{"id":212082,"date":"2020-02-23T16:44:03","date_gmt":"2015-12-06T22:44:13","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/project\/edge-computing\/"},"modified":"2020-11-12T19:40:46","modified_gmt":"2020-11-13T03:40:46","slug":"edge-computing","status":"publish","type":"msr-project","link":"https:\/\/www.microsoft.com\/en-us\/research\/project\/edge-computing\/","title":{"rendered":"Edge Computing"},"content":{"rendered":"<div>\n<p>Edge computing is where compute resources, ranging from credit-card-size computers to micro data centers, are placed closer to information-generation sources, to reduce network latency and bandwidth usage generally associated with cloud computing. Edge computing ensures continuation of service and operation despite intermittent cloud connections. Industries ranging from manufacturing to healthcare are eager to develop real-time control systems that use machine learning and artificial intelligence to improve efficiencies and reduce cost. We are exploring this new computing paradigm by identifying and addressing emerging technology and business model challenges.<\/p>\n<h2>A Brief History of Edge Computing<\/h2>\n<p>On October 29, 2008, we invited colleagues from academia and industry for a day-long brainstorming session about the future of cloud computing. Edge computing was conceived during those discussions. Attendees included Victor Bahl (organizer, Microsoft Research), Ram\u00f3n C\u00e1ceres (AT&T Labs), Nigel Davies (Lancaster University, U.K.), Mahadev Satyanarayanan (Carnegie Mellon University), and Roy Want (Intel Research). Following this meeting, we published the first paper on this topic, in IEEE Pervasive Computing (November 1, 2009) titled: <b><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/the-case-for-vm-based-cloudlets-in-mobile-computing\/\" target=\"_blank\" rel=\"noopener noreferrer\">The Case for VM-based Cloudlets in Mobile Computing<\/a><\/b>.<\/p>\n<p>The blog <b><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/lewisshepherd.wordpress.com\/2009\/12\/13\/why-a-cloudlet-beats-the-cloud-for-mobile-apps\/\" target=\"_blank\" rel=\"noopener noreferrer\">Why a Cloudlet Beats the Cloud for Mobile Apps<\/a><\/b> (December 13, 2009) was the first article to cover our ideas. In it are described two projects, Cloudlets, a joint project of Microsoft and Carnegie Mellon University; and MAUI (Mobile Assistance Using Infrastructure), a Microsoft Research project. In Cloudlets, we investigated fast virtual machine (VM) synthesis on the edge; in MAUI, we explored a .NET programming model for computational offloads to the edge. Many of the ideas we explored have withstood the test of time. For example, guarding against network disconnections, incorporating computing versus communications tradeoff, deciding which methods to offload and which to process locally. The papers describing Cloudlet and MAUI have been cited over 5,700 times. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/10-years-is-an-eternity-in-the-tech-world-but-we-are-just-getting-started\/\" target=\"_blank\" rel=\"noopener noreferrer\">Click here<\/a> for a colorful description of a ten year look-back<\/p>\n<p>Since then, having made the case for edge computing in the research community (see <b><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/devices-and-networking-summit-plenary-keynote-3-victor-bahl\/\" target=\"_blank\" rel=\"noopener noreferrer\">Faculty Summit keynote<\/a><\/b>), industry (see: <b><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/www.networkworld.com\/article\/2979570\/cloud-computing\/microsoft-researcher-why-micro-datacenters-really-matter-to-mobiles-future.html\" target=\"_blank\" rel=\"noopener noreferrer\">Network World interview<\/a><\/b>) and internally in Microsoft (see <b><a href=\"https:\/\/www.microsoft.com\/en-us\/internet-of-things\/intelligentedge\" target=\"_blank\" rel=\"noopener noreferrer\">Intelligent Edge<\/a><\/b>), we have been focusing on live-video analytics as the \u201ckiller\u201d app for edge computing. You can read all about it in a separate project page.<\/p>\n<hr>\n<p>Listen to Victor Bahl&#8217;s Podcast,\u00a0 <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/a-brief-history-of-networking-and-a-bit-about-the-future-too-with-dr-victor-bahl\/\"><em>A brief history of networking (and a bit about the future too)<\/em><\/a>,<br \/>where he shares some fascinating stories and gives an inside look at Edge Computing.<\/p>\n<hr>\n<h2>The Intelligent Edge<\/h2>\n<p>Microsoft product groups coined the term The Intelligent Edge. The Intelligent Edge is a capability that enables Microsoft customers to enjoy a seamless experience and compute capabilities wherever their data exists\u2014in the cloud or offline. Microsoft is making it easier for developers to build apps that use edge technology, by open sourcing the Azure IoT Edge Runtime, which allows customers to modify the runtime and customize applications.<\/p>\n<figure><img loading=\"lazy\" decoding=\"async\" width=\"975\" height=\"309\" class=\"alignnone wp-image-510995\" style=\"width: 80%\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2015\/12\/Types-of-Edge.jpg\" alt=\"Types of Edges\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2015\/12\/Types-of-Edge.jpg 975w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2015\/12\/Types-of-Edge-300x95.jpg 300w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2015\/12\/Types-of-Edge-768x243.jpg 768w\" sizes=\"(max-width: 975px) 100vw, 975px\" \/><\/figure>\n<p>Learn more about the <a href=\"https:\/\/www.microsoft.com\/en-us\/internet-of-things\/intelligentedge\" target=\"_blank\" rel=\"noopener noreferrer\">Intelligent Edge.<\/a><\/p>\n<h2>Recent Activity<\/h2>\n<ul>\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/youtu.be\/zWJtNM4T-Q0\" target=\"_blank\" rel=\"noopener noreferrer\">Keynote<\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/saiconference.com\/FICC2020\/Speakers\" target=\"_blank\" rel=\"noopener noreferrer\">Future of Information & Communication Conference (FICC) 2020 in San Francisco, CA (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, Mar. 5, 2020)<\/li>\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"http:\/\/www.ieee-icii.org\/key.html\" target=\"_blank\" rel=\"noopener noreferrer\">Keynote<\/a>, IEEE International Conference on Industrial Internet 2019 in Orlando, FL (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, Nov. 11, 2019)<\/li>\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/cra.org\/ccc\/wp-content\/uploads\/sites\/2\/2020\/05\/2019-Bahl-CCC-CRA-Video-Analytics.pdf\" target=\"_blank\" rel=\"noopener noreferrer\">Invited Talk<\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/cra.org\/ccc\/events\/wide-area-data-analytics\/\" target=\"_blank\" rel=\"noopener noreferrer\">CRA\/CCC Visioning Activity: Wide Area Data Analytics<\/a>, Washington DC, (Victor Bahl, Oct. 3, 2019)<\/li>\n<li>Keynote, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aiotworkshop.github.io\/2019\/keynote.html\" target=\"_blank\" rel=\"noopener noreferrer\">The 1<sup>st<\/sup> International Workshop on Artificial Intelligence of Things<\/a>, anchorage, Alaska, (Victor Bahl, Aug. 5, 2019)<\/li>\n<li><a>Keynote<\/a>, 20<sup>th<\/sup> IEEE Symposium on World of Wireless, Mobile & Multimedia Networks in Washington D.C.