickly and compactly, at times reaching a 10000$\times$ speedup compared to the state of the art.

cs.NI↗

arXiv2019

Internet Congestion Control via Deep Reinforcement Learning

We present and investigate a novel and timely application domain for deep reinforcement learning (RL): Internet congestion control. Congestion control is the core networking task of modulating traffic sources' data-transmission rates to efficiently utilize network capacity, and is the subject of extensive attention in light of the advent of Internet services such as live video, virtual reality, Internet-of-Things, and more. We show that casting congestion control as RL enables training deep network policies that capture intricate patterns in data traffic and network conditions, and leverage this to outperform the state-of-the-art. We also highlight significant challenges facing real-world adoption of RL-based congestion control, including fairness, safety, and generalization, which are not trivial to address within conventional RL formalism. To facilitate further research and reproducibility of our results, we present a test suite for RL-guided congestion control based on the OpenAI Gym interface.

cs.NI↗

arXiv2018

Expander Datacenters: From Theory to Practice

Recent work has shown that expander-based data center topologies are robust and can yield superior performance over Clos topologies. However, to achieve these benefits, previous proposals use routing and transport schemes that impede quick industry adoption. In this paper, we examine if expanders can be effective for the technology and environments practical in today's data centers, including the use of traditional protocols, at both small and large scale while complying with common practices such as over-subscription. We study bandwidth, latency and burst tolerance of topologies, highlighting pitfalls of previous topology comparisons. We consider several other metrics of interest: packet loss during failures, queue occupancy and topology degradation. Our experiments show that expanders can realize 3x more throughput than an equivalent fat tree, and 1.5x more throughput than an equivalent leaf-spine topology, for a wide range of scenarios, with only traditional protocols. We observe that expanders achieve lower flow completion times, are more resilient to bursty load conditions like incast and outcast and degrade more gracefully with increasing load. Our results are based on extensive simulations and experiments on a hardware testbed with realistic topologies and real traffic patterns.

cs.NI↗

arXiv2018

cISP: A Speed-of-Light Internet Service Provider

Low latency is a requirement for a variety of interactive network applications. The Internet, however, is not optimized for latency. We thus explore the design of cost-effective wide-area networks that move data over paths very close to great-circle paths, at speeds very close to the speed of light in vacuum. Our cISP design augments the Internet's fiber with free-space wireless connectivity. cISP addresses the fundamental challenge of simultaneously providing low latency and scalable bandwidth, while accounting for numerous practical factors ranging from transmission tower availability to packet queuing. We show that instantiations of cISP across the contiguous United States and Europe would achieve mean latencies within 5% of that achievable using great-circle paths at the speed of light, over medium and long distances. Further, we estimate that the economic value from such networks would substantially exceed their expense.

cs.NI↗

arXiv2016

Measuring and Understanding Throughput of Network Topologies

High throughput is of particular interest in data center and HPC networks. Although myriad network topologies have been proposed, a broad head-to-head comparison across topologies and across traffic patterns is absent, and the right way to compare worst-case throughput performance is a subtle problem. In this paper, we develop a framework to benchmark the throughput of network topologies, using a two-pronged approach. First, we study performance on a variety of synthetic and experimentally-measured traffic matrices (TMs). Second, we show how to measure worst-case throughput by generating a near-worst-case TM for any given topology. We apply the framework to study the performance of these TMs in a wide range of network topologies, revealing insights into the performance of topologies with scaling, robustness of performance across TMs, and the effect of scattered workload placement. Our evaluation code is freely available.

cs.NI↗

arXiv2015

Towards a Speed of Light Internet

In principle, a network can transfer data at nearly the speed of light. Today's Internet, however, is much slower: our measurements show that latencies are typically more than one, and often more than two orders of magnitude larger than the lower bound implied by the speed of light. Closing this gap would not only add value to today's Internet applications, but might also open the door to exciting new applications. Thus, we propose a grand challenge for the networking research community: building a speed-of-light Internet. Towards addressing this goal, we begin by investigating the causes of latency inflation in the Internet across the network stack. Our analysis reveals that while protocol overheads, which have dominated the community's attention, are indeed important, infrastructural inefficiencies are a significant and under-explored problem. Thus, we propose a radical, yet surprisingly low-cost approach to mitigating latency inflation at the lowest layers and building a nearly speed-of-light Internet infrastructure.

cs.NI↗

arXiv2014

A cost-benefit analysis of low latency via added utilization

Several recently proposed techniques achieve latency reduction by trading it off for some amount of additional bandwidth usage. But how would one quantify whether the tradeoff is actually beneficial in a given system? We develop an economic cost vs. benefit analysis for answering this question. We use the analysis to derive a benchmark for wide-area client-server applications, and demonstrate how it can be applied to reason about a particular latency saving technique --- redundant DNS requests.

cs.NI↗

arXiv2014

High Throughput Data Center Topology Design

With high throughput networks acquiring a crucial role in supporting data-intensive applications, a variety of data center network topologies have been proposed to achieve high capacity at low cost. While this literature explores a large number of design points, even in the limited case of a network of identical switches, no proposal has been able to claim any notion of optimality. The case of heterogeneous networks, incorporating multiple line-speeds and port-counts as data centers grow over time, introduces even greater complexity. In this paper, we present the first non-trivial upper-bound on network throughput under uniform traffic patterns for any topology with identical switches. We then show that random graphs achieve throughput surprisingly close to this bound, within a few percent at the scale of a few thousand servers. Apart from demonstrating that homogeneous topology design may be reaching its limits, this result also motivates our use of random graphs as building blocks to explore the design of heterogeneous networks. Given a heterogeneous pool of network switches, through experiments and analysis, we explore how the distribution of servers across switches and the interconnection of switches affect network throughput. We apply these insights to a real-world heterogeneous data center topology, VL2, demonstrating as much as 43% higher throughput with the same equipment.

