Figure 5 - uploaded by Matteo Varvello
Content may be subject to copyright.
Source publication
Recent advances in cloud computing have simplified the way that both software development and testing are performed. Unfortunately, this is not true for battery testing for which state of the art test-beds simply consist of one phone attached to a power meter. These test-beds have limited resources, access, and are overall hard to maintain; for the...
Similar publications
Large neural networks are difficult to deploy on mobile devices because of intensive computation and storage. To alleviate it, we study ternarization, a balance between efficiency and accuracy that quantizes both weights and activations into ternary values. In previous ternarized neural networks, a hard threshold Δ is introduced to determine quanti...
Citations
... This protocol is universally recognized by all devices [17]. It is designed to establish a dependable connection, accepting a data stream within local processes, breaking it down into segments not exceeding 64 Kbytes, and transmitting each segment as a separate ip datagram. ...
El siguiente documento presenta un dispositivo de conectividad inalámbrica diseñado específicamente para el sector educativo, que emplea tecnología IoT y Raspberry Pi. Su objetivo principal es ofrecer a las instituciones educativas una solución asequible que permita realizar presentaciones en el aula sin problemas y sin comprometer la calidad ni la facilidad de uso.Utilizando una metodología experimental, configuramos un ordenador monoplaca Raspberry Pi, transformando este dispositivo en una alternativa práctica a un ordenador de sobremesa tradicional. Esta configuración permite la reproducción de varios tipos de documentos, incluidos Word, Excel, PowerPoint y Publisher, ofreciendo a educadores y estudiantes una herramienta versátil para satisfacer sus necesidades académicas. La solución propuesta funciona a la perfección: mediante el uso de una aplicación móvil gratuita, los contenidos pueden transmitirse sin esfuerzo a dispositivos externos, como proyectores, televisores y monitores. De este modo se aborda un problema educativo cuyo principal inconveniente radica en el coste de los equipos.En última instancia, el objetivo no es sólo superar las barreras financieras en la educación, sino también impulsar la exploración innovadora para mejorar la conectividad y la automatización en diversos entornos.
... To handle big numbers, we use the high-performance Rust library of Arithmetic in Multiple Precision (RAMP) [29]. We perform a series of measurements on the power discharge (in mAh) of the battery of three Google Pixel devices (Table III), using the BatteryLab infrastructure [44] that operationalises a Monsoon High Voltage Power Monitor [32]. Additionally, we measured the CPU utilization (in %) and the end-to-end latency (in sec) of the smallest possible synergy. ...
Distributed (or Federated) learning enables users to train machine learning models on their very own devices, while they share only the gradients of their models usually in a differentially private way (utility loss). Although such a strategy provides better privacy guarantees than the traditional centralized approach, it requires users to blindly trust a centralized infrastructure that may also become a bottleneck with the increasing number of users. In this paper, we design and implement P4L: a privacy preserving peer-to-peer learning system for users to participate in an asynchronous, collaborative learning scheme without requiring any sort of infrastructure or relying on differential privacy. Our design uses strong cryptographic primitives to preserve both the confidentiality and utility of the shared gradients, a set of peer-to-peer mechanisms for fault tolerance and user churn, proximity and cross device communications. Extensive simulations under different network settings and ML scenarios for three real-life datasets show that P4L provides competitive performance to baselines, while it is resilient to different poisoning attacks. We implement P4L and experimental results show that the performance overhead and power consumption is minimal (less than 3mAh of discharge).
... The power meter enables voltage up to 13.5V and provides up to 6A of continuous current, at 5KHz sampling rate. It is considered a good candidate for this type of measures (total energy consumption) given its precision and previous usage in the field in similar scientific studies [18] [34]. We measure the current consumed over a period of time (profiling time), given an input value for voltage. ...
... We have released WPM as a web application integrated with BatteryLab which offers such testing capabilities to the public, in real time. This paper extends our previously published work [44] in many ways: ...
