diff --git a/content/events/2024/2024-09-24-training-session-introduction-to-quantum-computing-simulators-on-melu-xina.md b/content/events/2024/2024-09-24-training-session-introduction-to-quantum-computing-simulators-on-melu-xina.md new file mode 100644 index 0000000..f78aa2b --- /dev/null +++ b/content/events/2024/2024-09-24-training-session-introduction-to-quantum-computing-simulators-on-melu-xina.md @@ -0,0 +1,11 @@ +--- +contributor: scott +date: '2024-09-30T12:00:00' +starts: '2024-09-24T09:00:00+0500' +ends: '2024-09-24T17:00:00+0500' +title: 'Training Session: Introduction to Quantum Computing & Simulators on MeluXina' +external_url: 'https://www.luxprovide.lu/event/training-session-introduction-to-quantum-computing-simulators-on-meluxina/' +location: 'Digital Learning Hub Luxembourg' +--- + +Immerse yourself with captivating expert-led sessions and interactive hands-on demonstrations. diff --git a/content/events/2024/2024-09-30-heterogeneous-computing-with-sycl-and-one-api-on-cinec-as-leonardo.md b/content/events/2024/2024-09-30-heterogeneous-computing-with-sycl-and-one-api-on-cinec-as-leonardo.md new file mode 100644 index 0000000..d9499e2 --- /dev/null +++ b/content/events/2024/2024-09-30-heterogeneous-computing-with-sycl-and-one-api-on-cinec-as-leonardo.md @@ -0,0 +1,22 @@ +--- +contributor: scott +date: '2024-09-30T12:11:00' +starts: '2024-10-10T09:30:00+0200' +ends: '2024-10-10T11:30:00+0200' +title: 'Heterogeneous computing with SYCL and oneAPI on CINECA’s Leonardo' +external_url: 'https://epicure-hpc.eu/2024/09/25/heterogeneous-computing-with-sycl-and-oneapi-on-cinecas-leonardo/' +location: 'Online' +--- + +In this webinar attendees will learn the fundamentals of the SYCL programming model, exploring its key concepts and +capabilities, and how to leverage oneAPI tools to bridge SYCL code with proprietary programming models such as CUDA and +HIP, and deploy on a variety of hardware platforms. + +The webinar also features a detailed example of a successful CUDA to SYCL migration, showcasing the practical steps and +benefits of this approach. With ample time for questions and discussion, attendees will leave with a solid understanding +of SYCL and the confidence to apply it to their own HPC projects. + +Finally, the webinar also introduces developers to the SYCL oneAPI Hackaton at CINECA, a three-stage event (this +webinar, an online learning course, and a final on-site Hackaton) where attendees will learn how to effectively write +efficient heterogeneous SYCL code, and how to use oneAPI tools to deploy it on CINECA’s Leonardo, a Tier-0 EuroHPC +supercomputer ranked #7 on the Top500 list of the world’s most powerful supercomputers. diff --git a/content/events/2024/2024-09-30-introduction-to-one-api-sycl-2020-and-open-mp-offloading.md b/content/events/2024/2024-09-30-introduction-to-one-api-sycl-2020-and-open-mp-offloading.md new file mode 100644 index 0000000..a6c2fff --- /dev/null +++ b/content/events/2024/2024-09-30-introduction-to-one-api-sycl-2020-and-open-mp-offloading.md @@ -0,0 +1,17 @@ +--- +contributor: scott +date: '2024-09-30T12:11:00' +starts: '2024-09-23T08:45:00+0200' +ends: '2024-09-25T12:40:00+0200' +title: 'Introduction to oneAPI, SYCL2020 and OpenMP offloading' +external_url: 'https://oneapi.io/events/oneapi-devsummit-hosted-by-uxl-foundation/' +location: 'Online' +--- + +Most current HPC systems are heterogeneous and use accelerators. oneAPI is a standardized and portable programming model +adapted to heterogeneous computing. In this course we will provide an introduction to Intel's oneAPI implementation, +which supports two portable methods of heterogeneous computing: Data Parallel C++ (DPC) with SYCL and OpenMP for C, C++, +and Fortran. Both are portable on any Intel CPU and Intel based accelerator, but also other GPUs. The course will give +an introduction in these two programming methods, Intel's libraries like oneMKL and tools for performance analysis, +profiling, and debugging. Further an introduction to Intel's DPC compatibility Tool to facilitate code migration from +CUDA to SYCL and to Intel's MPI implementation support with GPU awareness completes the program. diff --git a/content/research_papers/2024/2024-09-25-mini-lb-a-performance-portability-study-of-lattice-boltzmann-simulations.md b/content/research_papers/2024/2024-09-25-mini-lb-a-performance-portability-study-of-lattice-boltzmann-simulations.md new file mode 100644 index 0000000..0bc3a58 --- /dev/null +++ b/content/research_papers/2024/2024-09-25-mini-lb-a-performance-portability-study-of-lattice-boltzmann-simulations.md @@ -0,0 +1,27 @@ +--- +contributor: scott +date: '2024-09-25T09:43:10' +title: 'miniLB: A Performance Portability Study of Lattice-Boltzmann Simulations' +external_url: 'https://www.arxiv.org/abs/2409.16781' +authors: + - Luigi Crisci, + - Biagio Cosenza + - Giorgio Amati + - Matteo Turisini +tags: + - sycl + - lbm + - cfd + - fluid-dynamics + - portability +--- + +The Lattice Boltzmann Method (LBM) is a computational technique of Computational Fluid Dynamics (CFD) that has gained +popularity due to its high parallelism and ability to handle complex geometries with minimal effort. Although LBM +frameworks are increasingly important in various industries and research fields, their complexity makes them difficult +to modify and can lead to suboptimal performance. This paper presents miniLB, the first, to the best of our knowledge, +SYCL-based LBM this http URL addresses the need for a performance-portable LBM proxy app capable of abstracting complex +fluid dynamics simulations across heterogeneous computing systems. We analyze SYCL semantics for performance portability +and evaluate miniLB on multiple GPU architectures using various SYCL implementations. Our results, compared against a +manually-tuned FORTRAN version, demonstrate effectiveness of miniLB in assessing LBM performance across diverse +hardware, offering valuable insights for optimizing large-scale LBM frameworks in modern computing environments. diff --git a/content/videos/2024/2022-05-27-bringing-hip-to-one-api.md b/content/videos/2024/2022-05-27-bringing-hip-to-one-api.md new file mode 100644 index 0000000..561d96d --- /dev/null +++ b/content/videos/2024/2022-05-27-bringing-hip-to-one-api.md @@ -0,0 +1,20 @@ +--- +contributor: scott +date: '2024-05-27T00:00:00' +title: 'Bringing HIP to oneAPI' +external_url: 'https://www.intel.com/content/www/us/en/developer/videos/hip-on-aurora.html' +image: ../../../static/images/videos/2022-05-27-bringing-hip-to-one-api.webp +type: presentation +tags: + - hip + - aurora + - hpc + - oneapi +featuring: + - name: Brice Videau + affiliation_at_video_production_time: Argonne National Laboratory +--- + +Derived from the University of Finland’s HIPCL, the new heterogeneous interface for portability (HIP) back end, +CHIP-SPV, can target Intel® GPUs through Level Zero or OpenCL™ Runtimes. Learn how HIP is implemented on Intel GPUs and +why it is important to increase the portability of the application. diff --git a/content/videos/2024/2024-09-12-one-dpl-multiarchitecture-parallel-c-programming-with-one-api-intel-software.md b/content/videos/2024/2024-09-12-one-dpl-multiarchitecture-parallel-c-programming-with-one-api-intel-software.md new file mode 100644 index 0000000..f5990a7 --- /dev/null +++ b/content/videos/2024/2024-09-12-one-dpl-multiarchitecture-parallel-c-programming-with-one-api-intel-software.md @@ -0,0 +1,34 @@ +--- +contributor: scott +date: '2024-09-12T12:11:00' +title: 'oneDPL: Multiarchitecture Parallel C++ Programming with oneAPI | Intel Software' +external_url: 'https://www.youtube.com/watch?v=SsXNONmiSJY' +type: presentation +tags: + - oneapi + - intel + - onedpl + - library +featuring: + - name: Nakita Shiledarbaxi + affiliation_at_video_production_time: Intel Corporation +--- + +The Intel® Developer Zone encourages and supports software developers developing applications for Intel hardware and +software products. The Intel Software YouTube channel is a place to learn tips and tricks, get the latest news, and +watch product demos from Intel and our many partners across multiple fields. You'll find videos covering the topics +listed below, and to learn more, you can follow the links provided! + +oneDPL, a part of the Intel® oneAPI Base Toolkit, is a flexible and optimized library for adding cross-platform +parallelism to your C++ applications for accelerated heterogeneous computing across the latest hardware from diverse +vendors. It lets you leverage the parallel programming capabilities of the SYCL framework, enhancing the performance of +C++ workloads. Its latest feature also grants developers a freedom to choose from the available compute resources for +executing their workloads, based on some pre-defined device selection policies provided by the library. + +With oneDPL, you can achieve remarkable performance improvements and acceleration in complex computations in real-world +applications such as financial modeling, scientific projects, real-time systems, web and cloud-based services, and much +more. + +This video will give you an overview of oneDPL and its component APIs and how C++ developers can benefit from the +multiarchitecture, vendor-independent, accelerated parallel computing capabilities that the library adds to their C++ +applications. diff --git a/static/images/videos/2022-05-27-bringing-hip-to-one-api.webp b/static/images/videos/2022-05-27-bringing-hip-to-one-api.webp new file mode 100644 index 0000000..b3e960a Binary files /dev/null and b/static/images/videos/2022-05-27-bringing-hip-to-one-api.webp differ