Abstract—Using data from 43 users across two platforms, we present a detailed look at smartphone traffic. We find that browsing contributes over half of the traffic, while each of email, media, and maps contribute roughly 10%. We also find that the overhead of lower layer protocols is high because of small transfer sizes. For half of the transfers that use transport-level security, header bytes correspond to 40 % of the total. We show that while packet loss is the main factor that limits the throughput of smartphone traffic, larger send buffers at Internet servers can improve the throughput of a quarter of the transfers. Finally, by studying the interaction between smartphone traffic and the radio power management policy, we find that the power consumption of the radio can be reduced by 35 % with minimal impact on the performance of packet exchanges. Categories and Subject Descriptors C.2.5 [Computer-Communication Networks] Local and Wide-

Conventional wisdom holds that storage is not a big contributor to application performance on mobile devices. Flash storage (the type most commonly used today) draws little power, and its performance is thought to exceed that of the network subsystem. In this paper we present evidence that storage performance does indeed affect the performance of several common applications such as web browsing, Maps, application install, email, and Facebook. For several Android smartphones, we find that just by varying the underlying flash storage, performance over WiFi can typically vary between 100 % to 300 % across applications; in one extreme scenario the variation jumped to over 2000%. We identify the reasons for the strong correlation between storage and application performance to be a combination of poor flash device performance, random I/O from application databases, and heavy-handed use of synchronous writes; based on our findings we implement and evaluate a set of pilot solutions to address the storage performance deficiencies in smartphones. 1

Advances in mobile phone technology have allowed phones to become a convenient platform for real-time assessment of a participants health and behavior. AndWellness, a personal data collection system, uses mobile phones to collect and analyze data from both active, triggered user experience samples and passive logging of onboard environmental sensors. The system includes an application that runs on Android based mobile phones, server software that manages deployments and acts as a central repository for data, and a dashboard front end for both participants and researchers to visualize incoming data in real-time. Our system has gone through testing by researchers in preparation for deployment with participants, and we describe an initial qualitative study plus several planned future studies to demonstrate how our system can be used to better understand a user’s health related habits and observations.

System energy models are important for energy optimization and management in mobile systems. However, existing system energy models are built in a lab setting with the help from a second computer. Not only are they labor-intensive; but also they do not adequately account for the great diversity in the hardware and usage of mobile systems. Moreover, existing system energy models are intended for energy estimation for time intervals of one second or longer; they do not provide the required rate for fine-grain use such as per-application energy accounting. In this work, we study a self-modeling paradigm in which a mobile system automatically generates its energy model without any external assistance. Our solution, Sesame, leverages the possibility of self power measurement through the smart battery interface and employs a suite of novel techniques to achieve accuracy and rate much higher than that of the smart battery interface. We report the implementation and evaluation of Sesame on a laptop and a smartphone. The experiment results show that Sesame is able to generate system energy models of 95 % accuracy at one estimation per second and of 88 % accuracy at one estimation per 10 ms, without any external assistance. Two fiveday field studies with four laptop and four smartphones users further demonstrate the effectiveness, efficiency, and noninvasiveness of Sesame.

With the recent advent of 4G LTE networks, there has been increasing interest to better understand the performance and power characteristics, compared with 3G/WiFi networks. In this paper, we take one of the first steps in this direction. Using a publicly deployed tool we designed for Android called 4GTest attracting more than 3000 users within 2 months and extensive local experiments, we study the network performance of LTE networks and compare with other types of mobile networks. We observe LTE generally has significantly higher downlink and uplink throughput than 3G and even WiFi, with a median value of 13Mbps and 6Mbps, respectively. We develop the first empirically derived comprehensive power model of a commercial LTE network with less than 6 % error rate and state transitions matching the specifications. Using a comprehensive data set consisting of 5-month traces of 20 smartphone users, we carefully investigate the energy usage in 3G, LTE, and WiFi networks and evaluate the impact of configuring LTE-related parameters. Despite several new power saving improvements, we find that LTE is as much as 23 times less power efficient compared with WiFi, and even less power efficient than 3G, based on the user traces and the long high power tail is found to be a key contributor. In addition, we perform case studies of several popular applications on Android in LTE and identify that the performance bottleneck for web-based applications lies less in the network, compared to our previous study in 3G [24]. Instead, the device’s processing power, despite the significant improvement compared to our analysis two years ago, becomes more of a bottleneck. Categories and Subject Descriptors C.2.1 [Network Architecture and Design]: wireless communication; C.4 [Performance of Systems]: measurement techniques,

