Intel have released the latest version of oneAPI, 2022.1.

The new toolkits have over 900 new features, capabilities, and improvements, helping your customers deliver high-performance, cross-architecture applications, and solutions.

Key highlights of the update are:

Cross-architecture Programming

  • Intel® oneAPI DPC++/C++ Compiler becomes the world’s first LLVM-based, unified compiler; it implements C++, SYCL, and Fortran to improve programming productivity for CPUs and GPUs.
  • Data-parallel Python technologies enable accelerated computing on CPUs and GPUs.
  • Intel® DPC++ Compatibility Tool now automatically migrates 90%-95% of CUDA cde to SYCL/DPC++.1

Advanced Development Productivity

  • Intel® VTune™ Profiler flame graph display improves the ability to visualise performance hotspots.
  • Intel® Advisor’s accelerator performance modeling estimates the benefits of offloading to a GPU before making code changes.
  • New Intel® oneAPI Rendering Toolkit features including cone telemetry, auxiliary feature denoising, and FP16 support provide more robust shapes and shorten rendering times. Additional support for upcoming Intel® Xe GPU plus real-time denoising further improve final-frame, production-quality rendering.
  • Expanded development environment support include deeper Microsoft VSCode integration, VS 2022 support, and WSL 2 for Linux development on Windows.

AI Training & Inference Performance Optimizations

  • Intel® Optimisation for TensorFlow is now 10x faster compared to the stock version on 3rd Gen Intel® Xeon® Scalable processors.2
  • New Intel® Extension for Scikit-learn speeds machine learning algorithms over 100x on Intel CPUs compared to the stock open-source version.3
  • Intel® Neural Compressor is introduced to increase inference performance through post-training optimisation techniques across multiple deep-learning frameworks.

Hardware Support

  • All toolkits are optimised to enable advanced features of the latest (and upcoming) hardware, including 12th Gen Intel® Core™ processors (codenamed Alder Lake) with AVX-VNNI, Sapphire Rapids with Intel® AMX, and upcoming Xe client and datacenter GPUs.

For more information on the latest update, get in touch with our Intel Business Development Manager, Nicolas Rannou. | +44 (0) 1364 655 219