Mobility footprint refers to the size, weight, and energy de- mand of the hardware that must be carried by a mobile user to be e*ective at any time and place. The ideal of a zero mobility footprint is achievable by encapsulating per- sonal computing state in a virtual machine (VM) and de- livering it over the Internet to a locally-obtained computer close to the user. In locations with poor Internet connectiv- ity, the demands placed on WAN bandwidth can result in unacceptable user experience. We show how this challenge can be overcome by using nascent smart phone technology as a trusted personal assistant called Horatio that serves as a self-cleaning portable cache for VM state. Since most users already carry cell phones for voice calls and texting, Hora- tio does not increase the size or weight aspects of a user's mobility footprint | there is only a small increase in the energy aspect. We have built an experimental prototype of Horatio, and measurements con*rm its ability to improve user experience even with current smart phone limitations.

Kraków, Poland

International Conference on Mobile Systems, Applications, and Services (MobiSys) 2009-04-22T20:16:49-04:00 224 7 6 8 Conference 8 2009-04-22T20:26:08-04:00 2009 Virtual Machine (VM) fork is a new cloud computing abstraction that instantaneously clones a VM into multiple replicas running on different hosts. All replicas share the same initial state, matching the intuitive semantics of stateful worker creation. VM fork thus enables the straightforward creation and efficient deployment of many tasks demanding swift instantiation of stateful workers in a cloud environment, e.g. excess load handling, opportunistic job placement, or parallel computing. Lack of instantaneous stateful cloning forces users of cloud computing into ad hoc practices to manage application state and cycle provisioning. We present SnowFlock, our implementation of the VM fork abstraction. To evaluate SnowFlock, we focus on the demanding scenario of services requiring on-the-fly creation of hundreds of parallel workers in order to solve computationally-intensive queries in seconds. These services are prominent in fields such as bioinformatics, finance, and rendering. SnowFlock provides sub-second VM cloning, scales to hundreds of workers, consumes few cloud I/O resources, and has negligible runtime overhead.

Nuremberg, Germany

3rd European Conference on Computer Systems (Eurosys) 2009-02-02T07:57:10-05:00 86 5 [Best Paper Award] 6 8 Conference 8 2009-08-13T15:46:58-04:00 2009 Cloud computing promises to provide researchers with the ability to perform parallel computations using large pools of virtual machines (VMs), without facing the burden of owning or maintaining physical infrastructure. However, with ease of access to hundreds of VMs, comes also an increased management burden. Cloud users today must manually instantiate, configure and maintain the virtual hosts in their cluster. They must learn new cloud APIs that are not germane to the problem of parallel processing. Those APIs usually take several minutes to perform their VMmanagement tasks, forcing users to keep VMs idling and pay for unused processing time, rather than shut VMs down and power them on as needed. Furthermore, users must still configure their cluster management framework to launch their parallel jobs. <br /> In this paper we show that all this management pain is unnecessary. We show how to combine a cloud API – SnowFlock – and a parallel processing framework – MPI – to truly realize the potential of the cloud. SnowFlock allows users to fork VMs as if they were processes, occupying in sub-second time multiple physical hosts. We exploit the synergy between this paradigm and MPI’s job management to completely hide all details of cloud management from the user. Maintaining a single VM and starting unmodified applications with familiar MPI commands, a user can instantaneously leverage hundreds of processors to perform a parallel computation. Besides making use of cloud resources trivial, we also eliminate the cost of idling – VMs exist only for as long as they are involved in computation.

Nuremberg, Germany

3rd Workshop on System-level Virtualization for High Performance Computing (HPCVirt 2009) 2009-02-26T10:52:02-05:00 88 5 7 12 Workshop 12 2009-02-26T10:58:01-05:00 2009 Securing interactions between devices that do not know each other a priori is an important and challenging task. We present Amigo, a technique to authenticate co-located devices using knowledge of their shared radio environment as proof of physical proximity. We present evaluation results that show that our technique is robust against a range of passive and active attacks. The key advantages of our technique are that it does not require any additional hardware to be present on the devices beyond the radios that are already used for communication, it does not require user involvement to verify the validity of the authentication process, and it is not vulnerable to eavesdropping.

patito

2008-08-14T13:55:48-04:00 26 International Journal of Security and Networks (IJSN), Special Issue on Secure Spontaneous Interaction 2 1 4-16 4 InderScience Publishers 9 Journal 9 2009-04-28T15:38:06-04:00 4 2009 Today, most web pages are designed for viewing on desktop computers with access to large screens and broadband connections. On the other hand, mobile devices face severe resource constraints, such as limitations in display size, battery life and network connectivity. This result in a significant degradation in the experience of mobile users as they go about browsing the World Wide Web. I will describe PageTailor and Usage-awaRe Interactive Content Adaptation (URICA), two techniques that let end users customize web content to suit their needs. PageTailor lets the end user customize a web page graphically by clicking on page elements, such as images, as part of her normal browsing activities on the mobile device. PageTailor records the user's customizations and automatically reapplies them on subsequent visits to the same page or to other, similar pages, on the same Web site. URICA learns how to adapt content from feedback provided by users. When serving content to a mobile user, URICA first makes an initial prediction on how to customize the content. Next, URICA allows users who are unsatisfied with the system's adaptation decision to take control of the adaptation process and make changes until the content is suitably adapted for their purposes. For example, a user may choose to remove a toolbar to improve readability, or ask the system to improve the resolution of a specific image. The successful adaptation is recorded and used in making future adaptation decisions for other users.

Waterloo, Canada

2009-02-09T23:18:18-05:00 87 13 8 14 Talk 14 2009-02-09T23:20:54-05:00 Google Tech Talk 2008 CILoS is an indoor localization system based on CDMA mobile phone signal fingerprinting. CDMA networks vary their transmission power to accommodate fluctuations in network load. This affects signal intensity and therefore limits the practicality of traditional fingerprinting approaches based on receiver signal strength (RSSI) measurements. Instead, CILoS uses fingerprints of signal delay that are robust to cell resizing. We demonstrate that CILoS achieves a median accuracy of 5 meters, and compares favourably to RSSI fingerprinting systems. We highlight the significance of wide fingerprints, constructed through scanning multiple channels, for achieving high localization accuracy. We also show that our system can accurately differentiate between floors of a multifloor building.

