March 2, 2010

TechFest 2010

Location: Redmond, WA, U.S.

Client + Cloud Computing for Research

Scientific applications have diverse data and computation needs that scale from desktop to supercomputers. Besides the nature of the application and the domain, the resource needs for the applications also vary over time—as the collaboration and the data collections expand, or when seasonal campaigns are undertaken. Cloud computing offers a scalable, economic, on-demand model well-matched to evolving eScience needs. We will present a suite of science applications that leverage the capabilities of Microsoft’s Azure cloud-computing platform. We will show tools and patterns we have developed to use the cloud effectively for solving problems in genomics, environmental science, and oceanography, covering both data and compute-intensive applications.

Learn more about this project

Cloud Faster

To make cloud computing work, we must make applications run substantially faster, both over the Internet and within data centers. Our measurements of real applications show that today’s protocols fall short, leading to slow page-load times across the Internet and congestion collapses inside the data center. We have developed a new suite of architectures and protocols that boost performance and the robustness of communications to overcome these problems. The results are backed by real measurements and a new theory describing protocol dynamics that enables us to remedy fundamental problems in the Transmission Control Protocol. We will demo the experience users will have with Bing Web sites, both with and without our improvements. The difference is stunning. We also will show visualizations of intra-data-center communication problems and our changes that fix them. This work stems from collaborations with Bing and Windows Core Operating System Networking.

Learn more about this project

Energy-Aware VMs and Cloud Computing

Virtual machines (VMs) become key platform components for data centers and Microsoft products such as Win8, System Center, and Azure. But existing power-management schemes designed at the server level, such as power capping and CPU throttling, do not work with VMs. VMmeter can estimate per-VM power consumption from Hyper-V performance counters, with the assistance of WinServer2008 R2 machine-level power metering, thus enabling power management at VM granularity. For example, we can selectively throttle VMs with the least performance hit for power capping. This demo compares VMmeter-based with hardware-based power-management solutions. We run multiple VMs, one of them being a high-priority video playback on a server. When a user requests power capping with our solution, the video playback will maintain high performance, while with hardware-capping solutions, we see reduced performance. We also will show how VMmeter can be part of System Center management packs.

Learn more about this project

The Future of Looking Back

We will examine the issue of family archiving and present a system designed to enable families to capture, manage, create, and store new kinds of digital memorabilia. The system, using Surface as its hub, shows how families can upload photos and videos quickly and easily, and also scan in physical memorabilia, such as children’s artwork or a child’s first pair of shoes. The system enables families to view these media in many flexible ways and to create new, compelling kinds of digital objects, such as multimedia scrapbooks and even a digital piñata. We further will show how this system would fit into a larger ecosystem of devices in the home and link to new kinds of media displays.

Learn more about this project

Greening Corporate Networks with Sleep Proxy

In a corporate network, most desktop machines always are left on, even when they are not in use for extended periods, such as at night. This is wasteful, bad for the environment, and bad for the corporate treasury. While Win7 provides aggressive sleep functionality, most users override it because they occasionally might want to access their machine remotely. Ideally, a desktop would go to sleep when not in use and awaken seamlessly when the user tries to access it. We have built a system to enable this. Our system consists of a “sleep server” that maintains the network presence of the sleeping machine and seamlessly awakens it on remote access. We do not require special hardware or changes to existing software. Our system is operational in Building 99 and has resulted in substantial savings in terms of money, power consumption, and carbon-dioxide emissions.

Learn more about this project

Inside the Cloud: New Cloud-Computing Interaction

With cloud computing, users can access their personal data anywhere and anytime. Cloud computing also will enable new forms of data to be provided for users, with applications ranging from Web data mining to social networks. But cloud computing necessitates new interaction metaphors and input-output technology. The cloud mouse is one such technology. Every user will have one. It will be a secure key to every user’s cloud data. And, with six degrees of freedom and with tactile feedback, the cloud mouse will enable users to orchestrate, interact with, and engage with their data as if they were inside the cloud.