(<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, June 12, 2019)<\/li>\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"http:\/\/cpslab.cs.mcgill.ca\/cpsiotweek2019\/keynote.html\" target=\"_blank\" rel=\"noopener noreferrer\">Keynote<\/a>, Cyber-Physical Systems & Internet-of-Things Week 2019 in Montreal, Canada (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, April 16, 2019)<\/li>\n<li>Invited Talk at the ASPLOS 2019 workshop &#8220;Unlocking the Power of Edge Computing&#8221; (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\" target=\"_blank\" rel=\"noopener noreferrer\">Ganesh Ananthanarayanan<\/a>, April 14, 2019)<\/li>\n<li>Plenary on \u201cAI\/ML for Communication Networks\u201d at the ICNC 2019 conference in Honolulu, HI (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, Feb. 19, 2019)<\/li>\n<li>&#8220;From Intelligent Cloud to Intelligent Edge&#8221; Workshop at UIUC CSL (Organizers: <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/shadi\/\">Shadi Noghabi<\/a> & <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/yushu\/\">Yuanchao Shu<\/a>, Feb. 7, 2019)<\/li>\n<li>Opening plenary at the 14th annual UIUC CSL Student Conference (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>, Feb. 6, 2019)<\/li>\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/www.nature.com\/articles\/s41928-018-0198-6\" target=\"_blank\" rel=\"noopener noreferrer\">Working on the edge<\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/www.nature.com\/natelectron\/\" target=\"_blank\" rel=\"noopener noreferrer\">Nature Electronics<\/a> Q\/A with <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a> (Jan. 15, 2019)<\/li>\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/arjmands\/\">Arjmand Samuel<\/a>, Principle Program Manager of Azure IoT Edge, gave the keynote at ACM\/IEEE SEC 2018 (Oct. 26, 2018)<\/li>\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/10-years-is-an-eternity-in-the-tech-world-but-we-are-just-getting-started\/\">Blog on the 10th anniversary of edge computing<\/a> by Victor Bahl (Oct. 19, 2018)<\/li>\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/shadi\/\">Shadi Noghabi<\/a> from UIUC joins Microsoft Research and our project as a full time Researcher (Oct. 1, 2018)<\/li>\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\">Ganesh Ananthanarayanan<\/a> named program co-chair of ACM\/IEEE SEC 2019 in Washington D.C (Sept. 9, 2018)<\/li>\n<li>Our <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/videoedge-processing-camera-streams-using-hierarchical-clusters\/\">VideoEdge<\/a> paper accepted with impressive review scores in ACM\/IEEE SEC 2018 (July 31, 2018)<\/li>\n<li>Check out our workshop: <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/at-the-bleeding-edge-of-intelligent-edges\/\">At the bleeding edge of Intelligent Edges<\/a> in Redmond WA (July 27, 2018)<\/li>\n<li>Edge was a hot-topic in <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/www.usenix.org\/conference\/hotcloud18\" target=\"_blank\" rel=\"noopener noreferrer\">HotCloud 2018<\/a> in Boston, MA. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\">Ganesh Ananthanarayanan<\/a> was PC co-chair of this USENIX workshop (July 9, 2018)<\/li>\n<li>Rashmi Misra, GM IoT & AI Solution and Rushmi Malaviarachchi, Partner GPM, Windoes IoT to keynote at MSR&#8217;s upcoming <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/at-the-bleeding-edge-of-intelligent-edges\/\">workshop<\/a><\/li>\n<\/ul>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Industries ranging from manufacturing to healthcare are eager to develop real-time control systems that use machine learning and artificial intelligence to improve efficiencies and reduce cost. We are exploring this new computing paradigm by identifying and addressing emerging technology and business model challenges.<\/p>\n","protected":false},"featured_media":507527,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"research-area":[13547],"msr-locale":[268875],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-212082","msr-project","type-msr-project","status-publish","has-post-thumbnail","hentry","msr-research-area-systems-and-networking","msr-locale-en_us","msr-archive-status-active"],"msr_project_start":"2008-10-29","related-publications":[498539,466809,391661,422301,367685,383729,323312,274497,422313,168760,168349,166573,510167,159073,158176,686664,688023,713815,585133,752782,603462,757663,606075,757669,630531,772063,634896,819391,646104,851380,661695,683163],"related-downloads":[],"related-videos":[],"related-groups":[],"related-events":[],"related-opportunities":[],"related-posts":[],"related-articles":[],"tab-content":[{"id":0,"name":"Research","content":"<div>\r\n<h2>Applications<\/h2>\r\nFrom the very beginning, we have maintained that the most compelling applications for edge computing are ones that require low latency responses or ones where the network to the cloud is expensive or inadequate. In this context, we asserted that the \u201ckiller app\u201d for edge computing is live video analytics. Along the way, other Microsoft researchers discovered precision agriculture to be a beautiful edge computing application as well. We are exploring both:\r\n\r\n[row][column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/live-video-analytics\/\">Live Video Analytics<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/live-video-analytics\/\"><img class=\"wp-image-509612 alignnone\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/Live_Analytics_480x280-300x175.jpg\" alt=\"Live Video Analytics Scenrios\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\nLarge-scale video processing is a grand challenge representing an important frontier for analytics, what with videos from factory floors, traffic intersections, police vehicles, and retail shops. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/live-video-analytics\/\">Read more<\/a>.\r\n\r\n[\/column] [column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/farmbeats-iot-agriculture\/\">Data Driven Agriculture<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/farmbeats-iot-agriculture\/\"><img class=\"alignnone wp-image-509597 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/MS_AI_Agriculture_DataVis_001-480-x-280.jpg\" alt=\"Photography depicts Microsoft's FarmBeats technology uses AI and IoT to help increase farm productivity.