oneAPI is a cross-industry, open, standards-based unified programming model that delivers a common developer experience across accelerator architectures—for faster application performance, more productivity, and greater innovation. The oneAPI industry initiative encourages collaboration on the oneAPI specification and compatible oneAPI implementations across the ecosystem.
The oneAPI Specification
The oneAPI specification extends existing developer programming models to enable a diverse set of hardware through language, a set of library APIs, and a low level hardware interface to support cross-architecture programming. To promote compatibility and enable developer productivity and innovation, the oneAPI specification builds upon industry standards and provides an open, cross-platform developer stack.
At the core of the oneAPI specification is DPC++, an open, cross-architecture language built upon the ISO C++ and Khronos SYCL standards. DPC++ extends these standards and provides explicit parallel constructs and offload interfaces to support a broad range of computing architectures and processors, including CPUs and accelerator architectures. Other languages and programming models can be supported on the oneAPI platform via the Accelerator Interface.
oneAPI provides libraries for compute and data intensive domains. They include deep learning, scientific computing, video analytics, and media processing.
The Hardware Abstraction Layer
The low-level hardware interface defines a set of capabilities and services that allow a language runtime to utilize a hardware accelerator.
Optimized Middleware & Frameworks
oneAPI Industry Specification
- Provide feedback on the oneAPI specification.
- Create implementations for new compute architectures.
- Suggest future directions for oneAPI and include oneAPI in upcoming research.
“With the growth of AI, machine learning, and data-centric applications, the industry needs a programming model that allows developers to take advantage of rapid innovation in processor architectures. TensorFlow supports the oneAPI industry initiative and its standards-based open specification. oneAPI complements TensorFlow’s modular design and provides increased choice of hardware vendor and processor architecture, and faster support of next-generation accelerators. TensorFlow uses oneAPI today on Xeon processors and we look forward to using oneAPI to run on future Intel architectures.”
“The future of advanced computing requires heterogeneous hardware to maximize the computing power needed for exascale-class workloads. The oneAPI industry initiative Intel is spearheading will ensure that programming across diverse compute architectures is greatly simplified.”
— Rick Stevens, associate laboratory director, Computing, Environment, and Life Sciences, Argonne National Laboratory and professor of computer science, University of Chicago
“The University of Cambridge Research Computing Service supports world-leading research in science, technology, and medicine where we see the need for large-scale data-driven discovery platforms with converged capability for simulation, data analytics, and AI. If these computation discovery platforms are to advance in step with the ambitions of the science projects that drive them, we will need ever more energy-efficient systems that rely increasingly on heterogeneous compute elements to help solve the world’s biggest scientific challenges. The promise of oneAPI to deliver a single programming environment across multiple compute architectures is a vital tool to unlock the promise of heterogeneous computing. Here science communities can leverage investments in code development across multihardware platforms helping advance performance gains from different hardware targets and also making future hardware targets more accessible.”
— Paul Calleja, director of Research Computing Service, Cambridge University
“One of the major problems facing developers today is disparate programming environments and little code re-use opportunities across different types of hardware. A single programming environment that could render code without sacrificing performance across multiple hardware types is a difficult, yet important challenge. Intel oneAPI appears to be a significant step in the right direction, promising code portability without compromising the ability to tune performance for CPUs and accelerators, and making hardware transitions considerably less risky and error prone. We are therefore considering oneAPI for high energy physics (HEP) workloads.”
— Federico Carminati, chief innovation officer, CERN openlab
“A unified programming model like Intel’s oneAPI can go a long way in accelerating the hardware and software ecosystems. We especially welcome how Intel is driving this as an open initiative and look forward to working closely with them to increase adoption in a collaborative manner.”
— Dani Pinkovich, algorithm group manager, GE Healthcare
“We are approaching the exascale era, which is defined by data growth and converged HPC, AI, and analytics workloads to unlock greater value, discovery, and accelerate innovation. Customers are requiring development tools to address varying and data-intensive workloads running on complex, diverse architectures. By continuing our long-standing partnership with Intel and supporting oneAPI, our customers are gaining tools to optimize applications and speed market delivery through unified programming and simplified software development across a range of HPE technologies, including compute solutions such as CPUs, GPUs, FPGAs, and AI accelerators.”
— Peter Ungaro, senior vice president and general manager, HPC and AI, HPE
“We eagerly look forward to the oneAPI initiative, and the effort to build a platform-inclusive programming approach that will help domain experts improve utilization across a variety of available hardware, as well as other emerging architectures in the future.”
— Dr. Manish Agarwal, Computer Services Centre, IIT Delhi
“SUSE* is the most customer-centric open source company in the world, and that’s why we look forward to the oneAPI initiative. Its goal of delivering a single multi-architecture programming environment based on an open specification and industry standards will benefit enterprise users globally. SUSE anticipates ongoing collaboration with Intel around this initiative as well as others, in order to help ease software development and deployment for our joint customer base.”
— Vojtech Pavlik, vice president, SUSE Labs
“As the leader of the GROMACS development team (one of the most widely used [high-performance computing] HPC codes in the world for molecular simulations) at Stockholm University and KTH, I’m delighted to strongly endorse Data Parallel C++, a cross-architecture development language, based on the first specification and road map. We think it’s outstanding to finally see a strong C++ enabled accelerator development environment, as well as the strong commitment to an open project and industry specifications. Furthermore, the aim of contributing parallel and accelerator-enabled functionality as part of future C++ standards is just as important, since this will be a revolution for portability.”
— Erik Lindahl, biophysics professor, GROMACS development team, Stockholm University and KTH