Hybrid Parallel Programming for HPC Clusters with MPI and DPC++

Modern high-performance computing (HPC) clusters can include various nodes that contain hardware accelerators such as GPUs and FPGAs. To take full advantage of internode, intranode, and accelerator-device-level parallelism, hybrid programming is required.

This webinar discusses how to do exactly that by using Data Parallel C++ (DPC++) with the message passing interface (MPI), which are supported in the Intel® oneAPI Base Toolkit and Intel® HPC Toolkit, respectively.

In this session, software specialists Karl Qi and Loc Nguyen cover the landscape, clarifying how MPI and DPC++, can be effectively used together to communicate between nodes and accelerate computation on a single node using available accelerators. Topics include:

• A brief overview of MPI
• Use the Intel® MPI Library with DPC++
• Compile and deploy applications on Linux* and Windows*
• Target DPC++ kernels for CPUs and GPUs
• Use MPI and DPC++ in the Intel® Developer Cloud

• Download the Intel oneAPI Base Toolkit that includes nearly 20 development tools and libraries for creating cross-architecture applications.
• Download the Intel HPC Toolkit that includes eight development tools and libraries for HPC computing, including the Intel MPI Library.
• Sign up for an Intel Developer Cloud account—a free development sandbox with access to the latest Intel hardware and oneAPI software.

• Learn more about the DPC++ language and the Intel® oneAPI DPC++ Library
• Subscribe to Code Together, an interview series that explores the challenges at the forefront of cross-architecture development. Each biweekly episode features industry VIPs who are blazing new trails through today’s data-centric world. Available wherever you get your podcasts.

Karl Qi
oneAPI technical evangelist, Intel Corporation

Karl focuses on enabling HPC and AI customers to create the optimal solution for their needs using the Intel® toolkits. He has a particular interest in software that can take advantage of the capabilities of heterogeneous parallel computing environments. Karl has a bachelor’s degree in electrical engineering from Cornell University.

Loc Nguyen
Software engineer, Intel Corporation

Loc Nguyen received an MBA from University of Dallas, a master’s degree in electrical engineering from McGill University, and a bachelor’s degree in electrical engineering from École Polytechnique de Montréal. He is a software engineer with Intel, with interests spanning across machine learning, computer networking, parallel computing, and computer graphics.