Seoul, South Korea

10th International Conference on Ubiquitous Computing (Ubicomp) 2008-08-12T23:02:29-04:00 19 10 6 8 Conference 8 2009-02-08T19:50:42-05:00 2008 Advances in electronic location technology and the coming of age of mobile computing have opened the door for location-aware applications to permeate all aspects of everyday life. Location is at the core of a large number of high-value applications ranging from the life-and-death context of emergency response to serendipitous social meet-ups. For example, the market for GPS products and services alone is expected to grow to US$200 billion by 2015. Unfortunately, there is no single location technology that is good for every situation and exhibits high accuracy, low cost, and universal coverage. In fact, high accuracy and good coverage seldom coexist, and when they do, it comes at an extreme cost. Instead, the modern localization landscape is a kaleidoscope of location systems based on a multitude of different technologies including satellite, mobile telephony, 802.11, ultrasound, and infrared among others. This lecture introduces researchers and developers to the most popular technologies and systems for location estimation and the challenges and opportunities that accompany their use. For each technology, we discuss the history of its development, the various systems that are based on it, and their trade-offs and their effects on cost and performance. We also describe technology-independent algorithms that are commonly used to smooth streams of location estimates and improve the accuracy of object tracking. Finally, we provide an overview of the wide variety of application domains where location plays a key role, and discuss opportunities and new technologies on the horizon.

2008-08-14T13:47:42-04:00 23 8 1 Morgan and Claypool Publishers Synthesis Lectures on Mobile and Pervasive Computing 11 Book 11 2009-02-08T20:18:26-05:00 2008

patito

2008-08-14T13:49:40-04:00 24 7 3 ACM Press 15 Collection 15 2009-04-28T14:56:28-04:00 2008 We introduce Impromptu Clusters (ICs), a new abstraction that makes it possible to leverage cloud-based clusters to execute short-lived parallel tasks, for example Internet services that use parallelism to deliver near-interactive responses. ICs are particularly relevant for resource-intensive web applications in areas such as bioinformatics, graphics rendering, computational finance, and search. In an IC, an application encapsulated inside a virtual machine (VM) is swiftly forked into multiple copies that execute on different physical hosts, and then disappear when the computation ends. SnowFlock, our IC prototype, offers near-interactive response times for many highly-parallelizable workloads, achieves subsecond parallel VM clone times, and has negligible runtime overhead.

2008-08-29T11:47:42-04:00 69 Department of Computer Science, University of Toronto 7 CSRG-TR578 10 10 Technical Report 10 2009-02-09T21:19:25-05:00 2008

Boston, MA

Poster at USENIX Annual Technical Conference 2008-09-19T13:38:57-04:00 75 7 9 13 Poster 13 2009-02-09T21:22:26-05:00 2008 Existing localization algorithms, such as centroid or fingerprinting, compute the location of a mobile device based on measurements of signal strengths from radio base stations. Unfortunately, these algorithms require tedious and expensive off-line calibration in the target deployment area before they can be used for localization. In this paper, we present Calibree, a novel localization algorithm that does not require off-line calibration. The algorithm starts by computing relative distances between pairs of mobile phones based on signatures of their radio envi- ronment. It then combines these distances with the known locations of a small number of GPS-equipped phones to estimate absolute locations of all phones, effectively spreading location measurements from phones with GPS to those without. Our evaluation results show that Calibree per- forms better than the conventional centroid algorithm and only slightly worse than fingerprinting, without requiring off-line calibration. More- over, when no phones report their absolute locations, Calibree can be used to estimate relative distances between phones.

Sydney, Australia

Sixth International Conference on Pervasive Computing (Pervasive) 2008-08-14T17:08:17-04:00 27 6 6 8 Conference 8 2009-02-09T21:52:20-05:00 2008 Virtual machine (VM) migration has been proposed as a building block for mobile computing. An important challenge for VM migration is to optimize the transfer of large amounts of disk and memory state. We propose a solution based on the opportunistic replay of user interactions with applications at the GUI level. Whereas this approach results in very small replay logs that economize network utilization, replay of user interactions on a VM at the migration target site can result in divergent VM state. Cryptographic hashing techniques are used to identify and transmit only the differences. We discuss the implementation challenges of this approach, and present encouraging results from an early prototype that show savings of up to 80.5% of bytes transferred.

Napa Valley, CA

9th IEEE Workshop on Mobile Computing Systems and Applications (HotMobile) 2008-08-14T17:37:40-04:00 28 3 7 12 Workshop 12 2009-04-22T20:28:52-04:00 2008 Virtual machines (VMs) are widely used to share hardware compute and storage resources among many users. In bioinformatics, typical applications leverage grid computing, where tasks are run in parallel on many machines (a cluster). Snowflock provides dynamic cloning of individual VMs into virtual clusters for bioinformatics grid computing.

Toronto, ON

Poster at the 16th Annual International Conference Intelligent Systems for Molecular Biology (ISMB) 2008-09-21T22:35:35-04:00 84 2 9 13 Poster 13 2009-02-09T22:00:21-05:00 2008 Accurate indoor localization has long been an objective of the ubiquitous computing research community, and numerous indoor localization solutions based on 802.11, Bluetooth, ultrasound and infrared technologies have been proposed. This paper presents the first accurate GSM indoor localization system that achieves median within floor accuracy of 4 m in large buildings and is able to identify the floor correctly in up to 60% of the cases and is within 2 floors in up to 98% of the cases in tall multi-floor buildings. We report evaluation results of two case studies conducted over a course of several years, with data collected from 6 buildings in 3 cities across North America. The key idea that makes accurate GSM-based indoor localization possible is the use of wide signalstrength fingerprints. In addition to the 6-strongest cells traditionally used in the GSM standard, the wide fingerprint includes readings from additional cells that are strong enough to be detected, but are too weak to be used for efficient communication. We further show that selecting a subset of highly relevant channels for fingerprinting matching out of all available channels, further improves the localization accuracy.

patito

2008-08-13T11:12:05-04:00 20 Pervasive and Mobile Computing Journal (PMC) 13 6 698-720 4 Elsevier 9 Journal 9 2009-02-08T20:39:59-05:00 3 2007 Snowbird is a middleware system based on virtual machine (VM) technology that simplifies the development and deployment of bimodal applications. Such applications alternate between phases with heavy computationalresource needs and phases rich in user interaction. Examples include digital animation, as well as scientific, medical, and engineering diagnostic and design tools. Traditionally, these applications have been manually partitioned into distributed components to take advantage of remote computational resources, while still providing low-latency user interaction. Instead, Snowbird lets developers design their applications as monolithic units within a VM, and automatically migrates the application to the optimal execution site to achieve short completion time and crisp interactive performance. Snowbird does not require that applications be written in a specific language, or use specific libraries, and it can be used with existing applications, including closed-source ones, without requiring recompilation or relinking. Snowbird achieves these goals by augmenting VM migration with an interaction-aware migration manager, support for graphics hardware acceleration, and a wide-area peer-to-peer storage system. Experiments conducted with a number of real-world applications, including commercial closed-source tools, show that applications running under Snowbird come within 4% of optimal compute time, and provide crisp interactive performance that is comparable to native local execution. Keywords: Bimodal Applications, Migration,