Mobile Assistance Using Infrastructure (MAUI)

Mobile devices have reached an impasse. Although the resources that can be integrated onto mobile handheld devices will continue to improve, faster CPUs, more RAM, faster wireless NICs, making substantial use of these resources will require a major breakthrough in battery technology. To bypass these limitations, MAUI (Mobile Assistance Using Infrastructure) is a system that enables fine-grained offloading of mobile applications to cloud-based infrastructure. By leveraging nearby infrastructure, MAUI enables a new class of resource-intensive applications, such as augmented reality, to run on mobile handheld devices. With MAUI, we enable resource-intensive .NET applications to run on Windows Mobile smartphones. We will demonstrate: A resource-intensive face-recognition application that consumes an order-of-magnitude less energy. A voice-based translation application that previously could not run using only the limited resources available on today’s smartphones.

Learn more about this project

Mobile-Search and Advertisement-Cache Architecture

We will show how to improve the mobile-search user experience by caching popular search results on mobile devices. First, a community-based cache is created by mining the most popular queries in mobile-search logs. Over time, the cache is personalized by adding all the new user search queries. An analysis of four months of mobile-search logs shows that, on average, 66 percent of the search queries submitted by a user can be answered by caching 2,500 links on a 1MB cache. Our prototype implementation in Windows Mobile demonstrates responses 16 times faster and 23 times more energy-efficient compared with querying through a 3G link. Our prototype also demonstrates how our caching architecture can enable monetization of mobile local search without hurting the mobile user experience. A rich set of ads is first cached on the phone. Because all the ads are locally stored, finding and displaying a mobile local ad is extremely fast, enabling us to display ads instantly to a mobile user as a query is being typed.

Learn more about this project

Mobile Surface

Mobile Surface is a novel interaction system for mobile computing. Our goal is to bring the Microsoft Surface experience to mobile scenarios and, more importantly, to enable 3-D interaction with mobile devices. We will demonstrate how to transform any surface, such as a coffee table or a piece of paper, into a Mobile Surface by using a mobile device and a camera-projector system. Besides the Surface, we will show 3-D object imaging, augmented reality, and multiple-layer 3-D information visualization. In particular, we have developed a system with the camera-projector component to scan 3-D objects in real time while doing normal projection. To visualize, 3-D data can be projected onto a surface formed by a piece of paper while maintaining the original scale as if it were printed on that paper, and a user can interact with the projected content with a hand. Mobile Surface enables you to interact with digital contents and information around you from anywhere.

Learn more about this project

Mobile-to-Mobile Networking in 3G Networks

Mobile devices increasingly are first-class computing devices, generating large amounts of data. Searching and sharing data securely across multiple devices can be a significant challenge. We have built Contrail, a communication abstraction for P2P communication on mobile phones. Communication in Contrail is purely asynchronous, coping with the fact that phones temporarily are disconnected from the network. Phones set up filters with other phones expressing their interest set. We will demonstrate the usage of Contrail with three applications: P2P content distribution, P2P search, and location-based group communication.

Learn more about this project

Natural User Interfaces with Physiological Sensing

Microsoft has innovated continually in developing novel interaction modalities, or natural user interfaces. Surface and Project Natal are two examples. While these modalities rely on sensors and devices situated in the environment, we believe there is a need for new modalities that enhance the mobile experience. We take advantage of sensing technologies that enable us to decode the signals generated by the body. We will demo muscle-computer interfaces, electromyography-based armbands that sense muscular activation directly to infer finger gestures on surfaces and in free space, and bio-acoustic interfaces, mechanical sensors on the body that enable us to turn the entire body into a tap-based input device.

Learn more about the Skinput project

Learn more about the Muscle-Computer Interfaces project

New Technologies for Multi-Image Fusion

As video and still cameras have become almost ubiquitous, people are taking increasingly more photographs and videos of the world around them. Often, the photographer’s intent is to capture more than what can be seen in a single photograph, and he or she instead takes a large set of images or a video clip to capture a large scene or a moment that extends over time. One can combine these images to produce an output that improves the input images, such as creating an image with a large field of view, a panorama, or a composite image that takes the best parts of the image, a photo montage. But creating these results is still non-trivial for many users. One challenge is in creating large-scale panoramas, for which the capture and stitching times can be long. In addition, when using consumer-level point-and-shoot cameras and camera phones, artifacts such as motion blur appear. Another challenge is combining large image sets from photos or videos to produce results that use the best parts of the images to create an enhanced photograph. We will present several new technologies that advance the state of the art in these areas and create improved user experiences. For panorama generation, we will demonstrate: ICE 2.0. Stitching of panoramas from video. Generating sharp panoramas from blurry videos. For generating composites, we will demonstrate: Video to snapshots. De-noising and sharpening using lucky imaging.

