Advancing HPC through oneAPI Heterogeneous Programming in Academia & Research – High-Performance Computing News Analysis

Intel oneAPI academic programs and centers of excellence

By Arti Gupta, Director oneAPI Academic and Community Outreach, Intel Corporation

The ever-increasing scale and speed of High Performance Computing (HPC) systems open up many new opportunities for researchers and data scientists. Today, the first exascale-ready HPC systems like the Department of Energy’s Aurora supercomputer are poised to deliver the goods. Upon full deployment at Argonne National Laboratory (ANL), Aurora is expected to exceed two exaFLOPS of double-precision computing power. While this system offers all the features required for demanding converged workloads, including AI training, modeling and simulation, a new development challenge arises. Modern HPC systems include heterogeneous architectures with specialized hardware components. Instead of coding for different languages ​​and proprietary approaches, developers prefer to write code once and know that it runs optimally on different architectures. For these reasons, traditional coding approaches for proprietary architectures applied to heterogeneous systems come with inherent complexities that can delay new science.

an API is helping solve this problem through an open industry initiative supported by over 100 organizations. oneAPI is an open, unified, cross-architecture programming model for CPUs and accelerator architectures (GPUs, FPGAs and others). The standards-based programming model simplifies software development and delivers uncompromised performance for accelerated computing without proprietary lock-in, while allowing integration with existing code. With oneAPI, developers can choose the best architecture for the specific problem they are trying to solve without having to rewrite software for the next architecture and platform.

Intel’s implementation of oneAPI includes the Intel® oneAPI Toolkits with advanced compilers, libraries, and analysis, debugging, and porting tools to help developers build cross-architecture code productively and accelerate performance. Intel is now leading efforts to help users in national labs, corporate organizations, and educational institutions use oneAPI successfully. Intel focuses on three academic initiatives to achieve this goal: the oneAPI Centers of Excellence, the oneAPI Educator Program, and the oneAPI Student Ambassador Program.

oneAPI Centers of Excellence

That Intel oneAPI Centers of Excellence The program is aimed at academic universities/research labs to support their efforts to simplify coding for multi-architecture systems. The centers deliver key software code optimizations and new implementations, port strategic applications to oneAPI, and develop curricula to drive adoption of oneAPI in the ecosystem. Intel offers the current 28 oneAPI Centers access to the Intel® Developer Cloud, which consists of the latest pre-production and current Intel CPU, GPU and FPGA architectures. Institutions in the program also benefit from Intel’s engineering resources and learning opportunities designed to get the most out of oneAPI.

Once the centers of excellence use these resources to optimize their codes for scientific applications in areas such as HPC simulations, biology or physics, the resulting code is freely available to any researcher who needs it. These centers regularly present their innovative work at conferences and events, publish research papers in peer-reviewed journals and offer oneAPI training for students. This collaborative effort makes it easier for others to benefit from real-world knowledge using oneAPI and best practices.

For example, Intel works closely with them KTH Royal Institute of Technology, Sweden, making the widely used molecular dynamics code – GROMACS – on an API future-proof solutions with a single SYCL code base for multi-architecture platforms across vendors. This center shows how a single SYCL code base allows running multiple devices such as Intel GPUs, AMD GPUs and NVIDIA GPUs.

Another example is the University of California, Davis center that develops deep learning tools for scientific visualization.

oneAPI Educator and Student Ambassador programs

The oneAPI Educator program and the accompanying oneAPI Student Ambassador program help professors and students to become familiar with the development of oneAPI and to apply this knowledge practically for cross-platform coding. As with the Centers of Excellence program, students and faculty receive technical support from Intel, access to the Developer Cloud, and an invitation to oneAPI developer events to share success stories and provide technical training and hands-on workshops on topics such as AI and code porting to SYCL, performance analysis and offloading code. However, there are some key differences between these academic programs.

The teacher centered oneAPI Educator Program provides academic institutions with ready-to-use curriculum, video presentations, homework, and quizzes to make the software development education and learning process more turnkey and effective. There are dozens of schools worldwide that are currently teaching oneAPI concepts in their classes, including the University of Southern California, Technion, and Loyola.

That oneAPI Student Ambassador Program follows a “pull-through” approach and encourages aspiring student developers in academia to lead innovative development and research projects, engage their peers and share oneAPI knowledge. Currently, participants include undergraduate, graduate and Ph.D. level student. Ambassadors conduct workshops tailored for student developers using content and training provided by Intel. After training on their campus, they can use their accumulated knowledge and resources to create workshops that help other students maximize their code for multi-architecture HPC and AI systems. There are students from schools around the world who have become oneAPI Student Ambassadors.

Learn more at Intel oneAPI training programs – Anyone interested in these programs is invited to apply.

Journey to Exascale

It’s exciting to imagine the breakthroughs that will result from exascale computing. Aurora and other exascale systems are opening the door to identifying new drugs, helping us to understand the mysteries of the universe, making nuclear power safer and more productive, and developing products that improve on those available today. At the moment, no one can foresee all the discoveries that will come in the years to come. However, it’s great to see that coding barriers have been removed, allowing scientists to focus more on their field and innovation instead of focusing on the complexities of coding. Then they can make optimal use of the HPC performance available to them.

Notices and Disclaimers

Intel technologies may require enabled hardware, software, or service activation.

No product or component can be absolutely safe.

Your costs and results may vary.

© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and marks may be claimed as the property of others.