Adaptive system-on-modules speed edge AI application deployments
The Kria adaptive SOMs, says the company, are a new method of bringing adaptive computing to AI and software developers. The first product available in the Kria SOM portfolio, the Kria K26 SOM, specifically targets vision AI applications in smart cities and smart factories.
The company says its SOM roadmap includes a full range of products, from cost-optimized SOMs for size and cost-constrained applications to higher performance modules that will offer developers more real-time compute capability per watt.
“Xilinx’s entrance into the burgeoning SOM market builds on our evolution beyond the chip-level business that began with our Alveo boards for the data center and continues with the introduction of complete board-level solutions for embedded systems,” says Kirk Saban, vice president, Product and Platform Marketing at Xilinx. “The Kria SOM portfolio expands our market reach into more edge applications and will make the power of adaptable hardware accessible to millions of software and AI developers.”
The Kria SOMs, says the company, harness the power, performance and flexibility advantages of its adaptable hardware, delivered as production-deployable, adaptive modules. Kria SOMs enable rapid deployment by providing an end-to-end board-level solution with a pre-built software stack. By allowing developers to start at a more evolved point in the design cycle compared to chip-down design, Kria SOMs are offered as being able to reduce time-to-deployment by up to nine months.
The Kria K26 SOM is built on top of the Zynq UltraScale+ MPSoC architecture, which features a quad-core Arm Cortex-A53 processor, more than 250 thousand logic cells, and an H.264/265 video codec. The SOM also features 4 GB of DDR4 memory and 245 IOs, which allow it to adapt to virtually any sensor or interface.
With 1.4 tera-ops of AI compute, says the company, the Kria K26 SOM enables developers to create vision AI applications offering more than 3x higher performance at lower latency and power compared to GPU-based SOMs, critical for smart vision applications like security, traffic and city cameras, retail analytics, machine vision, and vision guided robotics.
The Kria SOM portfolio couples the hardware and software platform with production-ready vision accelerated applications. These turnkey applications eliminate all the FPGA hardware design work and only require software developers to integrate their custom AI models, application code, and optionally modify the vision pipeline – using familiar design environments, such as TensorFlow, Pytorch or Café frameworks, as well as C, C++, OpenCL, and Python programming languages – enabled by the Vitis unified software development platform and libraries.
With this new accelerated-application paradigm for software-based design, says the company, it is announcing the first embedded app store for edge applications. Building out beyond the Alveo catalog of apps for data center, the Xilinx App Store now offers users a wide selection of apps for Kria SOMs from Xilinx and its ecosystem partners. Xilinx offerings are open source accelerated applications, provided at no charge, and range from smart camera tracking and face detection to natural language processing with smart vision.
Kria SOMs also enable customization and optimization for embedded developers with support for standard Yocto-based PetaLinux and, for the first time ever, the company says it is announcing a coming collaboration with Canonical to provide Ubuntu Linux support, the highly popular Linux distribution used by AI developers. This offers widespread familiarity with AI developers and interoperability with existing applications. Customers can develop in either environment and take either approach to production. Both environments will come pre-built with a software infrastructure and helpful utilities.
The Kria KV260 Vision AI Starter Kit is offered as an affordable and easy-to-use development platform for designing vision applications, right out of the box. The kit is purpose-built to support accelerated vision applications available in the Xilinx App Store and is offered as an excellent way to get up and running in less than an hour with no knowledge of FPGAs or FPGA tools.
Part of the Kria development experience, says the company, is a self-enabled path for exploration, design, and ultimately production deployment through a vast set of online resource. Hobbyists, makers, and commercial developers alike can accelerate through each phase of the design cycle with tutorial videos, training courses, and a vast ecosystem of providers offering accelerated applications, design services, and more.
The Kria KV260 Vision AI Starter Kit is priced at $199. When ready to move to deployment, users can seamlessly transition to the Kria K26 production SOM, including commercial and industrial variants priced at $250 or $350, respectively.
Kria K26 SOMs and the KV260 Vision AI Starter Kit can be ordered now. The KV260 Vision Starter Kit is available immediately, with the commercial-grade Kria K26 SOM shipping in May of 2021 and the industrial-grade K26 SOM shipping this summer. Ubuntu Linux on Kria K26 SOMs is expected to be available in July of 2021.