{"id":922377,"date":"2023-02-28T01:06:20","date_gmt":"2023-02-28T09:06:20","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/"},"modified":"2023-07-10T03:37:33","modified_gmt":"2023-07-10T10:37:33","slug":"intelligent-cloud-and-edge-group","status":"publish","type":"msr-group","link":"https:\/\/www.microsoft.com\/en-us\/research\/group\/intelligent-cloud-and-edge-group\/","title":{"rendered":"Intelligent Cloud and Edge Group"},"content":{"rendered":"<section class=\"mb-3 moray-highlight\">\n\t<div class=\"card-img-overlay mx-lg-0\">\n\t\t<div class=\"card-background  has-background-gable-green card-background--full-bleed\">\n\t\t\t\t\t<\/div>\n\t\t<!-- Foreground -->\n\t\t<div class=\"card-foreground d-flex mt-md-n5 my-lg-5 px-g px-lg-0\">\n\t\t\t<!-- Container -->\n\t\t\t<div class=\"container d-flex mt-md-n5 my-lg-5 \">\n\t\t\t\t<!-- Card wrapper -->\n\t\t\t\t<div class=\"w-100 w-lg-col-5\">\n\t\t\t\t\t<!-- Card -->\n\t\t\t\t\t<div class=\"card material-md-card py-5 px-md-5\">\n\t\t\t\t\t\t<div class=\"card-body \">\n\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\n\n<h1 class=\"wp-block-heading\" id=\"intelligent-cloud-and-edge-group\">Intelligent Cloud and Edge Group<\/h1>\n\n\n\n<p><\/p>\n\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n<\/section>\n\n\n\n\n\n<p>The Intelligent Cloud and Edge (ICE) Group is at the forefront of advancing the artificial intelligence infrastructure and addressing the fundamental system issues of combining Cloud and Edge. Our mission is to provide efficient, user-friendly, cross-platform artificial intelligence training and deployment technologies.<\/p>\n\n\n\n<p>Our group&#8217;s cutting-edge research areas include deep learning compilation frameworks, optimization of new hardware accelerators, system design for new types of workloads such as graph neural networks, Mixture-of-Experts (MoE), scientific computing, and software-hardware co-design optimization for new intelligent scenarios such as gaming and multimedia.<\/p>\n\n\n\n<p>We take pride in our contributions to the field, as evidenced by several research achievements published in top academic conferences such as OSDI, NSDI, ATC, etc. Our main achievements have been open-sourced as projects such as <a><\/a><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\">NNFusion<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Rammer<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Roller<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/sparta\">SparTA<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/antares\">Antares<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/tutel\">Tutel<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, etc., and some of these technologies have been applied to product lines such as Xbox, Bing, and Office.<\/p>\n\n\n\n<p>Our research directions are focused on solving the most pressing challenges in building a cost-effective AI infrastructure by harvesting both the power of Cloud and Edge. We aim to advance the field by<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>exploring new frontiers in deep learning compilation frameworks, e.g., tensor compilation, full model optimization, etc.<\/li>\n\n\n\n<li>designing large model inference systems combining Cloud and Edge<\/li>\n\n\n\n<li>developing decentralized large model inference and training systems<\/li>\n\n\n\n<li>exploring the capability of new hardware accelerator architecture, e.g., mesh-based AI accelerators<\/li>\n\n\n\n<li>software-hardware co-designed system for sparse model computation<\/li>\n\n\n\n<li>designing graph neural networks (GNN) and Mixture-of-Experts (MoE) systems,<\/li>\n\n\n\n<li>accelerating AI-based workload like databases, gaming, multimedia, etc.