Newport Beach, California

8th International Middleware Conference (Middleware) 2008-08-14T17:42:45-04:00 29 12 6 8 Conference 8 2009-04-22T20:30:16-04:00 2007 The resource impoverished environment on mobile devices results in a poor experience for users browsing the World Wide Web. Proxy-based middleware that transform content on the fly to better suit the resource conditions on a user’s device provide a promising solution to this problem. A key challenge in such systems is deciding how to adapt content, especially when the same content has multiple uses that have varying adaptation requirements. In this paper, we show that it is possible to provide fine grain adaptation of multi-purpose content by detecting correlations in the adaptation requirements of past users across multiple objects on a web site, and using this history to make adaptation predictions for users encountered subsequently. To evaluate our technique, we built prototype page layout and image fidelity adaptation systems, and used these to gather traces from users browsing multi-purpose web content in a laboratory setting. Our experimental results show that using correlations to make adaptation predictions can significantly reduce bandwidth consumption, browsing time, energy usage and user effort required to adapt content.

Newport Beach, California

8th International Middleware Conference (Middleware) 2008-08-14T22:28:38-04:00 30 12 6 8 Conference 8 2009-02-09T21:53:14-05:00 2007 We show that simple radio propagation and node mobility models widely used in MANET evaluation are not robust in indoor environments. Robust simulation models let researchers extrapolate simulation results and reach reliable conclusions about expected protocol performance. We experiment with two representative MANET routing protocols under dif- ferent mobility and radio propagation models with decreasing levels of complexity. We show that the effects of successive simplifications to the node mobility and radio propagation models are not consistent across protocols. Moreover, even within the same protocol, the effects on perfor- mance can vary erratically as simulation parameters change. Our results raise troubling questions about the soundness of evaluations of MANET routing protocols based on simple radio propagation and node mobility models.

patito

2008-08-13T16:43:59-04:00 22 Ad Hoc & Sensor Wireless Networks Journal 11 4 4 Old City Publishing, Inc. 9 Journal 9 2009-02-08T20:43:04-05:00 4 2007 Securing interactions between devices that do not know each other a priori is an important and challenging task. We present Amigo, a technique to authenticate co-located devices using knowledge of their shared radio environment as proof of physical proximity. We present evaluation results that show that our technique is robust against a range of passive and active attacks. The key advantages of our technique are that it does not require any additional hardware to be present on the devices beyond the radios that are already used for communication, it does not require user involvement to verify the validity of the authentication process, and it is not vulnerable to eavesdropping.

Innsbruck, Austria

9th International Conference on Ubiquitous Computing (Ubicomp) 2008-08-14T22:52:45-04:00 31 10 6 8 Conference 8 2009-02-09T21:54:54-05:00 2007 Many existing localization systems generate location predictions, but fail to report how accurate the predictions are. This paper explores the e*ect of revealing the error of location predictions to the end-user in a location *nding *eld study. We report *ndings obtained under four di*erent error visualization conditions and show signi*cant bene*t in revealing the error of location predictions to the user in location *nding tasks. We report the observed in uences of error on participants' strategies for location *nding. Additionally, given the observed bene*t of a dynamic estimate of error, we design practical algorithms for estimating the error of a location prediction. Analysis of the algorithms shows a median estimation inaccuracy of up to 50m from the predicted location's true error.

Innsbruck, Austria

9th International Conference on Ubiquitous Computing (Ubicomp) 2008-08-14T23:01:45-04:00 32 10 6 8 Conference 8 2009-02-09T21:55:51-05:00 2007 The popularity of handheld devices has created a flurry of research activity into new protocols and applications that can handle and exploit the defining characteristic of this new environment – user mobility. In addition to mobility, another defining characteristic of mobile systems is user social interaction. This paper investigates how mobile systems could exploit people’s social interactions to improve these systems’ performance and query hit rate. For this, we build a trace-driven simulator that enables us to re-create the behavior of mobile systems in a social environment. We use our simulator to study three diverse mobile systems: DTN routing protocols, firewalls preventing a worm infection, and a mobile P2P file-sharing system. In each of these three cases, we find that mobile systems can benefit substantially from exploiting social information.

Innsbruck, Austria

9th International Conference on Ubiquitous Computing (Ubicomp) 2008-08-14T23:11:02-04:00 33 10 6 8 Conference 8 2009-02-09T21:57:34-05:00 2007 This paper presents Haggle, an architecture for mobile devices that enables seamless network connectivity and application functionality in dynamic mobile environments. Current applications must contain significant network binding and protocol logic, which makes them inflexible to the dynamic networking environments facing mobile devices. Haggle allows separating application logic from transport bindings so that applications can be communication agnostic. Internally, the Haggle framework provides a mechanism for late-binding interfaces, names, protocols, and resources for network communication. This separation allows applications to easily utilize multiple communication modes and methods across infrastructure and infrastructure-less environments. We provide a prototype implementation of the Haggle framework and evaluate it by demonstrating support for two existing legacy applications, email and web browsing. Haggle makes it possible for these applications to seamlessly utilize mobile networking opportunities both with and without infrastructure.

Innsbruck, Austria

9th International Conference on Ubiquitous Computing (Ubicomp) 2008-08-14T23:17:18-04:00 34 10 6 8 Conference 8 2009-02-09T21:58:14-05:00 2007 Secure and spontaneous communication between wireless devices that come within close proximity of each other, but lack a pre-existing trust relationship -- devices that are previously unknown to each other -- is an important component of many future pervasive applications. For example, patrons at a bar, guests at a party or conference participants may use their mobile phones to exchange private contact information over Bluetooth or WiFi. Consumers may use their mobile devices as electronic wallets to pay for tickets at the train station or groceries at the store. A user may take advantage of resources available in the environment by pairing their mobile phone to a public full-sized display and keyboard, or share music by pairing their MP3 player to a friend's home entertainment system. In this talk, I will introduce a technique that authenticates devices in close proximity by using knowledge of their shared radio environment as proof of physical proximity. I will describe Amigo, a WiFi based prototype that is robust against a range of passive and active attacks. The key advantages of Amigo are that it does not require any additional hardware to be present on the devices beyond the radios that are already used for communication, it does not require user involvement to verify the validity of the authentication process, and it is not vulnerable to eavesdropping.

