Developer Resources from Intel and CERN*
CERN* took advantage of the Intel® oneAPI model that supports multiarchitecture programming via C++ with SYCL* to make use of hardware accelerators like GPUs in the ATLAS software framework.
Intel® oneAPI Base Toolkit components helped accelerate and optimize the code blocks. The SYCL API made it possible to achieve asynchronous running of interdependent tasks. Yet, achieving the same for CUDA* code required extra effort.
- Intel® Data Center GPU Max Series 1550 performed in competition with an NVIDIA* A100 GPU.
- Performance of SYCL code on an NVIDIA back end was very close to native CUDA* code that runs the same algorithm.
- Intel Data Center GPUs performed significantly better than what would be possible using even multiple traditional CPUs.
Get Started with the ATLAS Experiment by CERN
The ATLAS Experiment is a general-purpose particle physics experiment conducted by CERN at the Large Hadron Collider (LHC). It aims to study collisions of protons and heavy nuclei at high energy in order to understand the nature of the fundamental building blocks of matter in the universe.
Intel Case Study
The ATLAS Experiment implements heterogeneous particle reconstruction with Intel® tools.
"We're very happy to have the opportunity to collaborate with Intel and use the Intel tools to modernize key parts of our experiment's software. The Intel tools have been very valuable towards understanding and improving the performance of our software when running on GPUs."
— Zach Marshall, computing coordinator, ATLAS Experiment
"Scientists at the CERN Large Hadron Collider (LHC) project are working towards new scientific discoveries that will require them to analyze unprecedented data volumes on the most powerful HPC systems worldwide. The oneAPI concept of a unified programming model, built on open industry standard specifications, will allow for a seamless software development process for utilizing heterogeneous processing hardware infrastructures."
— Dr. Marcus Elsing, group leader, CERN ATLAS Processing Group
Intel® oneAPI Base Toolkit
Download the Stand-Alone Version
Develop performant code quickly and correctly across hardware targets, including CPUs, GPUs, and FPGAs, with this standards-based, multiarchitecture compiler.
Download the Stand-Alone Version
Efficiently debug cross-platform parallel and threaded applications developed in C, C++, SYCL*, OpenMP*, or Fortran.
Download the Stand-Alone Version
Find and fix performance bottlenecks and optimize application and system performance and configuration for HPC, cloud, IoT, media, storage, and more.
Download the Stand-Alone Version
Simplify the work of adding parallelism to complex applications, even if you're not a threading expert, with this advanced threading and memory management library.