<\/li>\n<\/ul>\n\n\n\n<p>Join us in our pursuit of pioneering AI system technology that will shape the future of the industry.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<p>\u667a\u80fd\u4e91\u7aef\u7cfb\u7edf\u7ec4\uff08Intelligent Cloud and Edge\uff09\u81f4\u529b\u4e8e\u7814\u7a76\u4e91\u7aef\u4e00\u4f53\u7684\u4eba\u5de5\u667a\u80fd\u57fa\u7840\u67b6\u6784\u53ca\u5176\u5173\u952e\u7cfb\u7edf\u95ee\u9898\uff0c\u4ee5\u63d0\u4f9b\u9ad8\u6548\u3001\u6613\u7528\u3001\u8de8\u5e73\u53f0\u7684\u4eba\u5de5\u667a\u80fd\u8bad\u7ec3\u548c\u90e8\u7f72\u6280\u672f\u3002\u5c0f\u7ec4\u76ee\u524d\u7814\u7a76\u65b9\u5411\u6db5\u76d6\u6df1\u5ea6\u5b66\u4e60\u7f16\u8bd1\u6846\u67b6\u3001\u65b0\u578b\u4eba\u5de5\u667a\u80fd\u52a0\u901f\u786c\u4ef6\u7684\u4f18\u5316\u3001\u9762\u5411\u65b0\u578b\u8d1f\u8f7d\uff08\u5982\u56fe\u795e\u7ecf\u7f51\u7edc\u3001MoE\u3001\u79d1\u5b66\u8ba1\u7b97\u7b49\uff09\u7684\u7cfb\u7edf\u8bbe\u8ba1\u3001\u65b0\u578b\u667a\u80fd\u573a\u666f\uff08\u5982\u6e38\u620f\u3001\u591a\u5a92\u4f53\u7b49\uff09\u7684\u8f6f\u786c\u4ef6\u534f\u540c\u4f18\u5316\u7b49\u3002\u5c0f\u7ec4\u7684\u591a\u9879\u7814\u7a76\u6210\u679c\u53d1\u8868\u5728OSDI\u3001NSDI\u3001ATC\u7b49\u9876\u7ea7\u5b66\u672f\u4f1a\u8bae\uff0c\u5176\u4e2d\u4e3b\u8981\u6210\u679c\u5747\u4ee5\u5f00\u6e90\u9879\u76ee\u7684\u5f62\u5f0f\u5bf9\u5916\u5f00\u653e\uff08\u5982<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\">NNFusion<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>\u3001<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Rammer<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>\u3001<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Roller<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>\u3001<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/antares\">Antares<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>\u3001<a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/tutel\">Tutel<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>\u7b49\uff09\uff0c\u90e8\u5206\u6280\u672f\u4e5f\u88ab\u5e94\u7528\u5728\u8bf8\u5982Xbox\u3001Bing\u3001Office\u7b49\u4ea7\u54c1\u7ebf\u4e2d\u3002<\/p>\n\n\n\n<p>\u7814\u7a76\u65b9\u5411\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u6df1\u5ea6\u5b66\u4e60\u7f16\u8bd1\u6846\u67b6\u548c\u7f16\u8bd1\u6280\u672f<\/li>\n\n\n\n<li>\u4e91\u7aef\u7ed3\u5408\u7684\u5927\u6a21\u578b\u7684\u63a8\u7406\u7cfb\u7edf\u8bbe\u8ba1\u4e0e\u4f18\u5316<\/li>\n\n\n\n<li>\u53bb\u4e2d\u5fc3\u5316\u7684\u5927\u6a21\u578b\u63a8\u7406\u548c\u8bad\u7ec3\u7cfb\u7edf<\/li>\n\n\n\n<li>\u65b0\u578b\u786c\u4ef6\u52a0\u901f\u52a0\u7684\u6027\u80fd\u4f18\u5316<\/li>\n\n\n\n<li>\u9762\u5411\u7a00\u758f\u6a21\u578b\u8ba1\u7b97\u7684\u8f6f\u786c\u4ef6\u534f\u540c\u4f18\u5316<\/li>\n\n\n\n<li>\u56fe\u795e\u7ecf\u7f51\u7edc\u548cMoE\u7cfb\u7edf\u7684\u8bbe\u8ba1\u4e0e\u4f18\u5316<\/li>\n\n\n\n<li>\u57fa\u4e8eAI\u7684\u65b0\u578b\u8d1f\u8f7d\uff08\u5982\u6570\u636e\u5e93\u3001\u6e38\u620f\u3001\u591a\u5a92\u4f53\uff09\u7684\u652f\u6301\u4e0e\u52a0\u901f\u7b49<\/li>\n<\/ul>\n\n\n\n<p><\/p>\n\n\n\n\n\n<h4 class=\"wp-block-heading\" id=\"some-on-going-projects\">Some on-going projects:<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/deep-learning-compiler-and-optimizer\/\">Deep Learning Compiler and Optimizer<\/a><\/li>\n<\/ul>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"some-open-sourced-projects\">Some open-sourced projects:<\/h4>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\">NNFusion<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.<\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Rammer<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: A DNN compiler technology that can generate an efficient static spatio-temporal schedule for a DNN at compile time to minimize scheduling overhead.