Data centers are increasingly being re-designed for workload consolidation in order to reap the benefits of better resource utilization, power savings, and physical space savings. Among the forces driving savings are server and storage virtualization technologies. As more consolidated workloads are concentrated on physical machines -- e.g., the virtual density is already very high in virtual desktop environments, and will be driven to unprecedented levels with the fast growing highcore counts of physical servers -- the shared storage layer must respond with virtualization innovations of its own such as de-duplication and thin provisioning. A key insight of this paper is that there is a greater synergy between the two layers of storage and server virtualization to exploit block sharing information than was previously thought possible. We reveal this via developing a systematic framework to explore the storage and virtualization servers interactions. We also quantitatively evaluate the I/O bandwidth and latency reduction that is possible between virtual machine hosts and storage servers using real-world trace driven simulation. Moreover, we present a proof of concept NFS implementation that incorporates our techniques to quantify their I/O latency benefits.

Numerous sensors in modern mobile phones enable a range of people-centric applications. This paper envisions a system called PhonePoint Pen that uses the in-built accelerometer in mobile phones to recognize human writing. By holding the phone like a pen, a user should be able to write short messages or draw simple diagrams in the air. The acceleration due to hand gestures can be translated into geometric strokes, and recognized as characters. We prototype the PhonePoint Pen on the Nokia N95 platform, and evaluate it through real users. Results show that English characters can be identified with an average accuracy of 91.9%, if the users conform to a few reasonable constraints. Future work is focused on refining the prototype, with the goal of offering a new user-experience that complements keyboards and touch-screens.

The costs of data loss and unavailability can be large, so businesses use many data protection techniques such as remote mirroring, snapshots, and backups to guard against failures. Choosing an appropriate combination of techniques is difficult because there are numerous approaches for protecting data and allocating resources. Storage system architects typically use ad hoc techniques, often resulting in overengineered expensive solutions or underprovisioned inadequate ones. In contrast, this paper presents a principled automated approach for designing dependable storage solutions for multiple applications in shared environments. Our contributions include search heuristics for intelligent exploration of the large design space and modeling techniques for capturing interactions between applications during recovery. Using realistic storage system requirements, we show that our design tool produces designs that cost up to two times less in initial outlays and expected data penalties than the designs produced by an emulated human design process. Additionally, we compare our design tool to a random search heuristic and a genetic algorithm metaheuristic, and show that our approach consistently produces better designs for the cases we have studied. Finally, we study the sensitivity of our design tool to several input parameters.

Mobius 2.3 is an extensible dependability, security, and performance modeling environment for large-scale discrete-event systems. It provides multiple model formalisms and solution techniques, facilitating the representation of each part of a system in the formalism that is most appropriate for it, and the application of the solution method or methods best-suited to estimating the system's behavior. Since its initial release in 2001, many advances have been made in Moumlbius's design and implementation that have strengthened its place in the modeling and analysis community. With almost a decade of widespread academic and industrial use, Moumlbius has proven itself to be useful in a wide variety of modeling situations. This paper documents the current feature set of Mobius 2.3, emphasizing recent significant enhancements.

A growing number of mobile phone applications utilize physical location to express the context of information. Most of these location-based applications assume GPS capabilities. Unfortunately, GPS incurs an unacceptable energy cost that can reduce the phone's battery life to less than nine hours. Alternate localization technologies, based on WiFi or GSM, improve battery life at the expense of localization accuracy. This paper quantifies this important tradeoff that underlies a range of emerging services. Driven by measurements from Nokia N95 phones, we develop an energy-efficient localization framework called EnLoc. The framework characterizes the optimal localization accuracy for a given energy budget, and develops prediction- based heuristics for real-time use. Evaluation on traces from real users demonstrates the possibility of achieving good localization accuracy for a realistic energy budget.

