Intel Omni-Path Architecture Adviser Newsletter

Intel® Omni-Path Architecture Click here to view this email as a web page. The Intel® Omni-Path Architecture Adviser Vol. II, Issue #4, April/May 2019 Intel® OPA This Month TACC's Stampede2 Helps Uncover Black Hole Optimizing Network Software for Scientific Applications New 2nd Generation Intel® Xeon® Scalable Platforms at HLRN Accelerating Deep Neural Network Training at BSC Intel® OPA Quick Links Intel OPA Home Page Intel OPA Blogs Intel® OPA Fabric Builders Intel OPA Webinars Intel® Select Solutions Intel Artificial Intelligence (AI) Intel AI Builders Intel High Performance Computing (HPC) Intel eXtreme Performance Users Group (IXPUG) TACC's Stampede2 Helps Uncover Black Hole On April 10, 2019, researchers from around the world shared the first image of a black hole centered in the galaxy Messier 87. Texas Advanced Computing Center's (TACC) Stampede2, with Intel® OPA is one of several supercomputers contributing to the extensive collaboration. Learn more about this groundbreaking work studying our universe. Read More > Optimizing Network Software for Scientific Applications Brookhaven National Laboratory researchers collaborated with Intel and University of Edinburgh engineers to optimize computationally intensive workloads. The team implemented improved MPI software to take full advantage of Intel OPA and parallel threads on Intel Xeon® Phi™ nodes. Read More > New 2nd Gen Intel® Xeon® Scalable Platforms at HLRN The North German Supercomputing Alliance (HLRN) chose 2nd Gen Intel® Xeon® Platinum platforms and Intel OPA for its newest HLRN4 supercomputer. This next-generation system will meet the ever-growing HPC and data analytics needs for universities and researchers from seven German states across diverse scientific disciplines. Watch the video to learn more. Read More > Accelerating Deep Neural Network Training at BSC Researchers at Barcelona Supercomputer Center are utilizing reduced precision floating point number formats to improve memory footprint and decrease bandwidth and processing time specifically to boost training of Deep Neural Networks. The team is using MareNorstrum4, a cluster with Intel Xeon Scalable processors and Intel OPA, to evaluate new algorithms with AI instructions to increase performance and reduce power. Read More > Intel OPA Training for Partners, Resellers, and Customers! - Where: Anytime, Anywhere! - How: Instructor Led Skype Sessions, Instructor Led On-Site Sessions, Optional Hands-On Labs, and Quarterly Workshops. - Sign Up: By contacting an Intel training representative at fabrictraining@intel.com Create World-Class Intel® Omni-Path Architecture-based Solutions with the Intel® Fabric Builders Community Technologies & Solutions Experience for yourself some of the world-class technologies and solutions that Intel® Fabric Builders have already created. Whether its high performance computing, artificial intelligence, or high performance data analytics, you can transform it with Omni-Path. Ecosystem Partners The Intel Fabric Builders ecosystem provides a collaborative network of partners dedicated to facilitating scientific and business innovation along your side. With Omni-Path Architecture technologies and Intel technical expertise, possibilities are limitless for ecosystem partners. Join the community here and learn more at the linked icons below Technologies & Solutions > Ecosystem Partners > About >

Still Need More HPC & AI? Check out the Intel® OPA Web Page, or visit us on social media Intel OPA home page

To subscribe to this newsletter, please click here. Subscribe

                                                           

This was sent to ivwinds.steeds@blogger.com because you are subscribed to Newsletters. To view and manage your marketing-related email preferences with Intel, please click here. © 2019 Intel Corporation Intel Corporation, 2200 Mission College Blvd., M/S RNB4-145, Santa Clara, CA 95054 USA. www.intel.com Privacy | Cookies | *Trademarks | Unsubscribe | Manage Preferences