Learn more about this project

OneAlbum-Find Your Photos in Your Friends’ Albums

Today, my album refers to a collection of photos I have taken. But many photos relevant to me—such as photos of me or my children—are in my friends’ albums. OneAlbum automatically finds relevant photos in my friends’ albums on social networks or in shared albums, brings them to my album, and shows them side-by-side with the photos I’ve taken. For example, if I was at a party, I’ll see all photos from that party—those I’ve taken and those my friends took. The technology behind OneAlbum is a novel, unsupervised face-recognition algorithm. It analyzes the photos in my album to find automatically the faces of people I most care about, based on frequency of their appearance; no tagging is required. Then, using the social-network graph and other information, OneAlbum crawls my friend’s albums looking for photos of people that interest me. The algorithm was tested on real large-scale albums including tens of thousands of photos and achieved accuracy rates as high as 90 percent.

Project Gustav: Immersive Digital Painting

Project posterGustav is a realistic painting-system prototype that enables artists to become immersed in the digital painting experience. The natural interface makes Gustav ideal for hobbyists and professional artists alike. Gustav achieves a high level of interactivity and realism by leveraging the computing power of modern GPUs, taking full advantage of multitouch and tablet input technology and our novel, natural media-modeling and brush-simulation algorithms. Our prototype provides convincingly realistic models for pastel and oil media, with more to come.

Learn more about this project

Saving Desktop Energy: WakeOnLAN & Virtualization

Saving desktop energy has been an area of focus at Microsoft Research India over the past year. WARP and LiteGreen are complementary projects under this theme. WARP is a composition of platform components that facilitates remote power-state management of a PC using the Wake-on-LAN and other system-management mechanisms. This system is virtually stateless and provides user interfaces for system management. Features include remote peer-to-peer wakeups and remote sleep, hibernate, or shutdown of a desktop and asynchronous machine-state transitions based on events published from user interfaces such as the Web, e-mail, SMS, and location-based services. Automatic desktop upgrades, centralized system control, delegation of power management, and auditing are also components of this system. LiteGreen is a system for saving energy from idle PCs in enterprises by exploiting short idle periods as well as long ones. To avoid user disruption, LiteGreen virtualizes the desktop computing environment and migrates it between the physical PC and a virtual-machine server, depending on whether the desktop computing environment is being used. Based on usage analysis of 120 desktops at Microsoft Research India, LiteGreen was able to deliver energy savings of 72 to 74 percent. When a user steps away from a PC, the desktop is migrated to a server and the PC is put to sleep. When the user returns, he or she is able to start using the desktop immediately.

Learn more about this project

Technologies for Rural Financial-Service Delivery

We will show two solutions that enable financial-service delivery to low-income individuals in developing countries. Both integrate the use of portable digital devices with paper-based tools that cost little and are widely used in developing countries. The first solution focuses on improving microfinance-record management through the use of pen-and-paper-based input on a low-cost digital slate device. Handwritten data on paper is digitized and processed simultaneously to provide instant user feedback, delivering gains in data quality and process efficiency. The second solution addresses a security concern in mobile-phone-based banking transactions. We use paper to facilitate secure PIN entry on mobile phones and to achieve a suitable tradeoff between security and usability in phone-assisted banking.

The Translating! Telephone

We will demonstrate a system for live speech-to-text and speech-to-speech translation of telephone calls. Douglas Adams’ Babelfish inspired dreams of unfettered universal communication. Though we are still far from achieving that goal, there are scenarios in which today’s limited accuracy can create value. Our goal in the telephone-call scenario is to provide an aid for cross-language communication in the event that no other means of communication exists. The system we will show makes extensive use of speaker-adaptation technologies to achieve reasonable, real-time speech-to-text transcription accuracy. This is then translated live using machine translation to provide speech-to-text translation and further fed into a text-to-speech system to realize speech-to-speech translation. The speech-to-text transcript and the translated transcript are shown to the users to enable validation of their intentions. This system will be demonstrated by a live conversation between German and English speakers.

Learn more about this project