\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\nWe believe that data, coupled with the farmer\u2019s knowledge and intuition, can help increase farm productivity and help reduce costs. However, getting data from the farm is difficult since there is often no power in the field... <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/farmbeats-iot-agriculture\/\">Read more<\/a>.\r\n\r\n[\/column][\/row]\r\n<h2>Highly Adaptive and Resilient Edges<\/h2>\r\n<b>Continuity of service<\/b> is a must-have attribute in mission- and safety- critical edge computing applications. For example, in telecommunication networks interruption of communication services is unexceptionable; in oil rigs (<a href=\"https:\/\/customers.microsoft.com\/en-us\/story\/exxonmobil-mining-oil-gas-azure\" target=\"_blank\" rel=\"noopener\">link<\/a>) where continuous monitoring of the safety of on-site workers and the health of multi-million-dollar equipment is a must have; in manufacturing, constantly looking-out for production errors that may lead to defective items is also a must-have. Downtime of edge computing servers, beyond an acceptable threshold, can result in accidents and significant financial loss. The challenge then is to keep everything operational, even when local technicians are absent. This requires edge servers to be up and running (24x7x365), which is particularly hard when edge-cloud connectivity is unreliable and expensive. We are developing <em>adaptive and resilient software solutions<\/em> that enable edge clusters to run continuously.\r\n\r\n[row][column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/paya\/\">Paya<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/paya\/\"><img class=\"alignnone wp-image-626226 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2019\/12\/project-paya_480x280.jpg\" alt=\"to describe Project Paya\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\n<em>Paya<\/em> is a state migration edge-tailored solution that ensures that high-availability is met for all edge-cloud applications based on the application\u2019s specific need. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/paya\/\">Read more<\/a>.\r\n[\/column]\r\n\r\n[column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/adaptation-at-the-edge\/\">LOKI<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/adaptation-at-the-edge\/\"><img class=\"alignnone wp-image-626136 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2019\/12\/project-loki_480x280.jpg\" alt=\"to describe Project LOKI\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\n<em>LOKI<\/em> is a suite of services and programming abstractions that simplify the development of adaptive edge-cloud &amp; multi-cloud applications, <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/adaptation-at-the-edge\/\">Read more<\/a>.\r\n[\/column][\/row]\r\n\r\nOur goal is to develop a software-based system and architecture that can help keep operations alive by healing the system automatically in the face of machine failures, network disconnections, dynamic application loads, or changes in capacity. Towards this north star, we have several ongoing projects that take on these problems.\r\n\r\n[row][column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/hybridkube\/\"> HybridKube<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/hybridkube\/\"><img class=\"alignnone wp-image-626223 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2019\/12\/HybridKube_480x280.jpg\" alt=\"to describe HybridKube\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\n<em>HybridKube<\/em> is a Kubernetes extension for optimal placement of applications across a edge-cloud environment. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/hybridkube\/\">Read more<\/a>.\r\n\r\n[\/column] [column class=\"m-col-12-24\"]\r\n[\/column][\/row]\r\n<h2>Offloading Computations<\/h2>\r\nWe have been exploring the fundamental trade-off between computation and communications to enable a new class of cpu-, gpu- and data-intensive applications that seamlessly augment the cognitive abilities of users by exploiting speech recognition, NLP, vision, machine learning, and augmented reality (Project Maui, Mobisys 2010). We have made significant progress in overcoming the energy and computation limitations of sensors, handhelds, and wearables. In subsequent research we demonstrated how important special-purpose workloads can also leverage cloud offload: for GPU-intensive rendering applications (Project Kahawai, MobiSys 2015) and deep neural network video stream processing (MCDNN, MobiSys 2016).\r\n\r\n[row][column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/mobile-assistance-using-infrastructure-maui\/\">Project MAUI<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/mobile-assistance-using-infrastructure-maui\/\"><img class=\"alignnone wp-image-509627 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/Project-Maui-480-x-280.png\" alt=\"Image of chess being played on phone to describe Project Maui\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\n<em>Mobile Assistance Using Infrastructure<\/em> (MAUI) was the first system to demonstrate fine-grained code offload to nearby edge server(s) with minimal programmer effort. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/mobile-assistance-using-infrastructure-maui\/\">Watch the video<\/a>.\r\n\r\n[\/column] [column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/kahawai\/\">Project Kahawai<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/kahawai\/\"><img class=\"alignnone wp-image-509786 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/kahawai-STILL.jpg\" alt=\"Image of person on computer\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\nKahawai enables high-quality gaming on mobile devices, such as tablets and smartphones, by offloading a portion of the GPU computation to server-side infrastructure. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/kahawai\/\">Watch the video<\/a>.\r\n\r\n[\/column][\/row]\r\n<h2>Geo-distributed Edge Analytics<\/h2>\r\nEdge servers located in thousands of locations and managed by the same administrative entity offer powerful computing resources for cloud providers. Our research on low-latency edge analytics explores how best to use these resources. For example, the old approach of aggregating all the data from sensors to a single data center negatively impacts the timeliness of the analytics. But, running queries over geo-distributed inputs using the current intra-DC analytics frameworks also results in high query response times because these frameworks cannot cope with the relatively low and variable capacity of the WAN links. Our Iridium system (SIGCOMM 2015) provides low latency geo-distributed analytics by optimizing placement of both data and tasks of the queries. Follow-on work (CLARINET, OSDI 2016) considers WAN links with heterogeneous and modest bandwidths, unlike intra-datacenter networks, when deriving query execution plans across the cloud and edge servers.