Xilinx, Inc., a leader in adaptive computing, charts out a new technology roadmap for future computing, unveiling a wide variety of new chip solitons at its virtual technology conference Xilinx Adapt 2021 that runs between September 7-16.
The new technology releases are designed to address its major applications markets from aerospace and defense to automotive to data center to industrial and healthcare to wires and wireless businesses.
Especially, the world’s largest FPGA chip maker bets big on automotive, health care and industrial IoT as well as live streaming market, as these applications are rapidly emerging as the fastest growing market segments.
A case in point is automotive LiDAR system market where about US$1 billion has been invested into up to 100 companies that are developing high-definition LiDAR sensors.
Healthcare and industrial IoT solution markets are also booming, as computing, 5G mobile communications, AI and new UI and UX technologies are converging to make workplaces both in factories and hospitals smarter and smarter.
Demand for live streaming services stays buoyant, as home-schooling and stay-at-home, and telecommuting are becoming a new norm of our lifestyle, getting people to stay connected to the Internet around the clock to work, study, and socialize at home
Xilinx’s new releases are good fit for those applications, including Alveo accelerator cards for Microsoft database analytics service, a software development kit (SDK) for live video transcoding, Kria robotics stack, a new Vitis library for medical ultrasound, and Zynq RFSoC DFE (digital front-end) solution.
“From the data center and 5G to robotics and space-grade devices, we’re looking forward to kicking off Xilinx Adapt 2021 to educate customers and developers on our newest products and tools across our diversified end markets. We’re excited to share how adaptive computing is helping customers realize great innovations and bring them to market faster than ever,” said Victor Peng, Xilinx president and CEO.
Aerospace & Defense
Xilinx is launching two new versions of its Versal ACAP products with the addition of the space-grade XQR Versal portfolio and the defense-grade XQ ruggedized Versal portfolio. The XQR Versal space-grade portfolio includes
Versal AI core and Versal Edge series devices, offering ruggedized, organic ball grid array packaging with extended qualification and burn-in supporting Mil-Std-883 Class B-grade flow and full radiation tolerance.
These XQR devices are designed to process on-board data processing and machine learning (ML) inference in satellite and space 2.0 applications, an important accelerated advancementbuilding upon the successful adoption of XQRKU060 launch in 2020.
The XQR Versal ACAP is a highly capable and adaptable device with support for full reconfiguration while in orbit. The space-grade XQR Versal AI Core device is targeted for availability in late Q2 2022.
The XQ ruggedized devices are heterogeneous multicore compute platforms equipped with adaptable industry-leading AI/ML capabilities to withstand the evolving algorithms and demands of harsh environmental requirements for aerospace and defense applications.
Building on Xilinx’s 30 years of heritage and continuous development in the industry, the latest portfolio includes the XQ AI Core, XQ AI Edge, XQ Prime and XQ Premium families.
The devices in the portfolio feature ruggedized packaging qualified to Mil-Std-883 Group D standards, tin-lead (Sn/Pb) content with full mitigation of tin-whiskering, and optional M-temp (-55°C to +125°C) support. The XQ ruggedized Versal portfolio is targeted for initial device availability in Q1 2022.
Xilinx FPGAs and adaptive SoCs bring tremendous value to LiDAR sensor architectures because of the capability to support advanced signal processing, point cloud pre-processing, and point cloud ML acceleration. Xilinx is the leading silicon supplier in the LiDAR market.
As automotive market adoption of LiDAR sensor is accelerating. Xilinx FPGAs and adaptive SoCs are being deployed in these systems because they provide extreme flexibility and adaptability, deliver high performance at low latency, and help meet fast time-to-market timeframes.
Xilinx data center platforms and plug-in board solutions continue to drive strong demand across hyperscale cloud customers and the fintech market.
Microsoft has announced that Azure is now utilizing their NP-VM FPGA-as-a-Service infrastructure, powered by Xilinx Alveoaccelerators, to dramatically improve the performance of Apache Spark on the Azure Synapse analytics platform.
Alveo acceleration delivers up to a 40x improvement in data parsing CSV and JSON file formats, from 15MB/second with a CPU core to 7.7GB/second with Alveo.
This eliminates a common bottleneck in Spark performance, and with 90 percent of Synapse Spark workloads consisting of CSV/JSON/Parquet files, Azure is delivering massively enhanced performance to their growing Synapse customer base.
Live Video Transcoding SDK
Xilinx has released the Xilinx Video Transcoding SDK, making it easier and faster than ever before for streamers to build new or migrate existing services to take advantage of Xilinx acceleration.
The Video Transcoding SDK is a complete solution, from runtime drivers to server resource management, and provides support for FFmpeg and C/C++ API, ensuring that customers can start developing streaming solutions instantly.
Paired with the Xilinx Alveo U30 media accelerator card, it delivers industry leading video transcoding performance with dramatically improved channel density, lower power and total cost of ownership compared to alternative products.
Industrial & Healthcare
Xilinx addresses the growing needs of scalable healthcare and industrial IoT platforms with heterogenous processing, I/O flexibility, hardware-based deterministic control, and comprehensive solutions.
Xilinx is expanding the scope of Kria System-on-Modules beyond vision AI and is now equipping roboticists with an easy-to-use, adaptive and hardware-accelerated set of functions and tools called the Kria robotics stack (KRS).
Built around the Robot Operating System (ROS), a framework for robot application development, KRS is comprised of an integrated set of utilities helping speed up ROS 2 and Gazebo, empowering faster, secure, and more complete robot behavior.
KRS ships with an expanding set of reference designs and over time, robotic-focused accelerated applications built with KRS will be made available via the Xilinx App Store.
KRS helps leverage hardware acceleration into Xilinx devices with high-performance and low latency, in a similar fashion to how Vitis AI helps AI developers implement machine learning inference models into Xilinx devices.
Vitis Libraries for Medical Ultrasound
For the first time, Xilinx has expanded its Vitis unified software platform to include libraries for ultra-fast imaging optimized and targeted for the innovative Versal AI engines.
These libraries target medical ultrasound applications, producing breakthrough performance resulting in greater than 1,000 frames-per-second in a 64-channel ultrasound machine with 200 lines of resolution using just a single Versal device.
Start Volume Shipments of Zynq RF SoC
From edge to cloud, Xilinx FPGAs and adaptive SoCs continue to be at the heart of wired and wireless networking equipment driving global 5G deployments.
Xilinx has begun shipping its new Zynq RFSoC DFE in volume to multiple radio customers worldwide including a top wireless system vendor.
Zynq RFSoC DFE is a new class of radio platforms that integrates hardened digital front-end application-specific blocks for 5G NR performance and power savings while also offering integrated programmable adaptive logic to enable a futureproof solution for evolving 5G 3GPP standards and O-RAN radio architectures.
Zynq RFSoC DFE is gaining momentum in the market because it optimizes power efficiency, performance and cost while still delivering the flexibility and time-to-market benefits of a programmable and adaptive SoC.