Mountain View, CA

2008-09-03T10:15:16-04:00 70 8 8 14 Talk 14 2009-02-09T21:49:23-05:00 Google Tech Talk 2007 Most pages on the Web are designed for the desktop environment and render poorly on the small screens available on handheld devices. We introduce Reusable End-User Customization (REUC), a technique that lets end users adapt the layout of Web pages by removing, resizing and moving page elements. REUC records the user’s customizations and automatically reapplies them on subsequent visits to the same page or to other, similar pages, on the same Web site. We present PageTailor, a REUC prototype based on the Minimo Web browser that runs on Windows Mobile PDAs. We show that users can utilize PageTailor to adapt sophisticated Web sites, such as Amazon, BBC and MSN, for browsing on a PDA. Moreover, the customizations remain effective for up to a year, even as the content of pages is updated, and can be reused across similar pages, limiting the customization effort required to browse a site.

San Juan, Puerto Rico

5th International Conference on Mobile Systems, Applications, and Services (MobiSys) 2008-08-14T23:24:40-04:00 35 7 6 8 Conference 8 2009-02-09T21:56:45-05:00 2007 This paper describes VMGL, a cross-platform OpenGL virtualization solution that is both virtual machine monitor (VMM) and graphics processing unit (GPU) independent. VMGL allows applications executing within virtual machines (VMs) to leverage hardware rendering acceleration, thus solving a problem that has limited virtualization of a growing class of graphics-intensive applications. VMGL also provides applications running within VMs with suspend and resume capabilities across GPUs from different vendors. Our experimental results from a number of graphics-intensive applications show that VMGL provides excellent rendering performance, coming within 14% or better of native graphics hardware acceleration. Further, VMGL’s performance is two orders of magnitude better than that of software rendering, the commonly available alternative today for graphics-intensive applications running in virtualized environments. Our results confirm VMGL’s portability across VMware Workstation and Xen (on VT and non-VT hardware), and across Linux (with and without paravirtualization), FreeBSD, and Solaris. Finally, the resource demands of VMGL align well with the emerging trend of multi-core processors.

San Diego, CA

3rd ACM SIGPLAN/SIGOPS Conference on Virtual Execution Environments (VEE) 2008-08-14T23:27:50-04:00 36 7 6 8 Conference 8 2009-02-09T22:03:35-05:00 2007 When a mobile user dials 911, a key to arriving to the emergency scene promptly is knowing the location of the mobile user. This paper presents SkyLoc, a GSM fingerprinting-based localization system that runs on a mobile phone and identifies the current floor of a user in tall multi-floor buildings. Knowing the floor in a tall building significantly reduces the area that emergency service personnel have to canvas to locate the individuals in need. We evaluated our system in three multi-floor buildings located in Washington DC, Seattle and Toronto. Our system identifies the floor correctly in up to 73% of the cases and is within 2 floors in 97% of the cases. The system is robust as it works for different network operators, when the training and testing sets were collected with different hardware and up to one month apart. In addition, we show that feature selection techniques that select a subset of highly relevant radio sources for fingerprint matching nearly double the localization accuracy of our system.

White Plains, NY

5th IEEE International Conference on Pervasive Computing and Communications (PerCom) 2008-08-14T23:30:48-04:00 37 4 6 8 Conference 8 2009-02-09T22:04:02-05:00 2007 With the proliferation of mobile devices, spontaneous interactions between co-located devices that do not know each other a priori will become commonplace. Securing these interactions against eavesdropping and man-in-themiddle attacks is an important and challenging task. In this paper, we postulate that mobile devices that are positioned in close proximity may be able to derive a shared secret to secure their communication by monitoring fluctuations in the signal strength of existing ambient radio sources (GSM cell towers or WiFi access points) in their common environment. We explore the feasibility of deriving locationbased secrets and describe two approaches for how such a secret could be used to secure spontaneous communication. Deriving location-based secrets is a hard problem because while the radio environment perceived by various devices in close proximity is similar, it is not identical.

Tucson, AZ

IEEE Workshop on Mobile Computing Systems and Applications (HotMobile) 2008-08-14T23:33:44-04:00 38 3 7 12 Workshop 12 2009-02-09T22:14:36-05:00 2007

patito

2008-08-14T13:52:28-04:00 25 2 3 IEEE Press 15 Collection 15 2009-02-08T20:25:39-05:00 2007 Over the past year, there have been several reports of malicious code exploiting vulnerabilities in the Bluetooth protocol. While the research community has started to investigate a diverse set of Bluetooth security issues, little is known about the feasibility and the propagation dynamics of a worm in a Bluetooth environment. This paper is an initial attempt to remedy this situation. We start by showing that the Bluetooth protocol design and implementation is large and complex. We gather traces and we use controlled experiments to investigate whether a large-scale Bluetooth worm outbreak is viable today. Our data shows that starting a Bluetooth worm infection is easy, once a vulnerability is discovered. Finally, we use trace-drive simulations to examine the propagation dynamics of Bluetooth worms. We find that Bluetooth worms can infect a large population of vulnerable devices relatively quickly, in just a few days.

Alexandria, VA

ACM Workshop on Rapid Malcode (WORM) 2008-08-15T00:01:57-04:00 41 12 7 12 Workshop 12 2009-02-09T22:15:04-05:00 2006 Recognition of everyday physical activities is difficult due to the challenges of building informative, yet unobtrusive sensors. The most widely deployed and used mobile computing device today is the mobile phone, which presents an obvious candidate for recognizing activities. This paper explores how coarse-grained GSM data from mobile phones can be used to recognize high-level properties of user mobility, and daily step count. We demonstrate that even without knowledge of observed cell tower locations, we can recognize mobility modes that are useful for several application domains. Our mobility detection system was evaluated with GSM traces from the everyday lives of three data collectors over a period of one month, yielding an overall average accuracy of 85%, and a daily step count number that reasonably approximates the numbers determined by several commercial pedometers.

Irvine, CA

8th International Conference on Ubiquitous Computing (UbiComp) 2008-08-15T00:07:25-04:00 42 10 6 8 Conference 8 2009-02-09T22:04:39-05:00 2006 Radio equipped mobile devices have enjoyed tremendous growth in the past few years. We observe that in the near future it might be possible to build a network that routes delay-tolerant packets by harnessing user mobility and the pervasive availability of wireless devices. Such a delay-tolerant network could be used to supplement wireless infrastructure or provide service where none is available. Since mobile devices in a delay-tolerant network forward packets to nearby users, the devices can use short-range radio, which potentially reduces device power consumption and radio contention. The design of a user mobility based delay-tolerant network raises two key challenges: determining the connectivity of such a network, and determining the latency characteristics and replication requirements of routing algorithms in such a network. To determine realistic contact patterns, we collected user mobility data by conducting two user studies. We outfitted groups of students with instrumented wireless-enabled PDAs that logged pairwise contacts between study participants over a period of several weeks. Experiments conducted on these traces show that it is possible to form a delay-tolerant network based on human mobility. The network has good connectivity, so that routes exist between almost all study participants via some multi-hop path. Moreover, it is possible to effectively route packets with modest replication.