\r\n<h2>ML for Edge<\/h2>\r\n<p style=\"text-align: left\">Our colleagues in Microsoft Research India are developing a library of efficient machine learning (ML) algorithms that can run on resource-constrained edge and IoT devices ranging from the Arduino to the Raspberry Pi. The thesis is that IoT devices and sensors don't have to be \u201cdumb\u201d i.e. they can do more than just sense their environment and transmit their readings to the cloud, which is where the traditional decision making intelligence resides. Instead, an alternative paradigm is where even tiny, resource-constrained IoT devices run ML algorithms locally. This enables important scenarios overcoming concerns around connectivity to the cloud, latency, energy, privacy, and security. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/resource-efficient-ml-for-the-edge-and-endpoint-iot-devices\/\">Read more<\/a><\/p>\r\n\r\n<h2>Networking<\/h2>\r\nCloud providers, such as Microsoft, have two types of edges: On-net edges or Off-net edges. On-net edges are generally easier to operate, manage and maintain as they are on the cloud provider\u2019s network. In contrast, Off-net edges are connected to the cloud via the Internet, which may include several ISPs. Managing and operating such edges can be challenging due to the vagaries of the Internet. We are investigating problems to improve the network connectivity to our Off-net edges and the networking between the edges and sensors. Furthermore, edges provide us an opportunity to (re) investigate old ideas around low-latency, secure, overlay networking.\r\n<h2>Security<\/h2>\r\nCloud companies spend large amounts of money to physically secure their millions of servers located in their many data centers. In contrast, edge computing servers may or may not be physically secured. This opens the possibility of malicious attacks on the edge and cloud infrastructure. While a lot has been done to physically secure assets in the cloud, we are investigating techniques to do the same for our edge assets. Security and trust require authenticity and integrity, so we are investigating the use of sensors and specialized hardware in combination with new programming abstractions and system support for building secure and trusted edges. This research builds on our prior work on trusted sensors (MobiSys 2012, ASPLOS '14) and recent product offering (Azure Sphere).\r\n<h2>Cloud Services<\/h2>\r\nBefore the dawn of edge computing, which has brought about a major cloud computing paradigm shift in the industry, we developed, deployed and operated a <em>cloud<\/em> <em>service-store <\/em>under the banner of <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/\">Project Hawaii<\/a>. With it we empowered developers to build sophisticated, cloud-enhanced applications for their resource constraint devices. Our cloud service store included a variety of services including: optical character recognition, speech-to-text, path prediction, social computing, language translation, relay, rendezvous, etc. for Windows, Android, &amp; IOS devices. Over 60 universities used our services as a teaching aid for senior and graduate-level mobile + cloud computing courses. 2015 onward similar cloud services were commercialized by all major cloud providers under the banner of cognitive services. Check out <a href=\"https:\/\/azure.microsoft.com\/en-us\/services\/cognitive-services\/\">Azure cognitive services<\/a><strong>. <\/strong>Historically speaking, Project Hawaii was the first to show how cloud\/edge can be used in conjunction with a resource-constraint mobile device to augment human abilities.\r\n\r\n[row][column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/\">Project Hawaii<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/\"><img class=\"wp-image-510386 alignnone\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/project-hawaii.jpg\" alt=\"Project Hawaii group photo\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\nThe Project Hawaii team - BACK ROW (left to right): Gleb Krivosheev, Philip Fawcett, Ronnie Chaiken; FRONT ROW (left to right): <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/arjmands\/\">Arjmand Samuel<\/a>, <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/padhye\/\">Jitu Padhye<\/a>, <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/alecw\/\">Alec Wolman<\/a>, <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/\">Read more<\/a>.\r\n\r\n[\/column] [column class=\"m-col-12-24\"]\r\n<h3><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/#!gallery\">Gallery<\/a><\/h3>\r\n<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/#!gallery\"><img class=\"alignnone wp-image-510383 size-full\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2018\/10\/project-hawaii-GALLERY.png\" alt=\"Gallery image of Project Hawaii\" width=\"480\" height=\"280\" \/><\/a>\r\n\r\nA utility tool developed by a student for on-the-go translations. Project Hawaii\u2019s OCR &amp; S2T services, and Bing Translator were used. Check out our Gallery for dozens of student created featured projects. <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/project-hawaii\/#!gallery\">Read more<\/a>.\r\n\r\n[\/column][\/row]\r\n<\/div>"},{"id":1,"name":"Keynotes","content":"Mar. 5, 2020 | <a href=\"https:\/\/saiconference.com\/FICC2020\/Speakers\" target=\"_blank\" rel=\"noopener\">Future of Information and Communication Conference (FICC) 2020<\/a> - Edge Computing for the (Telecom) Infrastructure | Victor Bahl | <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/03\/FICC-2020-Keynote.pdf\" target=\"_blank\" rel=\"noopener\">Download Presentation<\/a>\r\n\r\nNov. 11, 2019 | <a href=\"http:\/\/www.ieee-icii.org\/key.html\" target=\"_blank\" rel=\"noopener\">IEEE International Conference on Industrial Internet 2019<\/a> - Edge Computing: Where are we today and what\u2019s next? | Victor Bahl\r\n\r\nOct. 3, 2019 | <a href=\"https:\/\/cra.org\/ccc\/events\/wide-area-data-analytics\/\" target=\"_blank\" rel=\"noopener\">CRA\/CCC Visioning Activity: Wide Area Data Analytics<\/a> - Live Video Analytics (extracting actionable insights from cameras) | Victor Bahl | <a href=\"https:\/\/cra.org\/ccc\/wp-content\/uploads\/sites\/2\/2020\/05\/2019-Bahl-CCC-CRA-Video-Analytics.pdf\" target=\"_blank\" rel=\"noopener\">Download Presentation<\/a>\r\n\r\nAug. 5, 2019 | <a href=\"https:\/\/aiotworkshop.github.io\/2019\/keynote.html\" target=\"_blank\" rel=\"noopener\">The 1<sup>st<\/sup> International Workshop on Artificial Intelligence