To accelerate computational science on campus and beyond we investigate the Ray software framework across a range of different computing environments including 1) desktop systems; 2) campus compute clusters; 3) NSF-supported midscale computing and networking research infrastructures; and 4) hybrid campus and commercial compute clouds such as the Google Cloud Platform (GCP). Ray offers a single, unifying, open-source, distributed computing framework that promises to allow users to code once - using the Python language in Jupyter notebooks familiar to many researchers across disciplines - and easily deploy applications spanning diverse computing platforms. In this paper we specifically focus on the benefits of training researchers and students on using Ray on the FABRIC and CloudLab mid-scale research infrastructures, highlighting the ease of use of multiprocessing and hardware accelerators (e.g., GPUs) across heterogeneous hardware systems.

Deploying small cells with wireless backhaul promises to improve aggregate NextG cel-lular system capacity and coverage, and reduce the need for disruptive and costly new fiber installations. However, wireless backhaul architectures introduce design challenges and open research questions at multiple protocol layers, ranging from siting new cells to multi-hop topology management. To address siting challenges we introduce a new, open-source software-based Small Cell Topology Design Tool to help guide Mobile Network Operators (MNOs) and other stakeholders with planning the feasible place-ment of cells.

We describe a prototype of Science Traffic as a Service (STAAS), a decentralized, cooperative system to collect, filter and distribute a diverse collection of real and synthetic data traffic to the global experimental research testbed user community. Available on-demand to networking experimenters through a web dashboard, the tool promises to elevate traffic selection and distribution to a first class experimental instrumentation resource. We believe the alternatives to providing this service on large-scale federated testbeds are increasingly unworkable for experimenters. As backbone networks increasingly deploy 100–1000 Gbps communications links we are moving beyond the point where experimenters can reasonably be asked to independently, safely, and efficiently create test traffic that provides the realism that their investigations will demand. We seek to deploy our prototype at campuses and testbeds attached to the emerging FABRIC mid-scale networking research infrastructure. We describe prototype design, operation and implementation, and how it is integrated with existing campus networking infrastructure. We explain how remote experimenters will request and acquire network traffic to study. We detail our process for forwarding campus traffic onto the experimental testbed, while striving to preserve both the timing integrity of the flows and the data privacy of their payloads.†

Many promising networking research ideas in programmable networks never see the light of day. Yet, deploying research prototypes in production networks can help validate research ideas, improve them with faster feedback, uncover new research questions, and also ease the subsequent transition to practice. In this paper, we show how researchers can run and validate their research ideas in their own backyards---on their production campus networks---and we have seen that such a demonstrator can expedite the deployment of a research idea in practice to solve real network operation problems. We present P4Campus , a proof-of-concept that encompasses tools, an infrastructure design, strategies, and best practices---both technical and non-technical---that can help researchers run experiments against their programmable network idea in their own network. We use network tapping devices, packet brokers, and commodity programmable switches to enable running experiments to evaluate research ideas on a production campus network. We present several compelling data-plane applications as use cases that run on our campus and solve production network problems. By sharing our experiences and open-sourcing our P4 apps [28], we hope to encourage similar efforts on other campuses.

The envisioned dense deployment of millimeter wave small cells threatens to require expensive and potentially disruptive installation of fiber backhaul from each cell location to a Mobile Network Operator’s (MNO) nearest point-of-presence. Integrated Access and Backhaul (IAB) uses part of the wireless spectrum in lieu of fiber for the backhaul connection, promising to reduce the cost of fiber deployments. But IAB architectures also introduce design challenges and open research questions at multiple protocol layers, ranging from wireless self-interference to multi-hop topology management. To explore these issues we construct a replicable research testbed to investigate systems level architectural and performance issues in multi-hop IAB settings. We show how the Open Networking Foundation’s (ONF) Aether™ platform can be used to emulate alternative wireless backhaul architectures across a range of topologies and radio technologies. Using commercially available Citizens Broadband Radio Service (CBRS) radios and small cells, we evaluate the performance of a UE communicating over a multi-hop topology.1

Many promising networking research ideas never see the light of day. Yet, deploying research prototypes in production networks can improve networking practice, while providing the feedback necessary to make good ideas better quickly. We argue that academic networking researchers can and should demonstrate their research ideas in their own backyards—on their campus networks. We have found that using commodity programmable switches to create, deploy, and run experimental data-plane applications speeds the innovation pipeline between the conception of a novel idea and its deployment. We present our Camp4 infrastructure and several compelling applications running on our campus that solve production network problems. We also describe policies, strategies, and tactics for overcoming obstacles to working with production traffic. By sharing our experiences and open-sourcing our P4 apps, we hope to encourage similar efforts on other campuses.

