Match Every Application to Its Optimal Architecture with XPU
Applications are changing how we live and work. Ideally, every application’s performance, latency, and power needs are matched to the optimal hardware architecture (CPU, GPU, FPGA, ASIC), but this creates complexity. To help, Intel not only offers the widest array of hardware architectures in the industry, but a unified programming model to reduce the complexity.
Today’s Applications Demand an XPU Strategy
The “X” in “XPU” stands for any compute architecture that best fits the need of your application. Whether the application is for a PC, a car, or a robotic arm, you want a seamless experience.
Today’s applications require a mix of hardware architectures, with new languages, libraries, and tools to learn. Intel is the first to introduce an XPU strategy, offering the widest array of hardware architectures in the industry and a unified programming model to reduce the complexity.
With XPU, Intel is helping turn the promise of anywhere, anytime service delivery into a reality.
Harness the Power of an XPU Strategy
oneAPI Frees XPU Opportunity
oneAPI is an open, unified programming model that simplifies development across multiple architectures. It brings a productive, performant solution for code reuse across architectures and technology providers, overcoming the challenges of proprietary lock-in.
Every Application Gets the Right Hardware
Workloads have evolved to increasingly take advantage of new architectures. Intel offers a breadth and depth of IP that can be applied in unique ways to solve the compute challenges of today and tomorrow.
No Performance Left Behind
Intel brings deep industry partnership and more types of silicon than anyone to help you determine how your AI, HPC, IoT, network, and graphics applications can get the best performance out of every architecture type across the stack, wherever they need to run.
Getting the Balance Right
The ideal solution matches the application with its optimal hardware architecture, achieving the right performance, latency, and power for your needs. CPUs, GPUs, FGPAs, VPUs, ASICs, and even architectures like neuromorphic computing are part of the Intel XPU strategy.
See the Intel XPU Strategy in Action
Protecting Reefs with Project CORaiL
Project: CORaiL analyzes coral reef resiliency using underwater smart cameras and analytics powered by Intel® Movidius™ VPUs, Intel® FPGAs and CPUs, and the Intel® Distribution of the OpenVINO™ toolkit. The real-time assessment offers an unparalleled view into the coral reef’s health.
oneAPI Fosters Early Success
In one example of the value of oneAPI, a bioinformatics team ported 95 percent of hand-tuned CUDA code in one day. In another, a start-up in France turned a serial algorithm into a massively parallel application running on CPUs and GPUs by changing one line of code.
Innovate Like a Pro with Azure and Intel
Building, testing, and deploying applications isn’t always easy. What if you could use Microsoft Azure AI software tools, choose from the best hardware options Intel provides, and then quickly run your great idea from edge to cloud? Azure and Intel have got you covered.
Additional Resources
Lead the Heterogeneous Computing Future
Learn about Intel’s commitment to delivering a heterogeneous mix of scalar, vector, matrix, and spatial architectures deployed in CPU, GPU, specialized accelerator, and FPGA sockets.
Achieving the XPU Vision
Explore the key milestones Intel announced as part of its multiyear journey to offer an assortment of architectures with a unified software experience.
oneAPI Powers New Computing Era
Learn how the open, unified programming model is built on standards to simplify development and deployment of data-centric workloads.
Illuminating the Future at HPC DevCon
Listen as Intel Chief Architect Raja Koduri explains how Intel is using oneAPI to ease application development and accelerate innovation in the XPU era.