Intel Corporation TodayAnnounced a number of important scientific and technological progress，This is another milestone in Intel's years-long commitment to creating cross-architecture solutions through a unified software experience。in，Intel® oneAPI Gold toolkit will be officially delivered in December this year；Intel Software Stack Introduces New Features，As part of the company's hardware-software co-design approach。Simultaneously，Intel Officially Announces Its First Data Center Discrete Graphics Card。The server GPU is based on the Xe-LP microarchitecture，designed for high density、Designed for low-latency Android cloud gaming and streaming services。
Intel Senior Vice President、Chief Architect and Architecture、Raja Koduri, General Manager, Graphics and Software Division, said："Today is an important moment for Intel's oneAPI and XPU grand plans。With the release of oneAPI Gold version，Developer programming experience will be richer，oneAPI not only has CPU programming libraries and tools familiar to developers，Also includes programming libraries and tools for hybrid vector-matrix-space architectures。Simultaneously，We also introduced our first data center GPU based on the Xe-LP microarchitecture，To meet the needs of the rapidly growing cloud gaming and streaming market。”
importance：As the world enters the age of billions of smart devices，Data grows exponentially，Need to shift focus from individual CPUs to cross-CPUs、GPU、Hybrid architecture of FPGA and other accelerators，Intel calls it the "XPU" vision。Introducing Intel® Server GPUs is the latest step in Intel's expanding product portfolio in the XPU era。
This era of computing also requires a comprehensive software stack。passIntel oneAPI Toolkit，Developers can use a generic、Open, industry-standards-based programming model for accessing Intel XPUs。This not only unlocks the performance potential of the underlying hardware，At the same time, it can reduce software development and maintenance costs，and in terms of deploying accelerated computing，The Intel® oneAPI Toolkit is better than a dedicated、Vendor-specific solutions are less risky。
Intel launches oneAPI Gold toolkit：Intel oneAPI industry planFirst presented at the SuperComputing 2019 conference，This is Intel's unification、A Vision Proposed by a Simplified Cross-Architecture Programming Model：Capable of delivering uncompromising performance，Not bound by single-vendor-specific code constraints，And can realize the integration of the original code。With oneAPI，Developers can choose the best accelerator architecture for the specific problem they want to solve，without having to rewrite the software again for new architectures and platforms。
Intel oneAPI toolkit takes full advantage of advanced hardware performance and instruction，Such as Intel® AVX-512 (Advanced Vector Extensions) and Intel® Deep Learning Boost (Intel® DL Boost) for CPU，and XPU-exclusive features。oneAPI Toolkit is based on the long-proven Intel Developer Tools，Provide developers with familiar programming languages and standards，while maintaining full continuity with existing code。
today，Intel announced，Intel oneAPI Gold toolkit will be free on-premises and on Intel DevCloud in December，A commercial version will also be available with global support from Intel Technical Consulting Engineers。Intel will also migrate Intel® Parallel Studio XE and Intel® System Studio tool suites into oneAPI products。
also，Intel DevCloud platform allows developers to test code and workloads on various Intel architectures，New Intel® Iris® Xe GPU Hardware Now Added。Intel Iris Xe MAX graphics now also available for public access；Simultaneously，Intel Xe-HP is open to specific developers。
oneAPI is supported by the industry，Recently, Microsoft Azure and Google's TensorFlow have announced support for oneAPI；Numerous leading research institutions、companies and universitiessupport oneAPI。
besides，The Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign announced today，A new oneAPI Center of Excellence (CoE) will be established。They are using the oneAPI programming model to extend life science application NAMD to other computing environments。NAMD enables analogy to large biomolecular systems，is helping to solve global challenges like COVID-19。This Center of Excellence will collaborate with the Stockholm University (SERC) Center of Excellence for Research on GROMACS，Together with the Heidelberg University (URZ) Center of Excellence，Joint research on how to provide oneAPI support for GPUs from other manufacturers。
About Intel's New Server GPUs：Passed the first discrete display product for the data center，Intel Extends Rich Platform-Level Innovations to Enhance Cloud Gaming and Media Experiences。Take advantage of the combination of Intel® Xeon® Scalable processors and new Intel server GPUs，Plus open source and licensed Intel software components，i.e. lower total cost of ownership (TCO)，Provides high-density media transcoding for Android cloud games and OTT live video streaming、Low Latency Solution。
