AWS and Xilinx Highlight Growing FPGA-as-a-Service Adoption and Partner Collaboration at XDF 2019

11-21-2019 09:00 AM

Last month at the Xilinx Developer Forum (XDF) Americas 2019, Xilinx CEO Victor Peng was joined on stage by Gadi Hutt, Senior Director of Business Development at Amazon Web Services, and Vin Sharma, Head of Engineering for Amazon Sagemaker Neo and AWS AI, to discuss how leading companies are benefiting from the AWS and Xilinx technology partnership.

AWS1

From pharmaceutical companies like AstraZeneca analyzing 100,000 genomes in only 100 hours to enterprise network security provider Trend Micro launching high-performance Network Security Appliances, industry leaders choose to deploy accelerated cloud applications on Amazon EC2 F1 instances with Xilinx UltraScale+ FPGAs. Presently, AWS hosts the largest number of FPGA-powered Apps in the Cloud, and by popular demand, the availability of the F1 instance has been extended to its Canada (Central) AWS Region.

Additionally, Amazon SageMaker Neo provides model optimization as a service and a compact runtime to match the model with execution backend and sends sub-graphs to appropriate accelerators, either on-premise via Xilinx Alveo cards, in the cloud via Amazon EC2 F1 instances, or at the edge via Xilinx embedded FPGAs.

AWS XDF Americas announcements and highlights included:

  • AstraZeneca analyzed 100,000 genomes in only 100 hours using Amazon EC2 F1 instances with the power of Xilinx UltraScale+ FPGAs
  • Trend Micro announced Cloud Network Protection powered by Amazon EC2 F1 instances with Xilinx UltraScale+ FPGAs
  • Valtix announced the first FPGA-powered Cloud-Native Firewall on AWS, using Amazon EC2 F1 instances with Xilinx UltraScale+ FPGAs
  • AWS hosts the largest number of FPGA-powered Apps in the Cloud
  • Availability of AWS’s FPGA-based Amazon EC2 F1 instances, with integrated Xilinx UltraScale+ FPGAs, has been extended to its Canada (Central) AWS Region
  • Amazon SageMaker Neo liberates models from the framework they were trained in
  • Amazon SageMaker Neo provides model optimization as a service and a compact runtime to match the model with execution backend and sends sub-graphs to appropriate accelerators, either on-premise via Xilinx Alveo cards, in the cloud via EC2 F1 instances, or at the edge via Xilinx embedded FPGAs.
2019XDF-DAY01-180

To learn more, come see us at AWS re:Invent 2019, December 2-6 in Las Vegas, Nevada.

Visit the Xilinx booth #3709 in the Expo Hall to see the latest technology collaborations and demos with AWS and our solution partners. Xilinx technologies and speakers will also be featured in the following breakout sessions.

Chalk Talks:

Workshops:

Sessions:

We look forward to seeing you in Vegas!