To continue innovating in an age of at-scale computer systems research, the academic computing and networking systems research community must explore new approaches to addressing growing researcher demands to support larger size experiments. CloudJoin explores a transformational approach to scaling out successful Computing Research Infrastructures (CRI) into larger testbeds by creating hybrid cloud computing systems. We describe how to create a seamless, scalable, single experiment testbed that spans CloudLab and the Google Cloud Platform (GCP), while requiring no infrastructure changes. In addition to added elastic computing capacity, CloudJoin experiments benefit from easy access specialized hardware and cloud services and APIs to leverage world class data analytics and experimental infrastructure monitoring. In this work-in-progress, we show how to integrate the infrastructures by creating a Virtual Private Network between a CloudLab experiment and a GCP Virtual Private Cloud (VPC). To simplify understanding of large-scale experiment behavior, problem diagnosing and debugging, we also demonstrate how to use scalable, single dashboard cloud monitoring and logging tools across the hybrid testbed infrastructure. 1 1 This material is based upon work supported by the National Science Foundation under Grant No. CNS-1923692

A new generation of smartphone accessories are emerging to support 2-d and 3-d printing and model creation. These mobile, handheld devices require continuous self-location with position accuracy approaching 0.001 inch, well beyond the capabilities of commercial, consumer-grade wireless positioning technologies. We consider the problem of accurately tracking the motion of a handheld device that provides free- hand, high-resolution image `drawing' while being swiped across a fixed planar surface. We review the capabilities and limitations of existing short-range localization technologies, including optical navigation sensors, ultrasonic positioning devices, and Inertial Measurement Units (IMU). We describe the testing apparati we constructed to establish the ground truth position of a device. We show how combining the complementary capabilities of these sensors can accurately locate a handheld device, potentially enabling a new class of smartphone imaging peripherals and applications.

The demand for proximity-based services for mobile device users inside smart buildings has driven a surge in the installation of large-scale, ultradense Bluetooth Low Energy beacon deployments. But managing a large-scale beacon infrastructure presents new systems challenges, including managing battery life, replacing misplaced or stolen beacons, detecting and locating foreign or malicious beacons, and engineering beacon locations to optimize wireless resources. We study two problems whose solutions facilitate large-scale beacon network management. First, we develop a prototype system to demonstrate how conventionally static beacon advertisements can be made time-varying, and argue that time-varying advertisements can convey important device health and network state information. We then consider a fundamental question in site engineering; calculating upper limits on the spatial density of deployed beacons to optimize either the number of advertisements received, or mobile users reached. Using analysis and simulation, we show that with current wireless standards beacon separation should be at least 0.5 meters per advertisement channel. Knowledge of this bound facilitates the construction of topology design rules, and quantifies previously unknown limits on beacon density in ultra-dense deployments.

We introduce an indoor positioning technology based on an enterprise 3G UMTS radio access network and unmodified mobile phones. We describe the architecture and operation of a prototype that we have developed, and present empirical operational results from multiple enterprise settings. We discuss the relative advantages of using an enterprise RAN platform for indoor positioning, and examine both the challenges shared with alternative RF-based positioning technologies (e.g., WiFi), as well as the new and unique challenges that indoor cellular introduces. Finally, we present empirical results on location accuracy, and discuss how we seek to further improve accuracy using smartphone-based sensors and multiple radios.

We improve indoor positioning location accuracy by opportunistically enlisting wireless client devices to temporarily serve as unmanaged beacons (or access points). A two-phase positioning scheme is proposed; in the initial phase, a target node and its neighbors are located using managed APs with known locations. In the second phase, the target node's position is refined by neighbor nodes with uncertain locations serving as short-range wireless beacons themselves. Through simulations and analysis we explore the geographic conditions and noise models under which exploiting the availability of unmanaged beacons can improve location accuracy. We argue that as the spatial density of wireless devices grows, it is increasingly desirable to call on a preferred subset of neighbor nodes with uncorrelated anchor errors to serve as beacons to improve location accuracy.

