Shulou(Shulou.com)11/24 Report--

CTOnews.com, December 17 (Xinhua)-- Intel today launched the oneAPI 2023 tool (update log) before the release of the fourth-generation Xeon Scalable, Xeon CPU Max and Intel Data Center GPU.

Intel oneAPI 2023 tools can be used to write SYCL code that runs on a range of Intel hardware and other vendors' hardware. Version 2023 supports the upcoming Sapphire Rapids CPU, including the Xeon CPU Max series, and initially supports the AMX directive.

The full name of AMX instruction is Advanced Matrix Extensions, which is a new acceleration engine introduced by Intel on the upcoming fourth-generation Intel Xeon scalable processors, which significantly accelerates tensor computation in deep learning algorithms through instruction set-level support.

It contains a set of two-dimensional registers named TILE and hardware logic called TMUL (Tile Matrix Multiply Unit) to accelerate matrix operations based on TILE registers.

CTOnews.com learned that Intel oneAPI is a cross-architectural programming tool for CPU, GPU, and FPGA that frees itself from the economic and technical burden of proprietary programming models and achieves an efficient and agile way to accelerate computing.

OneAPI is an industry initiative to create an open, standards-based cross-architecture programming model that simplifies development in the face of a large number of data-centric workloads across a variety of architectures (CPU, GPU, FPGA, and other accelerators). It includes cross-architecture language Data Parallel C + + (SYCL based on ISO C + + and Khronos Group), advanced libraries, and community extensions. Many companies, research institutions, and universities support oneAPI. Keep abreast of the latest information on oneapi.com/news.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.