San Jose, CA

3rd International Conference on Mobile and Ubiquitous Systems: Networks and Services (MOBIQUITOUS) 2008-08-15T00:12:23-04:00 43 8 6 8 Conference 8 2009-02-09T22:05:16-05:00 2006 We introduce a novel infrastructure supporting automatic updates for dynamic content browsing on resource constrained mobile devices. Currently, the client is forced to continuously poll for updates from potentially di*erent data sources, such as, e-commerce, on-line auctions, stock and weather sites, to stay up to date with potential changes in content. We employ a pair of proxies, located on the mobile client and on a fully-connected edge server, respectively, to minimize the battery consumption caused by wireless data transfers to and from the mobile device. The client speci*es her interest in changes to speci*c parts of pages by highlighting portions of already loaded web pages in her browser. The edge proxy polls the web servers involved, and if relevant changes have occurred, it aggregates the updates as one batch to be sent to the client. The proxy running on the mobile device can pull these updates from the edge proxy, either on-demand or periodically, or can listen for pushed updates initiated by the edge proxy. We also use SMS messages to indicate available updates and to inform the user of which pages have changed. Our approach is fully implemented using two alternative wireless networking technologies, 802.11 and GPRS. Furthermore, we leverage our SMS feature to implement and evaluate a hybrid approach which chooses either 802.11 or GPRS depending on the size of the update batch. Our evaluation explores the data transfer savings enabled by our proxy-based infrastructure and the energy consumption when using each of the two networking capabilities and the hybrid approach. Our results show that our proxy system saves data transfers to and from the mobile device by an order of magnitude and battery consumption by up to a factor of 4.5, compared to the client-initiated continuous polling approach. Our results also show that the batching e*ect of our proxy reduces energy consumption even in the case where the user never visits the same page twice.

Uppsala, Sweden

4th International Conference on Mobile Systems, Applications, and Services (MobiSys) 2008-08-15T00:16:22-04:00 44 7 6 8 Conference 8 2009-02-09T22:06:10-05:00 2006 Automatic adaptation of content for mobile devices is a challenging problem because optimal adaptation often depends on the usage semantics of content, as well as the context of users (e.g., screen size of device being used, network connectivity, location, etc.). UsageawaRe Interactive Content Adaptation (URICA) is an automatic technique that adapts content for mobile devices based on usage semantics. URICA allows a user who is unsatisfied with the system’s current adaptation prediction to take control of the adaptation process and make changes until the content is suitably adapted for her purposes. The adaptation system learns from the user’s modifications and adjusts its prediction for future accesses by other users. This paper shows that it is possible to exploit user interaction to learn how to adapt content based on context. We introduce Feedback-driven Context Selection (FCS), an automatic technique that leverages user interaction to identify the context that has the most impact on adaptation requirements. We added contextawareness to URICA so that it makes adaptation predictions for a user based only on the history of the community of users that share the context identified by FCS. The result is an automatic adaptation system that provides fine grain adaptations that reflect both the user’s context and the content’s usage semantics. This level of fine grain adaptation was previously available only in content that was customized manually. Experiments with two context-aware URICA prototypes show that FCS correctly identifies the contextual characteristics that impact adaptation requirements, and that grouping users into communities based on context improves the performance of the adaptation system by up to 79%.

Uppsala, Sweden

4th International Conference on Mobile Systems, Applications, and Services (MobiSys) 2008-08-15T00:21:40-04:00 45 7 6 8 Conference 8 2009-02-09T22:06:44-05:00 2006

Uppsala, Sweden

Poster at 4th International Conference on Mobile Systems, Applications, and Services (MobiSys) 2008-09-19T14:00:28-04:00 79 7 9 13 Poster 13 2009-02-09T21:59:08-05:00 2006 Usage-awaRe Interactive Content Adaptation (URICA) is an automatic technique that adapts content for display on mobile devices based on usage semantics. URICA allows users who are unsatisfied with the system’s adaptation decision to take control of the adaptation process and make changes until the content is suitably adapted for their purposes. The successful adaptation is recorded and used in making future adaptation decisions. To validate URICA, we implemented a prototype system called Chameleon that performs fidelity adaptation on web images. We conducted a user study in which participants used Chameleon to browse image-rich web pages on bandwidth-limited cellular links and used the collected traces to evaluate our system. We show that Chameleon reduces the latency for browsing web content by up to 65% and reduces bandwidth consumption by up to 80%. Chameleon also allows users to exchange bandwidth consumption for user interaction based on their personal preferences.

Leuven, Belgium

1st EuroSys Conference 2008-08-15T00:23:52-04:00 46 5 6 8 Conference 8 2009-02-09T22:07:20-05:00 2006 In this paper, we argue that localization solution based on cellular phone technology, specifically GSM phones, is a sufficient and attractive option in terms of coverage and accuracy for a wide range of indoor, outdoor, and placebased location-aware applications. We present preliminary results that indicate that GSM-based localization systems have the potential to detect the places that people visit in their everyday lives, and can achieve median localization accuracies of 5 and 75 meters for indoor and outdoor environments, respectively.

Semiahmoo Resort, WA

7th IEEE Workshop on Mobile Computing Systems and Applications (WMCSA) 2008-08-15T00:26:56-04:00 47 5 7 12 Workshop 12 2009-02-09T22:15:58-05:00 2006 Iterative Adaptation is a novel approach to adaptation for resource-limited mobile and wireless environments that supports powerful application-specific adaptations without requiring modifications to the application s source code. Common productivity applications, such as browsers, word processors, and presentation tools, export APIs that allow external applications to control their operation. The novel premise in iterative adaptation is that these APIs are sufficient to support a wide range of adaptation policies for applications running on resource-limited devices. In addition to allowing adaptation without having to change the application s source code, this approach has a unique combination of advantages. First, it supports centralized management of resources across multiple applications. Second, it makes it possible to modify application behavior after the application has been deployed. This paper evaluates the extent to which existing APIs can be used for the purposes of adapting document-based applications to run on bandwidth-limited devices. In particular, we implement a large number of bandwidth adaptations for applications from the Microsoft Office and the OpenOffice productivity suites and for Internet Explorer. Although we find limitations in their APIs, we are able to implement many adaptation policies without much complexity and with good performance. Moreover, iterative adaptation achieves performance similar to an approach that implements adaptation by modifying the application, while requiring only a fraction of the coding effort.

patito

2008-08-13T16:32:49-04:00 21 IEEE Transactions on Parallel and Distributed Systems (TPDS) 11 10 4 IEEE Press 9 Journal 9 2009-02-09T22:24:05-05:00 16 2005 Recovery from intrusions is typically a very time-consuming operation in current systems. At a time when the cost of human resources dominates the cost of computing resources, we argue that next generation systems should be built with automated intrusion recovery as a primary goal. In this paper, we describe the design of Taser, a system that helps in selectively recovering legitimate file-system data after an attack or local damage occurs. Taser reverts tainted, i.e. attack-dependent, file-system operations but preserves legitimate operations. This process is difficult for two reasons. First, the set of tainted operations is not known precisely. Second, the recovery process can cause conflicts when legitimate operations depend on tainted operations. Taser provides several analysis policies that aid in determining the set of tainted operations. To handle conflicts, Taser uses automated resolution policies that isolate the tainted operations. Our evaluation shows that Taser is effective in recovering from a wide range of intrusions as well as damage caused by system management errors.

