Intel oneAPI Center of Excellence
It is highly desirable that scientific codes achieve high-performance on heterogeneous systems. These computers consist of a combination of CPU and GPU architectures, and so applications will need to run across a range of different types of processors. Applications must also run efficiently and achieve close to the best performance possible given the capabilities of the processors on which they execute, i.e., the code must be performance portable. Achieving this is challenging, and so our Intel oneAPI Center of Excellence will develop and share best practices on how to make applications performance portable at Exascale, in particular for applications written using standard parallel programming languages including SYCL. The project will make available code samples and other resources to help applications developers write performance portable codes.