Alibaba Cloud, the cloud computing arm of Chinese e-commerce giant Alibaba Group Holding Ltd., is teaming up with Intel Corp. in a pilot program that will see the chip maker’s new line of Field Programmable Gate Array chips made available to its customers as a service.

Accelerators such as Intel’s FPGAs are able to help speed up the performance of large-scale computing environments while keeping power consumption to a minimum. That’s because FPGA chips are programmable and can be customized for certain types of workloads, unlike standard microprocessors that are built for a specific purpose.

The technology has actually been around for quite a few years and has seen much use in high-performance computing and supercomputing environments, but it’s only recently emerged as an option in enterprise data centers. The potential is great, however. Intel shipped its first Xeon chips with Arria 10 FPGA accelerators only last year, saying it believes the technology could be deployed in up to 30 percent of the world’s data center servers by 2020.

With today’s move, Alibaba Cloud’s customers will be able to benefit from the substantial performance boost offered by Intel’s Arria 10 FPGAs paired with its Xeon processor-based servers, the companies said in a statement. They added that the pilot program would give enterprises a cheaper alternative to buying their own FPGA infrastructure, providing on-demand scalability of workload acceleration without the need for any major investments.

“Adding an FPGA-based acceleration offering means [customers] can access powerful computing without the cost or requirement of building out their own infrastructure,” said Jin Li, senior director of Alibaba Cloud. “This service greatly adds to our value as a leading provider of highly scalable cloud computing and data management services that provide businesses with flexible, reliable connectivity.”

Alibaba Cloud said Intel’s FPGAs can be customized to accelerate and scale for a variety of enterprise workloads including data encryption, machine learning and media transcoding.

Image: Alibaba/Intel