Brighton, UK

20th Symposium on Operating Systems Prinicples (SOSP) 2008-08-18T12:01:34-04:00 49 11 6 8 Conference 8 2009-02-09T22:07:46-05:00 2005 Accurate indoor localization has long been an objective of the ubiquitous computing research community, and numerous indoor localization solutions based on 802.11, Bluetooth, ultrasound and infrared technologies have been proposed. This paper presents the first accurate GSM indoor localization system that achieves median accuracy of 5 meters in large multifloor buildings. The key idea that makes accurate GSM-based indoor localization possible is the use of wide signal-strength fingerprints. In addition to the 6-strongest cells traditionally used in the GSM standard, the wide fingerprint includes readings from additional cells that are strong enough to be detected, but too weak to be used for eficient communication. Experiments conducted on three multifloor buildings show that our system achieves accuracy comparable to an 802.11-based implementation, and can accurately differentiate between floors in both wooden and steel-reinforced concrete structures.

Tokyo, Japan

7th International Conference on Ubiquitous Computing (UbiComp) 2008-08-18T12:03:58-04:00 50 10 6 8 Conference 8 2009-02-09T22:08:31-05:00 2005 As mobile devices become increasingly pervasive and commonly equipped with short-range radio capabilities, we observe that it might be possible to build a network based only on pair-wise contact of users. By using user mobility as a network transport mechanism, devices can intelligently route latency-insensitive packets using power-efficient shortrange radio. Such a network could provide communication capability where no network infrastructure exists, or extend the reach of established infrastructure. To collect user mobility data, we ran two user studies by giving instrumented PDA devices to groups of students to carry for several weeks. We evaluate our work by providing empirical data that suggests that it is possible to make intelligent routing decisions based on only pair-wise contact, without previous knowledge of the mobility model or location information.

English Lake District, UK

6th IEEE Workshop on Mobile Computing Systems and Applications (WMCSA) 2008-08-18T12:06:22-04:00 51 13 7 12 Workshop 12 2009-02-09T22:16:30-05:00 2004 Mobile devices are increasingly being used to access Web content but lack the resources for proper presentation to the user. To address this problem, content is typically adapted to be more suitable for a mobile environment. Community-Driven Adaptation (CDA) is a novel approach to automatic content adaptation for mobile devices that adapts content based on feedback from users. CDA groups users into communities based on common characteristics, and assumes that users of the same community have similar adaptation requirements. CDA learns how to adapt content by observing how members of a community alter adapted content to make it more useful to them. Experiments that consider the idealized case, where all users perform the same task, show that CDA can reduce wastage of network bandwidth by up to 90% and requires less user interaction to correct bad adaptation decisions compared with existing approaches to automatic content adaptation.

English Lake District, UK

6th IEEE Workshop on Mobile Computing Systems and Applications (WMCSA) 2008-08-18T12:09:28-04:00 52 13 7 12 Workshop 12 2009-02-09T22:17:09-05:00 2004 The interference range in multi-hop ad hoc networks (MANETs) is typically twice as large as the transmission range. This phenomena causes packets of a multi-hop flow to interference with each other as they are relayed over the multi-hop route. This interference, an instance of the notorious hidden terminal problem, is caused by simultaneous transmissions by down-stream nodes unaware of ongoing transmissions by up-stream nodes. DMAC is a novel MAC protocol that alleviates the hidden terminal problem by deferring further transmissions until the previously transmitted packets travel far enough to avoid interference with the newly transmitted packets. For simple chain topologies, DMAC improves the throughput of CBR and TCP flows by up to 100% and 60%, respectively. For random mobile topologies with up to 40 simultaneous flows, DMAC improves the throughput of TCP flows by up to 30%.

Tampa, FL

th International IEEE Workshop on Wireless Local Networks (WLN) 2008-08-18T12:11:58-04:00 53 12 7 12 Workshop 12 2009-02-09T22:17:38-05:00 2004 We evaluate the robustness of simplified mobility and radio propagation models for indoor MANET simulations. A robust simplification allows researchers to extrapolate simulation results and reach reliable conclusions about the expected performance of protocols in real life. We show that common simplified mobility and radio propagation models are not robust. Experiments with DSR and DSDV, two representative MANET routing protocols, show that the simplifications affect the two protocols in very different manners. Even for a single protocol, the effects on perceived performance can vary erratically as parameters change. These results cast doubt on the soundness of evaluations of MANET routing protocols based on simplified mobility and radio propagation models, and expose the urgent need for more research on realistic MANET simulation.

Santa Clara, CA

1st IEEE Communications Society Conference on Sensor and Ad Hoc Communications and Networks (SECON) 2008-08-18T12:13:01-04:00 54 11 6 8 Conference 8 2009-02-09T22:09:09-05:00 2004 The throughput in multi-hop ad hoc networks (MANETs) is highly dependent on the sending rate and the route length from the source node to the destination. Sending packets at the optimal rate for a given route length maximizes throughput in the network, whereas slightly increasing the sending rate over the optimal value may decrease throughput by up to 55%. This paper presents a novel cross-layer technique for flow control in lightly-loaded MANETs. The technique allows applications to send packets at the rate that maximizes throughput for a given route length. To achieve this, the routing layer notifies interested applications about routing changes, and the applications adaptively modify their sending rates based on the new route length to the destination. In static and mobile networks, this technique outperforms UDP-based flows with a fixed sending rate and doubles the throughput of TCP for networks with up to 2 concurrent flows.

