1. Home
  2. 사업현황
  3. 참여 프로젝트
  4. oneAPI

Intel oneAPI

oneAPI: Smart Path to Accelerated Computing
software.intel.com/oneapi

Problem Statement

Problems with Heterogeneous Architecture Programming 

Many data-centric workloads run best when deployed across a mix of heterogeneous architectures (CPU, GPU, FPGA, other accelerators).
However, different architectures have typically required unique languages, tools and libraries adding complexity for developers and limiting code reuse.
This makes it difficult to take advantage of cross-architecture solutions and inefficient to optimize application performance.

oneAPI

Simplify Cross-Architecture Development

Photo

Intel created a product implementation of oneAPI with the Intel® oneAPI Toolkits , which help developers efficiently build, analyze, and optimize high-performance, cross-architecture applications for Intel CPUs, GPUs and FPGAs.

• The Intel® oneAPI Base Toolkit is a core set of tools including best-in-class compilers, powerful performance libraries, and analysis and debug tools.

• Segment-specific Toolkitsaid developers in optimizing applications and solutions for HPC, AI, IoT, and rendering. For the latest news and updates visit Intel oneAPI News.

Note: The tools that comprised Intel® Parallel Studio XE and Intel® System Studio are now integrated into Intel’s oneAPI Products.
The Intel oneAPI toolkits are upward compatible supersets of the features in the previous studio products, so the studio products will no longer be offered to new customers in 2021.

oneAPI is an industry initiative creating an open, standards-based, cross-architecture programming model to simplify development for a wide range of data-centric workloads across a variety of architectures (CPU, GPU, FPGA, and other accelerators).
It includes a cross-architecture language Data Parallel C++ (based on ISO C++ and Khronos Group’s SYCL), advanced libraries, and community extensions.

Many companies, research organizations, and universities support oneAPI.

Stay updated at oneapi.com/news.

Photo

Primary Audience & Users

Who Needs It

• Developers wanting to maximize compute by building cross-architecture applications and solutions that take advantage of a variety of Intel systems and accelerators.

• Developers building data-centric applications to run on all types of platforms (edge-to-cloud) and architectures (Intel CPUs, GPUs, FPGAs) across a range of domains including HPC, AI, IoT, and visual computing.

• Developers with existing CUDA code who want to take advantage of other architectures via oneAPI and by porting their code to DPC++ code.

Benefits

What You Get

oneAPI and Intel oneAPI Toolkits together bring productive and performant heterogenous programming to developers.

Future-Ready Programming Model Provides Freedom of Choice

• An open alternative to single-vendor/proprietary lock-in enables easy architecture retargeting.

• Apply your skills to the next innovation, not rewriting software for the next hardware platform.

Realize All the Hardware Value

• Expose and exploit all the cutting-edge features of the latest hardware. − e.g. 3rd gen Intel® Xeon® Scalable processors enable Intel® Advanced Vector Extensions 512 (Intel® AVX-512) including Intel® Deep Learning Boost and bfloat16 acceleration for AI acceleration.

• Performance for Accelerated Architectures⎯Take full advantage of accelerated compute by maximizing performance across Intel CPUs, GPUs and FPGAs.

Confidently Develop Performant Code Quickly and Correctly

• Fast efficient development with a complete set of cross-architecture libraries and advanced tools.

• Shorter learning curve with familiar languages and standards; is interoperable with existing HPC programming standards including Fortran, C/C++, OpenMP and MPI, as well as Python with a rich set of optimized Python libraries.

• Easy integration with legacy code including migration of CUDA code to DPC++.

Product Brief

For more information, click the button below or the top menu.

Product Brief