China’s Alibaba Cloud and Intel on March 9 unveiled a pilot program to build a cloud-based FPGA (field programmable gate array) acceleration service.
The goal of the project is to enable cloud-services customers to have virtual access to a large set of compute resources to help them manage business, scientific and enterprise data application workloads more effectively.
A field-programmable gate array is an integrated circuit designed to be configured by a customer or a designer after manufacturing–hence “field-programmable.”
By using Intel Arria 10 FPGAs, Xeon processor-based servers and software development tools for application acceleration as a ready-to-go preconfigured infrastructure, Alibaba Cloud will provide systems designers cloud-based workload acceleration as an alternative to investing in on-premises FPGA infrastructure.
This service offers on-demand scalability of workload acceleration with FPGAs while reducing upfront investment risks and accelerating delivery of new infrastructure services, Alibaba said.
“We offer customers access to a number of services in the cloud, and adding an FPGA-based acceleration offering means they can access that powerful computing without the cost or requirement of building out their own infrastructure,” Alibaba Cloud Senior Director Jin Li said.
Alibaba Cloud is one of China’s largest online retailers and web services providers, similar to Amazon and Amazon Web Services in the U.S.
One of the key benefits of FPGAs is that they are programmable and can be customized to accelerate and scale for varying workloads, such as machine learning, data encryption and media transcode, the company said.
More information about Intel FPGAs for computing and storage can be found here in this free-of-charge report, “Data Centers are Short on Compute Capacity.” Registration is required.