Intel server GPU adopts Intel's most energy-efficient graphics architecture - Intel Xe-LP microarchitecture，Has low power consumption、Standalone System-on-Chip Design，And equipped with 128-bit pipeline and 8GB dedicated onboard low-power DDR4 memory。
By combining the power of Intel Server GPUs and Intel® Xeon® Scalable processors，Service providers can，Independently expand graphics card capacity，to support more streams and subscribers per system，while achieving a low total cost of ownership (TCO)。Through H3C XG310 X16 PCIe3.0 GPU expansion card - in 3/4 length、Package four Intel server GPU chips in a full-height form factor，It can support more than 100 concurrent users of Android cloud games in a typical dual card system。This number is scalable up to 160 concurrent users，Actual number depends on specific game and server configuration。
Developers can take advantage of the common API in the current Media SDK，This API will also be migrated to the oneAPI video processing library next year。current，Intel is working with including Gamestream、Cooperate with many software and service partners including Tencent and Ubitus，Bringing Intel Server GPUs to Market Together。
Fang Liang, assistant general manager of Tencent Xianyou Cloud Games, said："Intel is a very important partner for our Android cloud gaming solutions。Intel Xeon Scalable processors and Intel Server GPUs，Created a high density、low latency、Low power consumption、Low TCO Solutions，Allows us to generate over 100 game instances per dual card server，such as "Glory of the King"、"Legend Showdown"。”
Intel server GPUs based on the Xe-LP microarchitecture are now shipping。Together with the recently launched Intel® Iris® Xe MAX，The GPU will further enhance the visual computing experience for global users with the continuous in-depth development of Intel Xe architecture products and software projects。
Intel® Graphics Software Update：One of Intel's core strategies for scaling GPUs from entry-level graphics cards to high-performance computing (HPC)，is to implement the same code base。In order to achieve this goal，Intel's software stack now supports multiple generations of graphics cards，including the most recent11th Generation Intel® Core™ Mobile Processor Integrated Iris Xe GraphicsandIntel Iris Xe MAX discrete graphics card。Extend codebase to support more ubiquitous data center products on Linux，is the next critical step in the scalable Xe architecture strategy。Intel optimizes Linux drivers，Focus on code reuse between operating systems，And further focus on Linux 3D performance，Three fully validated and integrated distribution stacks are now available。
Intel announced today，Intel creates Project Flipfast to improve gaming experience on Linux OS。The Flipfast stack allows end users to run graphics applications on virtual machines，While maintaining native GPU performance and host integration integrity through zero-copy sharing between the virtual machine and the host。Flipfast stack driver boosts gaming performance，The technology will be directly usable in data center game streaming applications。
Intel also announced today，Intel® Implicit SPMD Program Compiler (ISPC) will run on the underlying hardware interface oneAPI Level Zero (Level Zero)。oneAPI level zero is the entire hardware abstraction layer，Tailored for devices in the oneAPI platform，provides the underlying、direct-to-hardware interface。ISPC supported by oneAPI is a variant of the C programming language，Single program、Multiple Data Programming，Used to accelerate the Intel® Osray Ray Tracing Engine on Intel CPUs。Intel adds Xe support to ISPC，to seamlessly accelerate Intel oneAPI Rendering Toolkit components (such as Osray)。
next step：11March 12-13，existoneAPI Developer Summit held onlinesuperior，innovator、Researchers and developers will showcase 40 collaborations and projects using oneAPI。Topic involves repurposing tests from COVID-19 drugs，to crop yield forecasting and more。At SuperComputing 2020 starting this week，Intel joins hands with industry leaders and research institutions to deliver keynote presentations、Science and Technology Conference、fireside chat、Demos and other events，Focus on oneAPI applications and Intel oneAPI tools。Full information on Intel's activities at SuperComputing 2020，Please visit intel.com。
Intel oneAPI and graphics software stack updates，and the introduction of Intel server GPUs，Marks a milestone step for Intel towards the era of XPU architecture。Based on Intel's six technology pillars innovation and heterogeneous architecture，And through a unified and scalable software abstraction layer based on open standards in oneAPI，These advances provide a solid foundation for a better experience。