Intel announces Gold release of OneAPI unified programming framework and launches new Intel Server GPU for datacenter applications.

Intel announced the gold release of its oneAPI toolkits alongside its first discrete graphics card for the data center, the Intel Server GPU. These announcements represent two of Intel's growth areas outlined in its "six pillar strategy" focused on expanding the definition of compute and leaning further into software. The oneAPI project is Intel's developer software initiative designed to provide developers with a unified environment that allows them to build applications across different compute architectures in a single tool. Intel leveraged its learnings while developing and operating OpenVINO to enable this experience.

The gold release of the oneAPI toolkits includes a set of tools and libraries for building specific applications and allows developers to reuse code across different hardware like CPUs, GPUs, FPGAs, or ML accelerators. The toolkits utilize new hardware capabilities and instructions like Intel AVX-512 and Intel DL Boost on CPUs. By December, the toolkits will be available for free within the Intel DevCloud, with commercial versions also available and supported by Intel's technical consulting engineers. The oneAPI project has gained support and endorsements from industry giants like Microsoft Azure and TensorFlow.

oneAPI, often described as "magic API," represents Intel's most ambitious software project with tremendous competitive advantage potential. With oneAPI, a developer can use a single API instead of writing on four or five different ones, allowing the developer to choose the best architecture for their problem and use it throughout the development duration. The abstraction layer makes it easier for developers to arbitrate between different compute units while still enabling the ability to drill into underlying code and make optimizations for absolute best performance and efficiency.

The Intel Server GPU is positioned as a low-cost per stream and energy-efficient solution intended not for HPC or AI but instead to power Android cloud gaming and high-density media transcoding and encoding. Built on the Xe-LP microarchitecture, a low-power, energy-efficient architecture, the GPU comes with a 128-bit pipeline and 8GB of dedicated low-power DDR4 memory in a fanless design with all cooling provided by the server. The Intel Server GPU can pair with Intel's Xeon Scalable processors and Intel's open-sourced and licensed software to complete the hardware and software combination.

Intel announced a cloud gaming solution with H3C called the H3C XG310 PCIe card, which utilizes four Intel Server GPUs. Each Intel Server GPU can support up to 20 Android games per GPU and 100 simultaneous users within a 2-card system, with the ability to scale up to 160 concurrent users depending on the server configuration. These announcements reflect Intel's continued commitment to graphics development, a focus that began when Raja Koduri joined the company in 2017 as Chief Architect, leading to the development of the Xe graphics architecture and solutions including the Intel Iris Xe MAX, now available in partner devices with Acer, Asus, and Dell.