cs.NI↗

arXiv2013

Shortest Paths in Microseconds

Computing shortest paths is a fundamental primitive for several social network applications including socially-sensitive ranking, location-aware search, social auctions and social network privacy. Since these applications compute paths in response to a user query, the goal is to minimize latency while maintaining feasible memory requirements. We present ASAP, a system that achieves this goal by exploiting the structure of social networks. ASAP preprocesses a given network to compute and store a partial shortest path tree (PSPT) for each node. The PSPTs have the property that for any two nodes, each edge along the shortest path is with high probability contained in the PSPT of at least one of the nodes. We show that the structure of social networks enable the PSPT of each node to be an extremely small fraction of the entire network; hence, PSPTs can be stored efficiently and each shortest path can be computed extremely quickly. For a real network with 5 million nodes and 69 million edges, ASAP computes a shortest path for most node pairs in less than 49 microseconds per pair. ASAP, unlike any previous technique, also computes hundreds of paths (along with corresponding distances) between any node pair in less than 100 microseconds. Finally, ASAP admits efficient implementation on distributed programming frameworks like MapReduce.

cs.DC↗

arXiv2013

Low latency via redundancy

Low latency is critical for interactive networked applications. But while we know how to scale systems to increase capacity, reducing latency --- especially the tail of the latency distribution --- can be much more difficult. In this paper, we argue that the use of redundancy is an effective way to convert extra capacity into reduced latency. By initiating redundant operations across diverse resources and using the first result which completes, redundancy improves a system's latency even under exceptional conditions. We study the tradeoff with added system utilization, characterizing the situations in which replicating all tasks reduces mean latency. We then demonstrate empirically that replicating all operations can result in significant mean and tail latency reduction in real-world systems including DNS queries, database servers, and packet forwarding within networks.

cs.NI↗

arXiv2013

Scalable Routing on Flat Names

We introduce a protocol which routes on flat, location-independent identifiers with guaranteed scalability and low stretch. Our design builds on theoretical advances in the area of compact routing, and is the first to realize these guarantees in a dynamic distributed setting.

cs.NI↗

arXiv2012

Finishing Flows Quickly with Preemptive Scheduling

Today's data centers face extreme challenges in providing low latency. However, fair sharing, a principle commonly adopted in current congestion control protocols, is far from optimal for satisfying latency requirements. We propose Preemptive Distributed Quick (PDQ) flow scheduling, a protocol designed to complete flows quickly and meet flow deadlines. PDQ enables flow preemption to approximate a range of scheduling disciplines. For example, PDQ can emulate a shortest job first algorithm to give priority to the short flows by pausing the contending flows. PDQ borrows ideas from centralized scheduling disciplines and implements them in a fully distributed manner, making it scalable to today's data centers. Further, we develop a multipath version of PDQ to exploit path diversity. Through extensive packet-level and flow-level simulation, we demonstrate that PDQ significantly outperforms TCP, RCP and D3 in data center environments. We further show that PDQ is stable, resilient to packet loss, and preserves nearly all its performance gains even given inaccurate flow information.

cs.NI↗

arXiv2012

Shortest Paths in Less Than a Millisecond

We consider the problem of answering point-to-point shortest path queries on massive social networks. The goal is to answer queries within tens of milliseconds while minimizing the memory requirements. We present a technique that achieves this goal for an extremely large fraction of path queries by exploiting the structure of the social networks. Using evaluations on real-world datasets, we argue that our technique offers a unique trade-off between latency, memory and accuracy. For instance, for the LiveJournal social network (roughly 5 million nodes and 69 million edges), our technique can answer 99.9% of the queries in less than a millisecond. In comparison to storing all pair shortest paths, our technique requires at least 550x less memory; the average query time is roughly 365 microseconds --- 430x faster than the state-of-the-art shortest path algorithm. Furthermore, the relative performance of our technique improves with the size (and density) of the network. For the Orkut social network (3 million nodes and 220 million edges), for instance, our technique is roughly 2588x faster than the state-of-the-art algorithm for computing shortest paths.

cs.SI↗

arXiv2012

Jellyfish: Networking Data Centers Randomly

Industry experience indicates that the ability to incrementally expand data centers is essential. However, existing high-bandwidth network designs have rigid structure that interferes with incremental expansion. We present Jellyfish, a high-capacity network interconnect, which, by adopting a random graph topology, yields itself naturally to incremental expansion. Somewhat surprisingly, Jellyfish is more cost-efficient than a fat-tree: A Jellyfish interconnect built using the same equipment as a fat-tree, supports as many as 25% more servers at full capacity at the scale of a few thousand nodes, and this advantage improves with scale. Jellyfish also allows great flexibility in building networks with different degrees of oversubscription. However, Jellyfish's unstructured design brings new challenges in routing, physical layout, and wiring. We describe and evaluate approaches that resolve these challenges effectively, indicating that Jellyfish could be deployed in today's data centers.

cs.NI↗