OpenMP* offload constructs are a set of directives for C++ and Fortran introduced in OpenMP 4.0 and further enhanced in later versions that allows developers to offload data and execution to target accelerators such as GPUs. The Intel® MPI Library is a multi-fabric message-passing library that can be used to build and run complex applications on HPC clusters including those with GPU offload resources. Both OpenMP offload support and the Intel® MPI Library are included in the Intel® oneAPI HPC Toolkit, and can be used to target Intel® Graphics including Xe GPUs. In this talk, we’ll discuss how OpenMP offload can be used in conjunction with the Intel® MPI Library, how to manage GPU resources with the Intel® MPI Library, and how to quickly get started using these technologies with the Intel® Devcloud for oneAPI. Karl Qi, oneAPI Technical Evangelist at Intel Corporation