Petascale computing requires I/O subsystems that can keep up with the dramatic computing power demanded by such systems. TOP500.org ranks top computers based on their peak compute performance, but there has not been adequate investigation of the current state-of-the-art and future requirements of storage area networks that support petascale computers. Dependable scaling of an I/O subsystem to support petascale computing is not as simple as adding more storage servers. In this paper, we present a stochastic activity network model that uses failure rates computed from real logs to predict the reliability and availability of the storage architecture of the Abe cluster at the National Center for Supercomputing Applications (NCSA). We then use the model to evaluate the challenges encountered as one scales the number of storage servers to support petascale computing. The results present new insights regarding the dependability challenges that will be encountered when building next-generation petabyte storage. Furthermore, we provide insight into a new design approach that will enable system designers to integrate the trace-based analysis of parameter values from real system data into their stochastic models to allow informed design choices.

Recent years have witnessed the impacts of distributed content sharing (Wikipedia, Blogger), social networks (Facebook, MySpace), sensor networks, and pervasive computing. We believe that significant more impact is latent in the convergence of these ideas on the mobile phone platform. Phones can be envisioned as people-centric sensors capable of aggregating participatory as well as sensory inputs from local surroundings. The inputs can be visualized in different dimensions, such as space and time. When plugged into the Internet, the collaborative inputs from phones may enable a high resolution view of the world. This paper presents the architecture and implementation of one such system, called Micro-Blog. New kinds of application-driven challenges are identified and addressed in the context of this system. Implemented on Nokia N95 mobile phones, Micro-Blog was distributed to volunteers for real life use. Promising feedback suggests that Micro-Blog can be a deployable tool for sharing, browsing, and querying global information.

A large majority of the Internet traffic relies on TCP as its transport protocol. In future, as the edge of the Internet continues to extend over the wireless medium, TCP (or its close variants) may not prove to be appropriate. The key reason is in TCP’s inability to discriminate congestion losses from transmission losses. Since transmission losses are frequent in wireless networks, the penalty from loss misclassification can become high, leading to performance degradation. This paper presents an eXtended Rate-based Transport Protocol (XRTP), designed to support communication over lossy wireless media. We depart from the ack-based rate control paradigm. Instead, we try to estimate the network conditions by injecting probe packets at the sender, and then observing the spacing between packets that arrive at the receiver. We show that these observations can be useful indicators of available bandwidth, network congestion, and even the cause of packet loss. The inferences from the observations are utilized to regulate the transmission rate at the sender, leading to desirable properties of congestion control and loss discrimination. Simulation results show the efficacy of our proposed rate-based protocol in lossy wireless environments.

The synergy of phone sensors (microphone, camera, GPS, etc.), wireless capability, and ever-increasing device density can lead to novel people-centric applications. Unlike traditional sensor networks, the next generation networks may be participatory, interactive, and in the scale of human users. Millions of global data points can be organized on a visual platform, queried, and sophistically answered through human participation. Recent years have witnessed the isolated impacts of distributed knowledge sharing (Wikipedia), social networks, sensor networks, and mobile communication. We believe that significant more impact is latent in their convergence, that can to be drawn out through innovations in applications. This demonstration, called Micro-Blog, is a first step towards this goal.

The costs of data loss and unavailability can be large, so businesses use many data protection techniques, such as remote mirroring, snapshots and backups, to guard against failures. Choosing an appropriate combination of techniques is difficult because there are numerous approaches for protecting data and allocating resources. Storage system designers typically use ad hoc techniques, often resulting in over-engineered, expensive solutions or under-provisioned, inadequate ones. In contrast, this paper presents a principled, automated approach for designing dependable storage solutions for multiple applications in shared environments. Our contributions include search heuristics for intelligently exploring the large design space and modeling techniques for capturing interactions between applications during recovery. Using realistic storage system requirements, we show that our design tool can produce designs that cost up to 3X less in initial outlays and expected data penalties than the designs produced by an emulated human design process

Simulation to obtain reliability and availability estimates has been widely used by system designers to evaluate and compare alternative choices before making design decisions. However, traditionally that approach worked only if significant computer resources were available or designers accepted a significant time delay between design iterations. In this paper, we present an alternative approach to compute measures of interest for a family of models that represent alternative design choices that is significantly more efficient than the traditional approach. The new approach combines the existing single-clock multiple-system simulation with adaptive uniformization. We achieve the speedup by simulating all the alternative configurations of the discrete-event model simultaneously while amortizing the cost of enabled event set management. That allows us to explore and evaluate multiple configuration settings of a discrete-event model at the same time, significantly increasing the number of alternative versions of the model that are explored in a given amount of time.