of Things<\/a> - Fueling Industry 4.0 with the Intelligent Edge | Victor Bahl\r\n\r\nJun. 12, 2019 | <a href=\"http:\/\/cs.ucf.edu\/wowmom2019\/keynotes\/\" target=\"_blank\" rel=\"noopener\">20<sup>th<\/sup> IEEE Symposium on a World of Wireless, Mobile and Multimedia Networks 2019<\/a> - Edge + Wireless: Technologies fueling Industry 4.0 | Victor Bahl\r\n\r\nApril 16, 2019 | <a href=\"http:\/\/cpslab.cs.mcgill.ca\/cpsiotweek2019\/index.html\" target=\"_blank\" rel=\"noopener\">Cyber-Physical Systems and Internet-of-Things Week 2019<\/a> - Edge computing and the fourth industrial revolution | Victor Bahl\r\n\r\nFebruary 6, 2019 | Fourteenth Annual University of Illinois Urbana Champaign CSL Student Conference - Better together: the intelligent edge + the intelligent cloud | Victor Bahl\r\n\r\nAugust 20, 2018 | <a href=\"https:\/\/conferences.sigcomm.org\/sigcomm\/2018\/workshop-bigdama.html\" target=\"_blank\" rel=\"noopener\">SIGCOMM Workshop on Big Data Analytics and Machine Learning <\/a> - Democratizing Video Analytics \u2013 The quest for the holy trinity of low latency, low cost, and high accuracy | Ganesh Ananthanarayanan\r\n\r\nAugust 20, 2018 | <a href=\"http:\/\/conferences.sigcomm.org\/sigcomm\/2018\/workshop-mecomm.html\" target=\"_blank\" rel=\"noopener\">SIGCOMM Workshop on Mobile Edge Communications<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/edge-computing-a-historical-perspective-and-direction\/\">Edge computing: a historical perspective &amp; direction<\/a> (slides) | Victor Bahl\r\n\r\nJuly 27, 2018 | Bleeding Edge of Intelligent Edge - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/intelligent-edge\/\">Edge computing: 10 years and counting<\/a> (video) | Victor Bahl\r\n\r\nOctober 23, 2017 | <a href=\"http:\/\/sn.committees.comsoc.org\/call-for-papers\/the-14th-ieee-international-conference-on-mobile-ad-hoc-and-sensor-systems-mass-2017\/\" target=\"_blank\" rel=\"noopener\">IEEE Fourteenth International Conference on Mobile Ad Hoc and Sensor Systems<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/live-video-analytics\/\">Live Video Analytics<\/a> (video) | Victor Bahl\r\n\r\nOctober 15, 2017 | <a href=\"http:\/\/www.sis.pitt.edu\/cic2017\/\" target=\"_blank\" rel=\"noopener\">Third IEEE International Conference on Collaboration and Internet Computing<\/a> - Democratizing Video Analytics | Victor Bahl\r\n\r\nSeptember 29, 2017 | <a href=\"https:\/\/engineering.buffalo.edu\/computer-science-engineering\/news-events\/cse50.html\" target=\"_blank\" rel=\"noopener\">Emerging Topics in Computing Symposium, University at Buffalo Computer Systems Engineering Dept. 50th Anniversary<\/a> - Live Video Analytics the Perfect Edge Computing Application | Victor Bahl\r\n\r\nDecember 10, 2016 | <a href=\"http:\/\/ipccc.org\/ipccc2017\/main.php?page=1\" target=\"_blank\" rel=\"noopener\">IEEE International Performance Computing and Communications Conference<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/democratization-of-streaming-video-analytics-the-emergence-of-edge-computing\/\">Democratization of Streaming Video Analytics &amp; the Emergence of Edge Computing<\/a> (video) | Victor Bahl\r\n\r\nMay 13, 2015 | <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/devices-and-networking-summit-2015\/\">Devices and Networking Summit 2015<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/devices-and-networking-summit-plenary-keynote-3-victor-bahl\/\">Cloud 2020: Emergence of Micro Data Centers for Latency Sensitive Computing<\/a> | Victory Bahl\r\n\r\nMarch 10, 2015 | <a href=\"http:\/\/www.wikicfp.com\/cfp\/servlet\/event.showcfp?eventid=39482\" target=\"_blank\" rel=\"noopener\">IEEE Wireless Communications and Networking Conference (WCNC) 2015<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/emergence-of-micro-datacenter-cloudlets-edges-for-mobile-computing\/\">Cloud 2020: Emergence of Micro Data Centers (Cloudlets\/Edges) for Latency Sensitive Computing<\/a> (slides) | Victor Bahl\r\n\r\nFebruary 19, 2015 | <a href=\"http:\/\/www.conf-icnc.org\/2015\/\" target=\"_blank\" rel=\"noopener\">IEEE International Conference on Computing, Networking and Communications (ICNC) 2015<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/emergence-of-micro-datacenter-cloudlets-edges-for-mobile-computing\/\">Cloud 2020: Emergence of Micro Data Centers (Cloudlets\/Edges) for Latency Sensitive Computing<\/a> (slides) | Victor Bahl\r\n\r\nJune 27, 2014 | <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/summer-school-2014\/\">MSR Summer School on Advances in Wireless Networking<\/a> - <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/cloudlets-for-mobile-computing-2\/\">Cloudlets for mobile computing<\/a> (slides) | Victor Bahl\r\n\r\nNovember 22, 2013 | <a href=\"http:\/\/cloudnet2013.ieee-cloudnet.org\/\" target=\"_blank\" rel=\"noopener\">2nd IEEE International Conference on Cloud Networking (Cloudnet) 2013<\/a> - Cloud 2020: Emergence of Micro Data Centers for Latency Sensitive Computing | Victor Bahl"},{"id":2,"name":"Additional news","content":"[row]\r\n\r\n[card title=\"How we created edge computing\" url=\"https:\/\/www.nature.com\/articles\/s41928-018-0198-6\"]Edge computing processes data on infrastructure that is located close to the point of data creation. Mahadev Satyanarayanan recounts how recognition of the potential limitations of centralized, cloud-based processing led to this new approach to computing.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>Nature Electronics<\/em>\u00a0| January 16, 2019<\/p>\r\n[\/card]\r\n\r\n[card title=\"How Microsoft Got into Edge Computing and Real-Time Video Analytics\" url=\"https:\/\/winbuzzer.com\/2018\/10\/23\/how-microsoft-got-into-edge-computing-and-real-time-video-analytics-xcxwbn\/\"]Microsoft has discussed how it has been developing edge computing over the last decade to create real-time processing and analytics.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>WinBuzzer<\/em> | October 23, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"Microsoft Azure Picks Dell EMC For Edge Compute And Storage\" url=\"https:\/\/www.forbes.com\/sites\/paulteich\/2018\/10\/02\/microsoft-azure-and-dell-emc\/\"]Microsoft created Data Box Edge to ingest and analyze sensor data close to the sensors. It is a standard rack-mounted server chassis design with Xeon dual-processor power with a moderate amount of storage.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>Forbes<\/em> | October 2, 2018<\/p>\r\n[\/card]\r\n\r\n[\/row]\r\n[row]\r\n\r\n[card title=\"Microsoft Azure enables a new wave of edge computing. Here\u2019s how.