Amazon Web Services, the clear market leader for public cloud infrastructure, is the first of the big cloud providers to launch Arm-based computing resources. Microsoft, Google, IBM and other companies compete with Amazon to provide public cloud infrastructure, which companies rely on to store their data and run their applications.
Across these clouds — and in corporate on-premises data centers — customers' computing workloads often run on Intel-based chips. But SoftBank-owned Arm, whose technology is widely used for chips in smartphones and tablets, has long been thought of as a potential alternative that could run while using less energy, which could lead to lower costs.
AWS' EC2 A1 computing instances rely on the Arm-based Graviton processor from Annapurna, AWS' chip-development group, Peter DeSantis, the group's vice president of global infrastructure and customer support, said during a keynote presentation at the AWS Reinvent user conference in Las Vegas on Monday. For certain workloads, like web servers, costs could be as much as 45 percent lower, DeSantis said.
Amazon bought Annapurna in 2015. Earlier, Amazon hired several people from Calxeda, a start-up that was working on Arm-based server systems.
While commerce still generates most of Amazon's revenue, AWS has become critical to the company's financial health. In the third quarter more than half of Amazon's operating profit came from AWS. While revenue from Amazon's online stores increased 10 percent year over year in the quarter, AWS had revenue growth of 46 percent. AWS now has more than 125 services available to customers, including the core EC2 computing service. The A1 instances are available now from four of AWS' data center regions around the world.
Earlier this year AWS announced the introduction of computing instances that use servers containing AMD chips.
Last year Microsoft demonstrated its Windows Server operating system running on Arm servers, but Arm-based computing is not currently possible from Microsoft's Azure public cloud.