Next week at the 30th IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM), being held at Cornell Tech on Roosevelt Island in New York City, Jim Dworkin, Vice President and General Manager of Intel’s Cloud & Enterprise Acceleration Division, will present a keynote titled “Intel FPGAs for Programmable, Accelerated Infrastructure and Applications for the Modern Data Center.” Dworkin’s keynote will provide an in-depth look at the use of Intel FPGAs for both application and infrastructure offload and acceleration in the data center.
His talk includes a dissection of the market dynamics that drive how modern data centers are architected in the so-called data center of the future. Millions of FPGAs have already been deployed in this way to enable programmable infrastructure and accelerate applications. Programmable infrastructure relies heavily on IPUs (infrastructure processing units) to offload infrastructure task overhead from server CPUs, thus freeing up the servers to run more revenue-generating application workloads. Accelerating applications using FPGAs in the data center of the future is poised to grow fast based on cloud native development methodologies and a DevOps deployment framework.
Dworkin will review the new Intel IPU roadmap, which was revealed just this week at the Intel Vision 2022 conference. He will show how these IPUs deliver a programmable server-in-a-server platform for infrastructure networking and storage innovation.
Cloud native applications ride on top of these infrastructure innovations. Dworkin will explain how FPGAs will be the next wave of processing unit for accelerating these applications using containerized hardware microservices. This should appeal to the audience of distinguished researchers working on field configurable computing machines.
The keynote is scheduled for 11:50 am on May 16.
For more information about next week’s FCCM conference, including registration information, click here.