\" url=\"https:\/\/azure.microsoft.com\/en-us\/blog\/microsoft-azure-enables-a-new-wave-of-edge-computing-here-s-how\/\" ]We are going through a technology transformation that is unlocking new scenarios that were simply not possible before. Smart sensors and connected devices are breathing new life into industrial equipment from factories to farms, smart cities to homes, while new devices are increasingly cloud connected by default \u2013 whether it\u2019s a car or a refrigerator.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Microsoft Azure Blog | September 24, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"Microsoft Targets Security, AI, Data, IoT and Edge Computing at Ignite\" url=\"https:\/\/www.channelpartnersonline.com\/2018\/09\/24\/microsoft-targets-security-ai-data-iot-and-edge-computing-at-inspire\/\"]Azure Data Box Edge is a physical network appliance, shipped by Microsoft, that helps businesses uses AI-enabled edge capabilities to analyze, process and transform data before uploading it to the cloud.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>Channel Partners Online<\/em> | September 24, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"The next wave of computing is the intelligent edge and intelligent cloud\" url=\"https:\/\/blogs.microsoft.com\/blog\/2018\/07\/23\/the-next-wave-of-computing-is-the-intelligent-edge-and-intelligent-cloud\/\" ]Take a look around your house, office or even the next store you visit, and you\u2019ll start to notice that internet-connected devices are bringing us closer than ever before to a world of ubiquitous computing and ambient intelligence.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Official Microsoft Blog | July 23, 2018<\/p>\r\n[\/card]\r\n\r\n[\/row]\r\n\r\n[row]\r\n[card title=\"New IoT and edge capabilities and programs to power partner and customer innovation\" url=\"https:\/\/blogs.microsoft.com\/iot\/2018\/07\/12\/new-iot-and-edge-capabilities-and-programs-to-power-partner-and-customer-innovation-simplify-journey-to-business-value\/\" ]Each year at our Inspire conference, we love hearing from our partners about the cloud solutions they are building, the business momentum they are seeing and the opportunities for growth. We heard early on that IoT\u2019s complexity was impeding...\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Microsoft IoT Blog | July 12, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"Azure IoT Edge generally available for enterprise-grade, scaled deployments\" url=\"https:\/\/azure.microsoft.com\/en-us\/blog\/azure-iot-edge-generally-available-for-enterprise-grade-scaled-deployments\/\"]Since we introduced Azure IoT Edge just over a year ago, we have seen many examples of the real-world impact from the factory floor to the farm to run cloud intelligence directly on IoT devices.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Microsoft Azure Blog | June 27, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"Advancing the future of society with AI and the intelligent edge\" url=\"https:\/\/blogs.microsoft.com\/blog\/2018\/05\/07\/advancing-the-future-of-society-with-ai-and-the-intelligent-edge\/\" ]The world is a computer, filled with an incredible amount of data. By 2020, the average person will generate 1.5GB of data a day, a smart home 50GB and a smart city, a whopping 250 petabytes of data per day.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Official Microsoft Blog | May 7, 2018<\/p>\r\n[\/card]\r\n\r\n[\/row]\r\n\r\n[row]\r\n\r\n[card title=\"Microsoft will invest $5 billion in IoT. Here\u2019s why.\" url=\"https:\/\/blogs.microsoft.com\/iot\/2018\/04\/04\/microsoft-will-invest-5-billion-in-iot-heres-why\/\" ]Today, we are announcing that we will invest $5 billion in the Internet of Things over the next four years. The reason we are doing this is simple: Our goal is to give every customer the ability to transform their businesses, and the world at large, with connected solutions.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\">Microsoft IoT Blog | April 4, 2018<\/p>\r\n[\/card]\r\n\r\n[card title=\"Could AI end car accidents?\" url=\"https:\/\/technical.ly\/dc\/2017\/09\/06\/ai-end-car-accidents\/\"]Imagine for a moment that, every week, four to five commercial airplanes crashed in America. In reality, a similar number of people die per week in traffic accidents, but, for the most part, those deaths don\u2019t resonate with us in the same way.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>Technical.ly<\/em> | September 6, 2017<\/p>\r\n[\/card]\r\n\r\n[card title=\"Microsoft Build 2017 buzzword bingo: On the edge\" url=\"https:\/\/www.zdnet.com\/article\/microsoft-build-2017-buzzword-bingo-on-the-edge\/\"]What Microsoft is doing in the edge computing\/distributed computing space may be a big topic at this week's Build 2017 developer conference. Here are a few clues as to why.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>ZDNet<\/em> | September 6, 2017<\/p>\r\n[\/card]\r\n\r\n[\/row]\r\n\r\n[row]\r\n\r\n[card title=\"Microsoft Research delivers cloud development kit for Windows Phone 7\" url=\"https:\/\/www.zdnet.com\/article\/microsoft-research-delivers-cloud-development-kit-for-windows-phone-7\/\" ]Microsoft Research has made available for download a developer preview of its Windows Phone 7 + Cloud Services Software Development Kit (SDK).The new SDK is related to Project Hawaii, a research initiative which I've blogged about before.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>ZDNet<\/em> | January 27, 2011<\/p>\r\n[\/card]\r\n\r\n[card title=\"Microsoft's new Hawaiian codenames are all about mobile\" url=\"https:\/\/www.zdnet.com\/article\/microsofts-new-hawaiian-codenames-are-all-about-mobile\/\" ]Oahu isn't Microsoft's only Hawaiian-themed code name. Project Hawaii from Microsoft Research, an initiative \"investigating how we can use the cloud to enhance how we use mobile devices.\"\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>ZDNet<\/em> | July 9, 2010<\/p>\r\n[\/card]\r\n\r\n[card title=\"Why a Cloudlet Beats the Cloud for Mobile Apps\" url=\"https:\/\/lewisshepherd.wordpress.com\/2009\/12\/13\/why-a-cloudlet-beats-the-cloud-for-mobile-apps\/\" ]Sure, you know cloud computing. You also know a bit about so-called \u201cprivate clouds,\u201d which enterprises and government agencies are exploring as an option to combine the power and scale of virtualized cloud architectures with security and control over data. But what do you know of Cloudlets? They may just be a key to the future of mobile computing.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>Shepherd's Pi<\/em> | December 13, 2009<\/p>\r\n[\/card]\r\n\r\n[\/row]\r\n[row]\r\n\r\n[card title=\"The Case for VM-based Cloudlets in Mobile Computing\" url=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/the-case-for-vm-based-cloudlets-in-mobile-computing\/\" ]Resource poverty is a fundamental constraint that severely limits the class of applications that can be run on mobile devices. This constraint is not just a temporary limitation of current technology, but is intrinsic to mobility. In this paper, we put forth a vision of mobile computing that breaks free of this fundamental constraint.