<\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/nnfusion\/tree\/osdi20_artifact\/artifacts\">Roller<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: A fast and efficient tensor compiler for DNN that can generate efficient kernels in&nbsp;<em>seconds <\/em>with a construction-based approach.<\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/sparta\">SparTA<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: an end-to-end optimization system to harvest the speeding up gain from the model sparsity.<\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/antares\">Antares<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: an automatic engine for multi-platform kernel generation and optimization. Supporting CPU, CUDA, ROCm, DirectX12, GraphCore, SYCL for CPU\/GPU, OpenCL for AMD\/NVIDIA, Android CPU\/GPU backends.<\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/microsoft\/tutel\">Tutel<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>: An Optimized Mixture-of-Experts Implementation.<\/li>\n<\/ul>\n\n\n\n\n\n<p><strong>Intern:<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/www.msra.cn\/zh-cn\/jobs\/interns\/intelligent-cloud-and-edge-group-research-intern?language=chinese\">Research Intern<span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/li>\n<\/ul>\n\n\n","protected":false},"excerpt":{"rendered":"<p>The Intelligent Cloud and Edge (ICE) Group is at the forefront of advancing the artificial intelligence infrastructure and addressing the fundamental system issues of combining Cloud and Edge. Our mission is to provide efficient, user-friendly, cross-platform artificial intelligence training and deployment technologies. Our group&#8217;s cutting-edge research areas include deep learning compilation frameworks, optimization of new [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_group_start":"","footnotes":""},"research-area":[13547],"msr-group-type":[243694],"msr-locale":[268875],"msr-impact-theme":[264846],"class_list":["post-922377","msr-group","type-msr-group","status-publish","hentry","msr-research-area-systems-and-networking","msr-group-type-group","msr-locale-en_us"],"msr_group_start":"","msr_detailed_description":"","msr_further_details":"","msr_hero_images":[],"msr_research_lab":[199560],"related-researchers":[{"type":"user_nicename","display_name":"Wei Cui","user_id":38859,"people_section":"Section name 0","alias":"weicu"},{"type":"user_nicename","display_name":"Peichen Xie","user_id":40624,"people_section":"Section name 0","alias":"peichenxie"}],"related-publications":[428202,581590,596290,651492,700210,831574,858267,923226,923232,941055,954468,954474],"related-downloads":[],"related-videos":[],"related-projects":[],"related-events":[],"related-opportunities":[],"related-posts":[963594],"tab-content":[],"msr_impact_theme":["Computing foundations"],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/922377","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-group"}],"version-history":[{"count":14,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/922377\/revisions"}],"predecessor-version":[{"id":954498,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/922377\/revisions\/954498"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=922377"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=922377"},{"taxonomy":"msr-group-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group-type?post=922377"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=922377"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=922377"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}