The 2022 Intel® oneAPI toolkits deliver productivity and performance through a complete set of advanced tools, including libraries, compilers, analysers, pre-optimised frameworks and debuggers. There are over 900 new and enhanced features that have been added over the past year, which strengthen every tool in the foundational and domain-specific toolkits. Intel has made them available to download or use in the Intel® DevCloud for free.
The new capabilities include the world’s first unified compiler, implementing C++, data-parallel Python for CPUs and GPUs, SYCL and Fortran, advanced accelerator performance modelling and tuning, and performance acceleration for AI and ray-tracing visualisation workloads. The Intel® DPC++ Compatibility Tool was improved to automatically migrate 90% to 95% of CUDA code to SYCL/DPC++.
The oneAPI cross-architecture programming model provides tools that aim to improve the velocity and productivity of code development when building cross-architecture applications.
Deep learning framework performance is accelerated ten times over earlier versions with the latest Intel® Optimisation for TensorFlow and Intel® Optimization for PyTorch. The Intel® Extension for Scikit-learn speeds up ML algorithms over 100 times on Intel CPUs over the stock open-source version. The company has introduced Intel® Neural Compressor, which can achieve increased inference performance through post-training optimisation techniques across multiple deep learning frameworks.
“The rich set of oneAPI technologies conforms to key industry standards, with deep technical innovations that enable applications developers to obtain the best possible run-time performance from the cloud to the edge. Multi-language support and cross-architecture performance acceleration are ready today in our oneAPI 2022 release to further enable programmer productivity on Intel platforms,” said Greg Lavender, Intel CTO, SVP and general manager of the Software and Advanced Technology Group.
The oneAPI toolkits are optimised to enable advanced features of the latest and upcoming new hardware, including 12th Gen Intel® Core™ processors with AVX-VNNI, NextGen Intel® Xeon® Scalable processors, codenamed Sapphire Rapids with Intel® Advanced Matrix Extension (Intel® AMX), and upcoming Xe client and data centre GPUs.
The pace of innovation is accelerating with cross-architecture computing driven forward with oneAPI across heterogeneous fabrics of CPUs, GPUs, FPGAs and other accelerators — all of which require software to bind these elements into a complete solution. Intel’s oneAPI allows developers to have the freedom to choose the best hardware for a solution without the proprietary programming models’ economic and technical burdens.