Montreal, Quebec

International Workshop on Network Design and Architecture (IWNDA) 2008-08-18T12:15:13-04:00 55 9 7 12 Workshop 12 2009-02-09T22:18:28-05:00 2004 The decoupling of producers and consumers in time and space in the publish/subscribe paradigm lends itself well to the support of mobile users who roam about the environment and have intermittent network connectivity. This paper identifies the factors that affect the performance of a distributed publish/subscribe architecture supporting mobility; formalizes mobility algorithms for distributed publish/ subscribe systems and develops and evaluates optimizations that reduce the costs associated with supporting mobility in publish/subscribe systems. In our analysis, we focus on the unicast traffic generated to support mobile users, as opposed to the regular multicast traffic used for event dissemination to stationary clients. We find that the network capacity must be doubled to handle the extra load of just 10% of mobile users.

Berkeley, CA

IEEE International Conference on Mobile Data Management (MDM) 2008-08-18T12:18:10-04:00 56 2 6 8 Conference 8 2009-02-09T22:09:44-05:00 2004 This paper introduces adaptation-aware editing and progressive update propagation, two novel mechanisms that enable authoring multimedia content and collaborative work on mobile devices. Adaptation-aware editing enables editing content that was adapted to reduce download time to the mobile device. Progressive update propagation reduces the time for propagating content generated at the mobile device by transmitting either a fraction of the modifications or transcoded versions thereof. With application-aware editing and progressive update propagation, an object present at a mobile device is characterized not only by a particular version, as in conventional replication, but also by a particular fidelity. We demonstrate that replication models can be extended to account for fidelity independently of the mechanisms used for concurrency control and consistency maintenance. As a result, the two techniques described in this paper can easily be added to any replication protocol, whether optimistic or pessimistic. We report on our experience implementing adaptation-aware editing and progressive update propagation. Experiments with two multimedia applications, an email reader and a presentation software package, show that both mechanisms can be added with modest programming effort and achieve substantial reductions in upload and download latencies.

San Francisco, CA

1st International Conference on Mobile Systems, Applications, and Services (MobiSys) 2008-08-18T12:20:48-04:00 57 6 6 8 Conference 8 2009-02-09T22:10:13-05:00 2003 Applications running on a mobile and wireless devices must be able to adapt gracefully to limited and fluctuating network resources. The variety of applications, platforms upon which they run, and desires of their users, require a variety of adaptation policies to be implemented and maintained. Therefore, it becomes important for adaptation policies to be easy to develop, to debug, and to compose together to form complex policies that satisfy the needs of mobile users. This paper presents the design, implementation, and evaluation of a simple programming language for expressing scheduling policies for transmission of multiple objects across a shared network connection. A key design component of our language is the ability to express constraints among the objects to be transmitted. A policy can make ordering constraints such as "all text objects are transmitted before any image objects" or a policy might express rules on the the relative bandwidth allocations across objects of different types. Because it is possible to express contradictory constraints, our system finds suitable approximate solutions when no precise solution is available.

Callicoon, NY

4th IEEE Worwshop on Mobile Computing Systems and Applications 2008-08-18T12:23:24-04:00 58 7 7 12 Workshop 12 2009-02-09T22:19:07-05:00 2002 Component-based adaptation is a novel approach for adapting applications to the limited availability of resources such as bandwidth and power in mobile environments. Component-based adaptation works by calling on the run-time APIs that modern component-based applications export. Because source code modification is not necessary, even proprietary applications such as productivity tools from Microsoft's Office suite can be adapted. Moreover, new adaptive behavior can be added to applications long after they have been deployed. Even if source code is available, development time for implementing adaptation is much reduced. In addition, the ease with which adaptations can be implemented in this framework has enabled me to explore new avenues in adaptation. First, I have developed the first adaptive system to support document editing and collaboration over bandwidth-limited links. The key insight gathered from this work is that support for adaptation is orthogonal to concurrency and consistency mechanisms, and therefore can be integrated easily in existing systems. Second, I have developed a hierarchical adaptive transmission scheduler to support coordinated multi-application adaptation. I have demonstrated the effectiveness of component-based adaptation by implementing a system called Puppeteer, which has allowed me to adapt widely deployed applications, such as productivity tools from Microsoft's Office suite and Sun Microsystems' OpenOffice suite. Although the APIs of these applications impose some limitations, I have been able to implement a wide range of adaptation policies for reading, editing, and collaboration, with modest implementation effort and good performance results.

Houston, Texas

PhD. Thesis 2008-08-14T23:40:07-04:00 39 Department of Electrical and Computer Engineering, Rice University 5 11 16 Thesis 16 2009-02-09T21:44:20-05:00 2002 Applications running on a mobile and wireless device most be able to adapt gracefully to limited and fluctuating network resources. When multiple applications run on the same device, control over bandwidth scheduling becomes fundamental in order to coordinate the adaptation of the multiple applications. This paper presents the design and evaluation of our Hierarchical Adaptive Transmission Scheduler (HATS), which adds control over bandwidth scheduling to the Puppeteer adaptation system. HATS enables the implementation of system-wide adaptation policies that can drastically improve the user's perception of network performance.

San Jose, CA

Multimedia Computing and Networking Conference (MMCN) 2008-08-18T12:25:48-04:00 59 2 6 8 Conference 8 2009-02-09T22:11:02-05:00 2002 In this paper, we demonstrate how component-based middleware can reduce the energy usage of closed-source applications. We first describe how the Puppeteer system exploits well-defined interfaces exported by applications to modify their behavior. We then present a detailed study of the energy usage of Microsoft's PowerPoint application and show that adaptive policies can reduce energy expenditure by 49% in some instances. In addition, we use the results of the study to provide general advice to developers of applications and middleware that will enable them to create more energy-efficient software.

Heidelberg, Germany

IFIP/ACM International Conference on Distributed Systems Platforms (Middleware) 2008-08-18T12:28:04-04:00 60 12 6 8 Conference 8 2009-02-09T22:11:36-05:00 2001 This paper presents the design of CoFi, a novel architecture for supporting document editing and collaborative work over bandwidth-limited clients. CoFi combines the previously disjoint notions of consistency and fidelity in a unified architecture. CoFi enables bandwidth-limited clients to edit documents that are only partially present at the client (because parts of the documents were lossily transcoded, or only a portion of the document was fetched), and to propagate modifications incrementally by progressively increasing their fidelity.

Atlanta, GA

Workshop on Application Models and Programming Tools for Ubiquitous Computing (UbiTools) 2008-08-18T12:30:18-04:00 61 10 7 12 Workshop 12 2009-02-09T22:19:43-05:00 2001

Schloss Elmau, Germany

8th IEEE Workshop on Hot Topics in Operating Systems (HotOS) 2008-08-18T12:32:25-04:00 62 6 7 12 Workshop 12 2009-02-09T22:20:58-05:00 2001 Puppeteer is a system for adapting component-based applications in mobile environments. Puppeteer takes advantage of the exported interfaces of these applications and the structured nature of the documents they manipulate to perform adaptation without modifying the applications. The system is structured in a modular fashion, allowing easy addition of new applications and adaptation policies. Our initial prototype focuses on adaptation to limited bandwidth. It runs on Windows NT, and includes support for a variety of adaptation policies for Microsoft PowerPoint and Internet Explorer 5. We demonstrate that Puppeteer can support complex policies without any modification to the application and with little overhead. To the best of our knowledge, previous implementations of adaptations of this nature have relied on modifying the application.

