Xilinx announced Zynq® UltraScale+™ RFSoC Gen 2 and Gen 3 last week, which established Xilinx as an uncontested provider of monolithic, direct-RF, single-chip adaptable radio platforms, now with greater RF performance and scalability.

Zynq UltraScale+ RFSoC devices leverage the highly successful 16nm Zynq UltraScale+ MPSoC platform and add high-performance RF sampling data converters and reconfigurable Soft-Decision Forward Error Correction (SD-FEC). Zynq UltraScale+ RFSoC Gen 2 device, sampling now with production scheduled for June 2019, pushes the analog input /output frequency to 5GHz to support the n79 5G NR bands in Asia and is available in a 16T/16R configuration. Zynq UltraScale+ RFSoC Gen 3 devices extend the input / output frequency to 6GHz covering the full sub-6GHz band and increase ADC sample rate to 5GSPS from 4GSPS and resolution to 14-bit from 12-bit. 

DesignCon 2019 has come and gone, and with it came a flurry of new interconnects targeting a variety of forward-looking serial technology. Xilinx partnered with four connector vendors to present six demonstrations showing what architects can look forward to using in their designs. From the near term 32G PCIe® style and emerging 58Gb/s PAM4 long reach interconnects to tomorrow’s 112G PAM4 topologies, Xilinx had SerDes technologies on display to fit everyone’s next generation wish list. In the following two weeks, we are going to release a series of Xilinx PAM4 technology demos from DesignCon. Stay tuned!

The most critical AI video analytics applications are the ones that protect human health, preserve human life, and secure property, and complex applications like that need deterministic low latency performance. Not only that, but their complexity can be extremely hardware intensive, which can challenge organizational budgets.

Massive growth in data traffic driven by cloud services and the rollout of 5G is creating challenges to keep up with burgeoning bandwidth requirements within networks. Line card port densities in routers and switches, optical standard evolution, and optical network bandwidth updates are major constraints to cost effectively keep pace with bandwidth needs. The transition to 58G transceivers is an important step to deliver 400G+data rates on the same existing footprint.

What do you get when you combine the world's most popular microprocessor with the biggest ecosystem and the world's most powerful programmable logic devices? Unparalleled flexibility, performance, scalability, and real-time determinism. This combination allows increased protocol support, I/O expansion, hardware acceleration, and future-proofed designs.

Financial trading instructions have gone through significant changes over the past two decades, with the evolution of electronic trading, algorithms, smart order routers, the rise of HFT, and ever more complex strategies in search of alpha. 

Innovation is recognized when it disrupts a market. Network Interface Cards (NICs) have existed since the 1980s, and around the turn of the century, various companies began innovating in this area. Their initial focus was on High-Performance Computing (HPC). While businesses ran on Ethernet, the most powerful HPC clusters in the world were created using Myrinet or Infiniband. The NICs for these networks were architected to bypass the operating system (OS) kernel and communicate directly with HPC applications delivering substantially greater performance. To make this happen, these NICs applied extensive computational resources at the edge of the network, enabling them to offload common communications tasks from the host CPU.

The volume of data that healthcare providers collect is increasing and overwhelming the approaches currently being used to analyze it. At the same time, the ability to connect across tables and business entities and identify hidden relationships and patterns offers tantalizing breakthroughs in patient care and outcome improvements as well as efficiencies and cost savings for the providers.

To improve the performance of compute infrastructure, and to keep up with the expanding requirements of data analytics and AI, many enterprises are looking to hardware acceleration as an integral solution.

And the result of that? ... Making Transcoding Easy to Develop and Deploy ... The Xilinx U30 SDK enables Straight Real-time Transcoding to convert one media asset from ... Want to learn more?