Intel® oneAPI Products: Gold Release Coming in December

Senior oneAPI Product Marketing Engineer, Intel Corporation

Over the last year, leading technologists across industry, academia and government have increasingly rallied around oneAPI, a single programming model that aims to unify software development across CPUs, GPUs, FPGAs and other accelerators. September marked the release of the oneAPI v1.0 specification, paving an open road for cross-architecture development through a language, set of library APIs and a low-level hardware interface, as well as a reference stack.

Today, nearly a year after its beta launch, we are excited to celebrate the announcement of the gold release of Intel’s oneAPI implementation, coming in December. This implementation—comprised of a family of toolkits that address data-intensive workloads—marries decades of proven software development tools expertise and advanced hardware capabilities with the heterogeneity of oneAPI. And because these toolkits incorporate familiar languages and standards, they protect your investment with full compatibility with existing code, while offering a quick learning curve.

A big thank you to those who have evaluated this beta implementation over the last year and provided us with your valuable input, whether by kicking the tires in the Intel® DevCloud, downloading one or more of the beta toolkits, or both. Since its initial beta launch, we’ve introduced continuous enhancements.

The gold release of Intel® oneAPI products will include powerful toolkits comprised of industry-leading compilers, performance libraries and analysis tools, and optimized AI frameworks and analytics libraries, including:

  • Intel® oneAPI Base Toolkit: the foundational kit of 20+ tools PLUS the new language Data Parallel C++ and CUDA code migration tool.
  • Toolkits for specialized workloads—HPC, IoT, AI, Data Analytics, Rendering, and more.

Today you can test your code on the latest Intel® hardware in the Intel® DevCloud, including the recently introduced Intel® Iris® Xe MAX GPU as well as Intel® Stratix® 10 FPGAs and the latest generations of Intel® Xeon® CPUs. Intel Iris Xe MAX graphics, the first member of the Xe discrete GPU family, offers a nice proxy development platform for Intel data center Xe GPUs, and Intel® Stratix® 10 FPGAs deliver 2X performance gains over previous-generation, high-performance FPGAs. The DevCloud environment has also received some great new enhancements, like direct access to Jupyter notebooks.

Stay tuned to @IntelDevTools for more news about the gold release in December!

Performance varies by use, configuration, and other factors. Learn more at www.Intel.com/PerformanceIndex.