Spontaneous and robust mobile device location authentication can be realized by supplementing existing 802.11x access points (AP) with small cells. We show that by transferring network traffic to a mobile computing device associated with a femtocell while remotely monitoring its ingress traffic activity, any internet-connected sender can verify the cooperating receiver's location. We describe a prototype non-cryptographic location authentication system we constructed, and explain how to design both voice and data transmissions with distinct, discernible traffic signatures. Using both analytical modeling and empirical results from our implementation, we demonstrate that these signatures can be reliably detected even in the presence of heavy cross-traffic introduced by other femtocell users.

Certain location-based services seek to spontaneously authenticate user location without the need to have a pre-existing relationship with each user, or with each location provider. We introduce an intelligent infrastructure-based solution that provides spontaneous, rapid, and robust mobile device location authentication by supplementing existing 802.11x APs with femtocells. We show that by transferring data to a mobile computing device associated with a femtocell while remotely monitoring its traffic activity, a sender can verify the cooperating receiver’s location. We describe a prototype femtocell-based location authentication system we constructed, and explain how to use rate-control to construct data transmissions with distinct traffic signatures that can be reliably detected even in the presence of heavy cross-traffic introduced by other femtocell users. Neither mobile operators nor location providers need be aware that an authentication is taking place.

Certain location-based services seek to spontaneously authenticate user location without the need to have a pre-existing relationship with each user, or with each location provider. We introduce an intelligent infrastructure-based solution that provides spontaneous, rapid, and robust mobile device location authentication by supplementing existing 802.11x APs with femtocells. We show that by transferring data to a mobile device associated with a femtocell while remotely monitoring its traffic activity, a sender can verify the cooperating receiver’s location. We explain how to design data transmissions with distinct traffic signatures that can be rapidly and reliably detected even in the presence of heavy cross-traffic introduced by other femtocell users. Neither mobile operators nor location providers need be aware that an authentication is taking place.

Providers of location-based services seek new methods to authenticate the location of their clients. We propose a novel infrastructure-based solution that provides spontaneous and transaction-oriented mobile device location authentication via an integrated 802.11× wireless access point and 3G femtocell access system. By simply making a voice call while remotely monitoring femtocell activity, a calling party can verify a (co-operating) called party's location even when the participants have no pre-existing relationship. We show how such a traffic signature can be reliably detected even in the presence of heavy cross-traffic introduced by other femtocell users. We describe how the verification proceeds without revealing details of the authentication - or even the parties involved - to the location provider.

The growing use of virtualization technologies in settings such as multi-tenant compute and storage clouds challenges us to both specify and enforce the isolation of clients sharing network and compute resources. In this paper we propose a novel analytical measure of performance isolation for shared resource systems serving multiple traffic flows or computing workloads. By basing our isolation measure on well-known results from network calculus, we demonstrate how isolation metrics can be calculated for flows traversing both individual system elements and networks of those elements. We argue that this measure facilitates the design of systems capable of ensuring that clients can realize a specified isolation target. We present illustrative examples of how our quantitative isolation measure can be used to compare the isolation properties of alternative instantiations of resource sharing systems ranging from experimental testbeds to dynamically-instantiated compute clouds. Finally, we show how next generation resource allocators can be designed to preserve the isolation of clients by either routing newly arriving flows, or re-arrange existing flows.

Mobile device users are increasingly incented to falsify their locations to retain location privacy while capturing economic benefits such as location-based retail discounts. Location spoofing is easily achieved with several widely-used location services that rely on smartphone applications to convey GPS coordinates, IP addresses, or WiFi Positioning System radio environment data. In earlier work we introduced a network infrastructure-based system that provides spontaneous, rapid, and robust mobile device location authentication by supplementing existing 802.11x APs with off-the-shelf femtocells. The proposed system has the property of leveraging mobile operator infrastructure, without requiring operator participation in either providing or authenticating location. In this paper we present a security analysis of the location authentication system. We assess its resistance to DoS attacks, identify various approaches for a mobile user to deceive a location verifier with and without the assistance of a colluder, and explore the tradeoffs between cost and complexity in mounting such attacks. Finally, we identify a collection of system modifications and countermeasures to anticipated attacks designed to decrease location authentication system vulnerabilities and increase privacy protection.