\r\n<div style=\"height: 7px\"><\/div>\r\n<p style=\"font-size: 15px\"><em>IEEE Pervasive Computing<\/em> | November 1, 2009<\/p>\r\n[\/card]\r\n\r\n[\/row]"},{"id":3,"name":"Engagements","content":"<h2>Events (we helped organize)<\/h2>\r\n[row]\r\n\r\n[card title=\"IEEE Special issue on Pervasive Computing at the Edge\" url=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/edge-computing\/#!engagements\"]October 2020 | IEEE Pervasive Magazine[\/card]\r\n\r\n[card title=\"UIUC\/CSL Workshop on From Intelligent Cloud to Intelligent Edge\" url=\"http:\/\/studentconference.csl.illinois.edu\/overview\/corporate-day\/\" ]February 7, 2019 | UIUC, Illinois, USA[\/card]\r\n\r\n[card title=\"ACM\/IEEE Symposium on Edge Computing\" url=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/acm-ieee-sec-2018\/\" ]October 25-27, 2018 | Bellevue, Washington, USA[\/card]\r\n\r\n[\/row]\r\n\r\n[row]\r\n\r\n[card title=\"At the bleeding edge of Intelligent Edges\" url=\"https:\/\/www.microsoft.com\/en-us\/research\/event\/at-the-bleeding-edge-of-intelligent-edges\/\" ]July 27, 2018 | Redmond, Washington, USA[\/card]\r\n\r\n[card title=\"10th USENIX Workshop on Hot Topics in Cloud Computing\" url=\"https:\/\/www.usenix.org\/conference\/hotcloud18\" ]July 9, 2018 | Boston, Mass, USA[\/card]\r\n\r\n[card title=\"UW\/MSR Summer Institute on Unpacking the Future of IoT\" url=\"https:\/\/www.cs.washington.edu\/mssi\/2017\/\" ]July 31 - August 3, 2017 | Snoqualmie, Washington, USA USA[\/card]\r\n\r\n[\/row]\r\n\r\n[row]\r\n\r\n[card title=\"NSF Workshop on Grand Challenges in Edge Computing\" url=\"http:\/\/iot.eng.wayne.edu\/edge\/goals.php\" ]October 26, 2016 | Washington, DC, USA[\/card]\r\n\r\n[card title=\"International Workshop on Mobile Cloud Computing and Services (MCS)\" url=\"https:\/\/www.sigmobile.org\/workshops\/mcs\/\" ]2010-2015[\/card] [\/row]\r\n<h2>Research Community Service<\/h2>\r\n<h3>Special Issue<\/h3>\r\n<p style=\"padding-left: 15px\">2020 | <strong>Guest Editor <\/strong>(<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): <a href=\"https:\/\/www.computer.org\/digital-library\/magazines\/pc\/call-for-papers-pervasive-computing-at-the-edge-cfp\">Call for Papers: Pervasive Computing at the Edge<\/a>, IEEE Pervasive Magazine<\/p>\r\n\r\n<h3>Journals<\/h3>\r\n<p style=\"padding-left: 15px\">2017 - Present | <strong>Associate Editor <\/strong>(<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): <a href=\"https:\/\/tiot.acm.org\/\" target=\"_blank\" rel=\"noopener\">ACM Transactions on Internet of Things<\/a>\r\n2017 - Present | <strong>Associate Editor <\/strong>(<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): <a href=\"https:\/\/www.computer.org\/web\/tsc\/about\" target=\"_blank\" rel=\"noopener\">IEEE Transactions on Service Computing&lt;\/a&gt;\r\n2013 - Present | <strong>Advisory Board Member <\/strong>(<\/a><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a><a href=\"https:\/\/www.computer.org\/web\/tsc\/about\" target=\"_blank\" rel=\"noopener\">): <\/a><a href=\"http:\/\/ieee-iotj.org\/editorial-board\/\" target=\"_blank\" rel=\"noopener\">IEEE Internet of Things Journal<\/a><a href=\"https:\/\/www.computer.org\/web\/tsc\/about\" target=\"_blank\" rel=\"noopener\">\r\n2007-2018 | <strong>Editorial Board Member <\/strong>(<\/a><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a><a href=\"https:\/\/www.computer.org\/web\/tsc\/about\" target=\"_blank\" rel=\"noopener\">): <\/a><a rel=\"noopener\">Foundations and Trends\u00ae in Networking<\/a><\/p>\r\n\r\n<h3>Conferences &amp; Workshops<\/h3>\r\n<p style=\"padding-left: 15px\">2019 | <strong>Program Committee Co-Chair<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\">Ganesh Ananthanarayanan<\/a>): The Fourth ACM\/IEEE Symposium on Edge Computing\r\n2018 | <strong>Program Committee Co-Chair<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): <a href=\"http:\/\/acm-ieee-sec.org\/2018\/\" target=\"_blank\" rel=\"noopener\">The Third ACM\/IEEE Symposium on Edge Computing<\/a>\r\n2018 | <strong>Invited Speaker<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\">Ganesh Ananthanarayanan<\/a>): <a href=\"https:\/\/ewh.ieee.org\/conf\/sarnoff\/2018\/\" target=\"_blank\" rel=\"noopener\">IEEE Sarnoff Symposium<\/a>\r\n2018 | <strong>Program Committee Co-Chair<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/ga\/\">Ganesh Ananthanarayanan&lt;<\/a>): <a href=\"https:\/\/www.usenix.org\/conference\/hotcloud18\" target=\"_blank\" rel=\"noopener\">10th USENIX Workshop on Hot Topics in Cloud Computing<\/a>\r\n2016 | <strong>Advisor &amp; Steering Committee Member<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): <a href=\"http:\/\/iot.eng.wayne.edu\/edge\/goals.php\" target=\"_blank\" rel=\"noopener\">NSF Workshop on Grand Challenges in Edge Computing<\/a>\r\n2014 \u2013 Present | (Founding) <strong>Steering Committee Member<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): ACM\/IEEE Symposium on Edge Computing\r\n2010-2015 | (Founding) <strong>Steering Committee Member<\/strong> (<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>): ACM workshop on Mobile Cloud Computing and Services (MCS)<\/p>\r\n\r\n<h3>Distinguished Seminars (on Edge Computing)<\/h3>\r\n<p style=\"padding-left: 15px\">September 2018 | <strong>University of Southern California<\/strong>, Los Angles, CA | Ganesh Ananthanarayanan\r\nSeptember 11, 2017 | <strong>Rice University<\/strong>, Houston, Texas | Victor Bahl\r\nApril 28, 2017 | <strong>Washington University St. Louis<\/strong>, St. Louis, Missouri | Victor Bahl\r\nDecember 17, 2014 | <strong>Sorbonne Universit\u00e9<\/strong>, Paris, France | Victor Bahl\r\nNovember 20, 2014 | <strong>University College of London<\/strong>, London, U.K.| Victor Bahl\r\nOctober 3, 2014 | <strong>Yale University<\/strong>, New Haven, Connecticut | Victor Bahl<\/p>\r\n\r\n<h3>Panels (on Edge Computing)<\/h3>\r\n<p style=\"padding-left: 15px\">November 7, 2019 | <strong>Edge Computing: Where are we today and what\u2019s next?<\/strong> | &lt;a target=_blank href=\"http:\/\/acm-ieee-sec.org\/2019\/keynote%20and%20panel.php\"ACM\/IEEE Symposium on Edge Computing | Washington DC, USA | Panelist: <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>\r\nFebruary 19, 2019 | <strong>AI\/ML for Communication Networks<\/strong> | IEEE Intl. Conf. on Computing, Networking &amp; Communication | Honolulu, Hawaii, USA | Panelist: <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a>\r\nOctober 12, 2017 | <strong>Enabling Technologies for Edge Computing<\/strong> | <a href=\"http:\/\/acm-ieee-sec.org\/2017\/\" target=\"_blank\" rel=\"noopener\">Second ACM\/IEEE Symposium on Edge Computing <\/a> | San Jose, California, USA | Panelist: <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/bahl\/\">Victor Bahl<\/a><\/p>"},{"id":4,"name":"Outreach","content":"<div>\r\nThe <strong>Intelligent Edge<\/strong> is a continually expanding set of connected systems and devices that gather and analyze data\u2014close to your users, the data, or both.