San Francisco, CA

3rd USENIX Symposium on Internet Technologies and Systems (USITS) 2008-08-18T12:34:19-04:00 63 4 6 8 Conference 8 2009-02-09T22:12:25-05:00 2001 Microsoft Office, the most popular office productivity suite, produces large documents that can result in long download latencies for platforms with limited bandwidth. To reduce latency and improve the user's experience, these documents need to be adapted for transmission on a limited-bandwidth network. To identify opportunities for adaptation, we characterize documents created by three popular applications from the Microsoft Office suite: Word, PowerPoint, and Excel. Our study encompasses over 12,500 documents retrieved from 935 different Web sites. Our main conclusions are: 1) Microsoft Office documents are large and require adaptation on bandwidth-limited clients; 2) embedded objects and images account for the majority of the data in these documents, with image types being the most popular non-text content, suggesting that adaptation efforts should focus on these elements; 3) compression considerably reduces the size of these documents; and 4) the internal structure of these documents (pages, slides, or sheets) can be used to download elements on demand and reduce user-perceived latency.

Seattle, Washington

4th USENIX Windows Systems Symposium 2008-08-18T12:36:04-04:00 64 9 6 8 Conference 8 2009-02-09T22:13:00-05:00 2000 In this paper, we examine the causes and effects of contention for shared data access in parallel programs running on a software distributed shared memory (DSM) system. Specifically, we experiment on two widely-used, page-based protocols, Princeton's home-based lazy release consistency (HLRC) and TreadMarks. For most of our programs, these protocols were equally affected by latency increases caused by contention and achieved similar performance. Where they differ significantly, HLRC's ability to manually eliminate load imbalance was the largest factor accounting for the difference. To quantify the effects of contention we either modified the application to eliminate the cause of the contention or modified the underlying protocol to efficiently handle it. Overall, we find that contention has profound effects on performance: eliminating contention reduced execution time by 64% in the most extreme case, even at the relatively modest scale of 32 nodes that we consider in this paper.

Rochester, NY

5th Workshop on Languages, Compilers, and Run-time Systems for Scalable Computers (LRC) 2008-08-18T12:39:22-04:00 65 6 7 12 Workshop 12 2009-02-09T22:20:22-05:00 2000 Recent developments in office productivity suites make it easier for users to publish rich compound documents on the Web. Compound documents appear as a single unit of information but may contain data generated by different applications, such as text, images, and spreadsheets. Given the popularity enjoyed by these office suites and the pervasiveness of the Web as a publication medium, we expect that in the near future these compound documents will become an increasing proportion of the Web’s content. As a result, the content handled by servers, proxies, and browsers may change considerably from what is currently observed. Furthermore, these compound documents are currently treated as opaque byte streams, but future Web infrastructure may wish to understand their internal structure to provide higher-quality service. In order to guide the design of this future Web infrastructure, we characterize compound documents currently found on theWeb. Previous studies ofWeb content either ignored these document types altogether or did not consider their internal structure. We study compound documents originated by the three most popular applications from the Microsoft Office suite: Word, Excel, and PowerPoint. Our study encompasses over 12,500 documents retrieved from 935 differentWeb sites. Our main conclusions are: 1. Compound documents are in general much larger than current HTML documents. 2. For large documents, embedded objects and images make up a large part of the documents’ size. 3. For small documents, XML format produces much larger documents than OLE. For large documents, there is little difference. 4. Compression considerably reduces the size of documents in both formats.

2008-08-18T12:42:24-04:00 66 Rice Computer Science 12 TR99-351 10 10 Technical Report 10 2009-02-09T21:50:08-05:00 1999 Caching in the Placeless Documents system poses new challenges because users can attach active properties to documents. Active properties can modify the document's content as seen by the user. Thus, the caching mechanisms must take into account that a document's content not only depends on when the document was last modified, but also on the set of personal and universal properties attached to the document and the information on which these properties depend. Interestingly, active properties can be used to help caches manage their content by notifying them of events that affect cache consistency, by providing caches with document-specific verifiers to further check check on a document's consistency, and by returning information that can aid in decisions of which documents to cache.

Rio Rico, AZ

7th IEEE Workshop on Hot Topics in Operating Systems (HotOS) 2008-08-18T12:51:06-04:00 68 4 7 12 Workshop 12 2009-02-09T22:02:29-05:00 1999 We demonstrate the profound effects of contention on the performance of page-based software distributed shared memory systems, as such systems are scaled to a larger number of nodes. Programs whose performance scales well experience only minor increases in memory latency, do not suffer from contention, and show a balanced communication load. In contrast, programs that scaled poorly suffered from large memory latency increases due to contention and communication imbalance. We use two existing protocols, Princeton's home-based protocol and the TreadMarks protocol, and a third novel protocol, Adaptive Striping. For most of our programs, all three protocols were equally affected by latency increases and achieved similar performance. Where they differ significantly, the communication load imbalance, which is caused by the read accesses to pages that have multiple readers following one or more writers, is the largest factor accounting for the difference.

Houston, Texas

M.Sc. Thesis 2008-08-14T23:52:21-04:00 40 Department of Electrical and Computer Engineering, Rice University 2 11 16 Thesis 16 2009-02-09T21:42:04-05:00 1999 In this paper, we compare the performance of two multiple-writer protocols based on lazy release consistency. In particular, we analyze the performance of Princeton's home-based protocol and TreadMarks' protocol on a 32-processor platform. We found the performance of the two protocols to be strikingly similar for most of our application suite. The homeless protocol performed better for two of the applications while the home-based protocol outperformed TreadMarks for one application. We show, however, that for the latter, the performance gap is not a consequence of the disparity in message count. It is, rather, a product of differing abilities of the protocols to balance communication among nodes. These results differ from a previous study conducted on the two protocols. We attribute the difference to: (1) a platform with a different ratio of memory bandwidth to network bandwidth and (2) lazy diffing and request overlapping, two features of the original TreadMarks implementation that were not considered in the previous study.

Orlando, FL

5th International Symposium on High-Performance Computer Architecture 2008-08-18T12:47:49-04:00 67 2 6 8 Conference 8 2009-02-09T22:13:52-05:00 1999