Advances in cloud computing have simplified the way that both software development and testing are performed. This is not true for battery testing for which state of the art test-beds simply consist of one phone attached to a power meter. These test-beds have limited resources, access, and are overall hard to maintain; for these reasons, they often sit idle with no experiment to run. In this paper, we propose to share existing battery testbeds and transform them into vantage points of BatteryLab, a power monitoring platform offering heterogeneous devices and testing conditions. We have achieved this vision with a combination of hardware and software which allow to augment existing battery test-beds with remote capabilities. BatteryLab currently counts three vantage points, one in Europe and two in the US, hosting three Android devices and one iPhone 7. We benchmark BatteryLab with respect to the accuracy of its battery readings, system performance, and platform heterogeneity. Next, we demonstrate how measurements can be run atop of BatteryLab by developing the "Web Power Monitor" (WPM), a tool which can measure website power consumption at scale. We released WPM and used it to report on the energy consumption of Alexa's top 1,000 websites across 3 locations and 4 devices (both Android and iOS).
... Thus any particular connectivity/peering of the Azure network might influence our cloud experiments. Ideally, the testbed should be deployed across multiple cloud providers to mitigate any artifact of a single provider, or even encompass distributed edge-based platforms provisioned across heterogeneous access networks (e.g., residential [20,38], campus [2] and enterprise networks [11,12]). In fact, moving the evaluation platform to the edge would allow us to extend the list of target videoconferencing systems to study. ...
Since the outbreak of the COVID-19 pandemic, videoconferencing has become the default mode of communication in our daily lives at homes, workplaces and schools, and it is likely to remain an important part of our lives in the post-pandemic world. Despite its significance, there has not been any systematic study characterizing the user-perceived performance of existing videoconferencing systems other than anecdotal reports. In this paper, we present a detailed measurement study that compares three major videoconferencing systems: Zoom, Webex and Google Meet. Our study is based on 48 hours' worth of more than 700 videoconferencing sessions, which were created with a mix of emulated videoconferencing clients deployed in the cloud, as well as real mobile devices running from a residential network. We find that the existing videoconferencing systems vary in terms of geographic scope, which in turns determines streaming lag experienced by users. We also observe that streaming rate can change under different conditions (e.g., number of users in a session, mobile device status, etc), which affects user-perceived streaming quality. Beyond these findings, our measurement methodology can enable reproducible benchmark analysis for any types of comparative or longitudinal study on available videoconferencing systems.
... In the same way as Espresso can automatically generate testing code from human input, Cappuccino automatically generates automation for third party apps. We integrated Cappuccino with Batterylab [27] -a research platform for battery measurements -to realize a fully transparent and extensible browser testing suite. ...
... Fortunately, the research community has recently released BatteryLab [27], a testbed which largely simplifies battery measurements. In short, BatteryLab consists of a set of remote devices connected to power meters where experimenters can run ad-hoc experiments. ...
Mobile web browsing has recently surpassed desktop browsing both in term of popularity and traffic. Following its desktop counterpart, the mobile browsers ecosystem has been growing from few browsers (Chrome, Firefox, and Safari) to a plethora of browsers, each with unique characteristics (battery friendly, privacy preserving, lightweight, etc.). In this paper, we introduce a browser benchmarking pipeline for Android browsers encompassing automation, in-depth experimentation, and result analysis. We tested 15 Android browsers, using Cappuccino a novel testing suite we built for third party Android applications. We perform a battery-centric analysis of such browsers and show that: 1) popular browsers tend also to consume the most, 2) adblocking produces significant battery savings (between 20 and 40% depending on the browser), and 3) dark mode offers an extra 10% battery savings on AMOLED screens. We exploit this observation to build AttentionDim, a screen dimming mechanism driven by browser events. Via integration with the Brave browser and 10 volunteers, we show potential battery savings up to 30%, on both devices with AMOLED and LCD screens.
