Back in April of 2023 we announced the AMD Alveo™ MA35D media accelerator to much fanfare, if you’re not yet up to speed on that, or you just want to relive the excitement, you can find the announcement here. After showcasing our newest accelerator at the NAB (National Association of Broadcasters) Show, we were able to realize the culmination of years of learnings, planning, and good old-fashioned hard work in launching the industry’s first 5nm ASIC-based media accelerator, purpose-built for interactive streaming at scale. With that we are pleased to announce the expansion of the Alveo Media Acceleration portfolio alongside the AMD Alveo™ U30. With the launch and early-access program now behind us, it’s time to really get started.

AMD Xilinx Video SDK 3.0 brings additional enhancements to the already feature-rich Video SDK 2.0. It is now available to Alveo™ U30 media accelerator card users through  on-premises setups (support by AWS’s Amazon EC2 VT1 instances coming soon). With Video SDK 3.0 and a multimedia framework/toolkit (such as GStreamer, FFmpeg, and XMA), users can continue to cost-effectively scale their live video transcoding with even greater ease.

AMD unveils the Alveo™ X3 series network cards designed for first turnkey deployment and custom implementation paths for low latency trading applications.

The Xilinx App Store makes it easy for you to evaluate, purchase, and deploy accelerated applications. You can be deploying effortlessly in the cloud with AWS or running your apps on-premise in under 10 minutes!

This week at the SC20 virtual conference, Xilinx is presenting a technology demonstration showcasing the integration of Xilinx Alveo accelerator cards with the AMD ROCm™ open software platform.

Acquiring Solarflare brings Xilinx both market-leading technology and additional expertise in networking hardware, software, firmware and drivers. 

Victor Peng discusses how leading companies are benefiting from the AWS and Xilinx technology partnership.

AMD and Xilinx announce a new world record for AI Inference

Last summer, we kicked off the first-ever Xilinx Adaptive Computing Challenge, where we challenged developers and startups to create exciting new applications using Vitis or Vitis AI on select Xilinx hardware platforms. We announced the winners for both the developer  and startup  contests earlier this year and we're extremely proud of the innovative projects and solutions that were built.  We are excited to announce that you can now find the Deepfield-SR solution by BLUEDOT , the first place startup winner, on the new Xilinx App Store! 

One of the most critical bottlenecks in building hardware architectures for complex datapath workloads is the speed of the memory subsystem. With the two latest additions to  Alveo™ accelerator card portfolio, the Alveo U50 and Alveo U280, we set out to eliminate that bottleneck by integrating high bandwidth memory gen2 (HBM2) into each. HBM2 is stacked DRAM memory, located in-package to shrink the power and PCB footprint—but it also provides astounding memory bandwidth of 460GB/s. This amount of memory bandwidth is enough to eliminate the memory bottlenecks seen in complex compute and data intensive workloads, and it fully unlocks the incredibly parallel capability of FPGAs.