Cloud services accessed through mobile devices suffer from high network access latencies and are constrained by energy budgets dictated by the devices ’ batteries. Radio and battery technologies will improve over time, but are still expected to be the bottlenecks in future systems. Non-volatile memories (NVM), however, may continue experiencing significant and steady improvements in density for at least ten more years. In this paper, we propose to leverage the abundance in memory capacity of mobile devices to mitigate latency and energy issues when accessing cloud services. We first analyze NVM technology scaling trends, and then propose a cloud service cache architecture that resides on the mobile device’s NVM (pocket cloudlet). This architecture utilizes both individual user and community access models to maximize its hit rate, and subsequently reduce overall service latency and energy consumption. As a showcase we present the design, implementation and evaluation of PocketSearch, a search and advertisement pocket cloudlet. We perform mobile search characterization to guide the design of PocketSearch and evaluate it with 200 million mobile queries from the search logs of m.bing.com. We show that PocketSearch can serve, on average, 66 % of the web search queries submitted by an individual user without having to use the slow 3G link, leading to 16x service access speedup. Finally, based on experience with PocketSearch we provide additional insight and guidelines on how future pocket cloudlets should be organized, from both an architectural and an operating system perspective.

Displays based on organic light-emitting diode (OLED) technology are appearing on many mobile devices. Unlike liquid crystal displays (LCD), OLED displays consume dramatically different power for showing different colors. In particular, OLED displays are inefficient for showing bright colors. This has made them undesirable for mobile devices because much of the web content is of bright colors. To tackle this problem, we present the motivational studies, design, and realization of Chameleon, a color adaptive web browser that renders web pages with power-optimized color schemes under user-supplied constraints. Driven by the findings from our motivational studies, Chameleon provides end users with important options, offloads tasks that are not absolutely needed in real-time, and accomplishes real-time tasks by carefully enhancing the codebase of a browser engine. According to measurements with OLED smartphones, Chameleon is able to reduce average system power consumption for web browsing by 41 % and is able to reduce display power consumption by 64% without introducing any noticeable delay.

We report the first field study of real-world tracking via mobile apps in which we measured how 20 participants were tracked over three weeks as they exercised their Android smartphone apps. We instrumented the phones with dynamic taint tracking to record communications that exposed identifying information, and inspected web cookie databases. We find that 36 % of the sites (655 out of 1824) that our study apps are programmed to contact are tracking users. Of these sites, 37 % track users with persistent identifiers (mostly AndroidId and IMEI) derived from an identifying string unique to the user’s device. This is privacy risk as these IDs are long-lived and enable cross-application and cross-site profiling of the user; they are often sent without encryption and with geo-location too. Advertising and analytics services are widely used, being embedded in 57 % of apps and tracking every single participant in our study. Most of these sites are heavy trackers: 25 % of their tracking is done with persistent identifiers and geo-location is gathered by half of the top 10 advertisers. To one participant’s surprise, over 600 geocoordinates were exposed to a third-party advertising site during the study. Our participants expected to be tracked but wanted greater transparency and the ability to opt out in cases when there was no perceived value. To let them opt out, we prototype a privacy control that selectively shadows access to identifiers and other sensitive personal data by third-party components. 1.

In this paper, we study the Location-based Reporting Tool (LRT), asmartphoneapplicationforcollectinglarge-scalefeedback from mobile customers. Using one-year data collected from one of the largest cellular networks in the US, we compare LRT feedback to the traditional customer feedback channel – customer care tickets. Our analysis shows that, due to the light-weight design, LRT encourages customers to report more problems from anywhere and at any time. In addition, we find LRT users access network services more intensively than other mobile users, and hence aremore likely to experience and are more sensitive to network problems. All these render LRT feedback a valuable information source for early detection of emerging network problems.

The ultimate goal of a computer design is to satisfy the end user. However, the design and optimization of computer architectures have largely left the user out of the loop. In this dissertation, I make the case that with modern computer architectures it is becoming increasingly important to take the end user into account. I then propose three specific aspects of the end user that should be explored when incorporating the end user into loop; (1) user perception, (2) user state, and (3) user activity. First, I show that that computer architects should study the end user’s perception of performance relative to actual hardware performance. User studies show that for satisfaction across different users. This variation represents opportunity for optimizing computer architectures subject to individual user satisfaction. Second, I make the case for measuring user state via empathic input devices, input devices providing a computer with information about user state. I demonstrate that three example empathic input devices (eye tracking, a galvanic skin response sensor, and force sensors) can be useful for understanding changes in user satisfaction for driving power optimizations. Third, I show