\r\n\r\nMicrosoft product groups coined the term The Intelligent Edge. The Intelligent Edge is a capability that enables Microsoft customers to enjoy a seamless experience and compute capabilities wherever their data exists\u2014in the cloud or offline. Microsoft is making it easier for developers to build apps that use edge technology, by open sourcing the Azure IoT Edge Runtime, which allows customers to modify the runtime and customize applications.\r\n\r\n<h2>Collaborations<\/h2>\r\n[row][column class=\"m-col-8-24\"]\r\n<h3 style=\"text-align: center\">Video Analytics over 5G Project<\/h3>\r\n<h3 style=\"text-align: center\">Princeton University<\/h3>\r\n<img class=\"aligncenter wp-image-656070 size-medium\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/02\/MSC17_dataCenter_046-300x200.jpg\" alt=\"\" width=\"300\" height=\"200\" \/>\r\n\r\nInternet-enabled cameras pervade daily life, generating a huge amount of data, but most of the video they generate is transmitted over wires and analyzed offline with a human in the loop.\u00a0 As a result, the amount of coverage and level of application accuracy that today\u2019s surveillance camera systems can provide is necessarily limited.\u00a0 Work has commenced both on scaling the stream processing behind video analytic systems and leveraging certain aspects of 5G technologies such as small cells.\u00a0 These ideas have in turn enabled exciting new applications for computing on the edge.\u00a0 The maturation of deep learning techniques and the complete 5G portfolio of technologies create exciting new opportunities to tackle even more challenging problems in video analytics.\r\n\r\nThis research is centered around <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/live-video-analytics\/\">Live Video Analytics <\/a>\u00a0occurring over a 5G network with multiple cameras and research program that leverages (a) The full suite of 5G technologies, including rapid mobility handover, small cells, and millimeter-wave (24 and 60 GHz) radio frequencies, and (b) Massive arrays of video cameras, backed by deep learning algorithms to process video jointly across the entire array of cameras.\u00a0\u00a0\u00a0 The use of 5G wireless links to each camera enables an unprecedented amount of wireless capacity to the edge devices, enabling buffering to be relegated to the edge device rather than situated onboard the camera\r\n\r\n[\/column]\r\n\r\n[column class=\"m-col-8-24\"]\r\n<h3 style=\"text-align: center\">Living Edge Lab<\/h3>\r\n<h3 style=\"text-align: center\">Carnegie Mellon<\/h3>\r\n<img class=\"wp-image-656067 size-medium aligncenter\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/02\/Azure-Stack-Edge-300x169.png\" alt=\"\" width=\"300\" height=\"169\" \/>\r\n\r\nCarnegie Mellon University and Microsoft are collaborating on a joint effort to innovate in edge computing, an exciting field of research for intensive computing applications that require rapid response times in remote and low-connectivity environments.\r\n\r\nBy bringing artificial intelligence to the \"edge,\" devices such as connected vehicles, drones or factory equipment\u00a0are able to quickly learn and respond to their environments, which is critical in scenarios like search and rescue, disaster recovery and safety.\r\n<p class=\"Normal1\">To enable discovery in these areas and more, Microsoft will contribute edge computing products to Carnegie Mellon for use in its\u00a0<a href=\"http:\/\/openedgecomputing.org\/lel.html\">Living Edge Laboratory<\/a>, a testbed for exploring applications that generate large data volumes and require intense processing with near-instantaneous response times.<\/p>\r\n[\/column][\/row]\r\n<\/div>"}],"slides":[],"related-researchers":[{"type":"user_nicename","display_name":"Ganesh Ananthanarayanan","user_id":31834,"people_section":"Researchers & Engineers","alias":"ga"},{"type":"user_nicename","display_name":"Victor Bahl","user_id":31167,"people_section":"Researchers & Engineers","alias":"bahl"},{"type":"user_nicename","display_name":"Landon Cox","user_id":37527,"people_section":"Researchers & Engineers","alias":"lacox"},{"type":"user_nicename","display_name":"Alex Crown","user_id":30817,"people_section":"Researchers & Engineers","alias":"acrown"},{"type":"user_nicename","display_name":"Sanjeev Mehrotra","user_id":33516,"people_section":"Researchers & Engineers","alias":"sanjeevm"},{"type":"guest","display_name":"Zephyr Yao","user_id":508607,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Enrique  Saurez Apuy","user_id":509735,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Jack Kolb","user_id":510146,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Shadi Noghabi","user_id":508610,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Chien-Chun Hung","user_id":509723,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Giulio Grassi","user_id":509717,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Kiryong Ha","user_id":508613,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Aakanksha Chowdhery","user_id":510152,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Tiffany Chen","user_id":510158,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Eduardo Cuero","user_id":508616,"people_section":"Interns","alias":""},{"type":"guest","display_name":"Kyle Jamieson","user_id":508598,"people_section":"Collaborators","alias":""},{"type":"guest","display_name":"Giovanni Pau","user_id":508601,"people_section":"Collaborators","alias":""},{"type":"guest","display_name":"Mahadev Satyanarayanan","user_id":508604,"people_section":"Collaborators","alias":""}],"msr_research_lab":[199565],"msr_impact_theme":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/212082"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-project"}],"version-history":[{"count":57,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/212082\/revisions"}],"predecessor-version":[{"id":659748,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/212082\/revisions\/659748"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/507527"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=212082"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=212082"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=212082"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=212082"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=212082"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}