... We have argued that raw CPU usage is not robust (See §V). Further, power consumption cannot be accurately monitored accurately using resources only available within Web browsers [42]. Finally, while we have also shown the effectiveness of WebSocket activity in detecting miners, CoinSpy shows how all three behavioral resources of CPU, Network, and Memory can actually be combined to perform more robust detection. ...
... The S9 also has twice as much memory (4 GB when J3 has 2 GB) and a larger battery (3,000 mAh when the battery of J3 is 2,600 mAh). The low-end device (J3) is part of Batterylab [73,74], a distributed platform for battery measurements. It follows that fine grained battery measurements (via a Monsoon High Voltage Power Monitor [75] directly connected to the device's battery) are available for this device. ...
CAPTCHA systems have been widely deployed to identify and block fraudulent bot traffic. However, current solutions, such as Google's reCAPTCHA, often either (i) require additional user actions (e.g., users solving mathematical or image-based puzzles), or (ii) need to send the attestation data back to the server (e.g., user behavioral data, device fingerprints, etc.), thus raising significant privacy concerns. To address both of the above, in this paper we present ZKSENSE: the first zero knowledge proof-based bot detection system, specifically designed for mobile devices. Our approach is completely transparent to the users and does not reveal any sensitive sensor data to the service provider. To achieve this, ZKSENSE studies the mobile device's motion sensor outputs during user actions and assess their humanness locally with the use of an ML-based classifier trained by using sensor data from public sources and data collected from a small set of volunteers. We implement a proof of concept of our system as an Android service to demonstrate its feasibility and effectiveness. In our evaluation we show that ZKSENSE detects bots without degrading the end-user experience or jeopardizing their privacy, with 91% accuracy across a variety of bot scenarios, including: (i) when the device is resting (e.g., on a table), (ii) when there is artificial movement from the device's vibration, and (iii) when the device is docked on a swinging cradle.
Since the outbreak of the COVID-19 pandemic, videoconferencing has become the default mode of communication in our daily lives at homes, workplaces and schools, and it is likely to remain an important part of our lives in the post-pandemic world. Despite its significance, there has not been any systematic study characterizing the user-perceived performance of existing videoconferencing systems other than anecdotal reports. In this paper, we present a detailed measurement study that compares three major videoconferencing systems: Zoom, Webex and Google Meet. Our study is based on 62 hours’ worth of more than 1.1K videoconferencing sessions, which were created with a mix of emulated videoconferencing clients deployed in the cloud, as well as real mobile devices running from a residential network over two separate periods with nine months apart. We find that the existing videoconferencing systems vary in terms of geographic scope and resource provisioning strategies, which in turns determine streaming lag experienced by users. We also observe that streaming rate can change under different conditions (e.g., available bandwidth, number of users in a session, mobile device status), which affects user-perceived streaming quality. Beyond these findings, our measurement methodology enables reproducible benchmark analysis for any types of comparative or longitudinal study on available videoconferencing systems.
Advances in cloud computing have simplified the way that both software development and testing are performed. This is not true for battery testing for which state of the art test-beds simply consist of one phone attached to a power meter. These test-beds have limited resources, access, and are overall hard to maintain; for these reasons, they often sit idle with no experiment to run. In this paper, we propose to share existing battery testbeds and transform them into vantage points of BatteryLab, a power monitoring platform offering heterogeneous devices and testing conditions. We have achieved this vision with a combination of hardware and software which allow to augment existing battery test-beds with remote capabilities. BatteryLab currently counts three vantage points, one in Europe and two in the US, hosting three Android devices and one iPhone 7. We benchmark BatteryLab with respect to the accuracy of its battery readings, system performance, and platform heterogeneity. Next, we demonstrate how measurements can be run atop of BatteryLab by developing the “Web Power Monitor” (WPM), a tool which can measure website power consumption at scale. We released WPM and used it to report on the energy consumption of Alexa’s top 1,000 websites across 3 locations and 4 devices (both Android and iOS).KeywordsBatteryTest-bedPerformanceAndroidiOS
