February 2021 Newsletter

Greetings IXPUG members around the globe. As we welcome 2021, we are pleased to announce the newly elected IXPUG Leadership board for the 2021-2023 term:

President: R. Glenn Brook, University of Tennessee Knoxville
Vice President: Thomas Steinke, Zuse Institute Berlin
Secretary: Amit Ruhela, Texas Advanced Computing Center

IXPUG Webinars

Migrating from CUDA-only to Multi-Platform DPC++ Thursday, February 18 | 8:00–9:00AM PT

In our next live webinar Migrating from CUDA-only to Multi-Platform DPC++, Steffen Christgau (Zuse Institute Berlin (ZIB)) will expand on his experiences with portable code and case study first shared in his lightning talk at the IXPUG Annual Conference 2020. Check it out, then join us for the live webinar Thursday, February 18 | 8:00–9:00AM PT. Registration is free to all IXPUG members and guests.

Webinar details

Steffen Christgau, research associate in the Algorithms for Innovative Architectures research group of the Supercomputing Department at ZIB, will demonstrate how an existing CUDA stencil application code can be migrated to DPC++ with the help of the Compatibility Tool. He will highlight and discuss the crucial differences between the two programming environments in the context of migrating the tsunami simulation easyWave. The discussion also includes steps for making the code to compliant with the SYCL standard. During the talk, he will also show that the migrated code can run on a wide range of platforms starting from CPUs, over GPUs, to FPGAs. Marius Knaust (ZIB) will join to answer FPGA-related questions from the audience.

Performance Optimization with Large Data Sets In-Memory: Applying AI in HPC Applications and Simulations At-Scale Thursday, March 11 | 8:00–9:00AM PT

Save the date for this live webinar with Meena Arunachalam, Principal Engineer, Intel Architecture, Graphics and Software. View the full abstract and register HERE.

Events

HPC Asia 2021

IXPUG’s first virtual event in the Asia region, IXPUG Workshop at HPC Asia 2021, took place January 22, 2021. We are grateful to Organizing Co-Chairs Taisuke Boku (University of Tsukuba) and Toshihiro Hanawa (The University of Tokyo) and the entire Program Committee for their contributions and planning resulting in this half-day, open workshop on high performance computing applications, systems, and architecture with Intel technologies. Here are quick links to replay the keynote and other sessions:

• Keynote Address: Advancing HPC Together
• High Performance Simulations of Quantum Transport using Manycore Computing
• Distributed MLPerf ResNet50 Training on Intel Xeon Architectures with TensorFlow
• Invited Talk: oneAPI Industry Initiative for Accelerated Computing
• Single-Precision Calculation of Iterative Refinement of Eigenpairs of a Real Symmetric-Definite Generalized Eigenproblem by Using a Filter Composed of a Single Resolvent
• A Comparison of Parallel Profiling Tools for Programs Utilizing the FFT
• Efficient Parallel Multigrid Method on Intel Xeon Phi Clusters

IXPUG Annual Conference YouTube Playlist

Did you know that the IXPUG Annual Conference 2020 individual sessions are now conveniently available to replay via our YouTube playlist? Check out the keynotes by Michael Wong (Codeplay) “The Landscape of Modern Parallel Programming Using Open Standards” and “How HPC Developers Using SYCL Need to Deal with Data Movement” and subscribe to the channel. Also download slides on any conference topics of your interest.

Other Events and Replays

• Intel® AI Workshop 2021 (February 11, 2020 online) co-organized by KAUST Supercomputing Core Lab and Intel
• HPC User Forum Special Event on Big Memory (December, 2020 replay)
• oneAPI Developer Summit 2020 (November, 2020 replay) ecosystem developer talks

Podcasts and Articles

• Podcast: Expanding the SYCL 2020 Ecosystem with hipSYCL and DPC++ – Aksel Alpay, Engineer at Heidelberg University Computing Center and Jeff Hammond, Principal Engineer at Intel. Open standards underpin the future of heterogeneous programming, and as Intel Principal Engineer Jeff Hammond points out, SYCL has proven in many ways that “standards don’t cost you a thing and you get portability.” A modern C++ language, SYCL has demonstrated amazing growth and now boasts four implementations—ComputeCpp, triSYCL, hipSYCL and DPC++. hipSYCL Creator Aksel Alpay explains some of the differences between these implementations, and how hipSYCL is evolving to incorporate DPC++ features contributed to the SYCL 2020 specification, including unified shared memory (USM), reductions, group algorithms and subgroups.
• Podcast: Igniting the Next Generation of Deep Learning – Penporn Koanantakool, Google Senior Software Engineer, and Ramesh, Intel Principal Engineer, share how their teams are collaborating to optimize TensorFlow for the latest Intel technologies using oneAPI Deep Neural Network Library (oneDNN).
• Article: DPC++ Foundations Code Sample Walk-Through by Dylan Benito, Intel Community. This sample walk-through features vector_add as a program that demonstrates oneAPI concepts and functionality, using the Data Parallel C++ (DPC++) programming language. The program adds two arrays of integers together using hardware acceleration.
• Article: Analyzing Memory and Threading Correctness for GPU-Offloaded Code by Kevin O’Leary, Lead Technical Consulting Engineer, and Michael Tutin, Software Architect, Intel Corporation
• Article: Solving a 2D Heat Equation Using Data Parallel C++ (DPC++) by Graham McKenzie, Systems Field Application Engineer, Intel Corporation

Resources

• oneAPI specification: https://www.oneapi.com/documentation/ As part of the oneAPI industry initiative, oneAPI elements have been open sourced to make it easy for developers to take advantage of oneAPI in their projects.
• Open access book: Data Parallel C++: Mastering DPC++ for Programming Heterogeneous Systems using C++ and SYCL (new, free, includes code examples)
• Looking for presentations from our past Workshops, Meetings, and BOFs? Check the IXPUG Resources page for a searchable repository of slides and recordings from all of our events.

ABOUT IXPUG

• IXPUG is an independent users group whose mission is to provide a forum for the free exchange of information that enhances the usability and efficiency of scientific and technical applications running on HPC, AI, and data analytics systems using advanced Intel technologies. IXPUG is administered by representatives of member sites that operate large Intel-based HPC systems.
• If you are interested in serving on the IXPUG Steering Committee, please contact us! For more information, see the IXPUG Charter or contact This email address is being protected from spambots. You need JavaScript enabled to view it..
• What technologies do you want to learn more about in 2021? Drop us a note at This email address is being protected from spambots. You need JavaScript enabled to view it. anytime—we want to hear from members!