Amazon Web Services has made it possible to use its home-baked, Arm-powered Graviton2 CPUs with its Lambda serverless functions.
The server-renter on Wednesday slipped out news that Gravitons are an option alongside its x86 offerings from Intel and AMD.
It will come as no surprise that AWS reckons its own silicon slays on price – the company claims Graviton2 can deliver "up to 19 percent better performance at 20 percent lower cost".
But wait: there's more! AWS asserted that "Workloads using multithreading and multiprocessing, or performing many I/O operations, can experience lower execution time and, as a consequence, even lower costs."
- IBM's former Chinese Power Systems partner sues for theft of customer data
- Russian Arm SoC now shipping in Russian PCs running Russian Linux
- Google is designing its own Arm-based processors for 2023 Chromebooks – report
AWS reckons it's not hard to put that claim to the test if your functions have no binary dependencies – a happy lack that will often be the case for functions written with Node.js and Python, or functions compiled to Java bytecode.
Under those circumstances the cloud colossus suggests "changing the architecture of a Lambda function is like flipping a switch". Or, to be more accurate, changing your runtime settings to select
arm64 instead of
Serverless functions are not always lightweights that can get away with modest compute resources. Indeed, in December 2020 AWS tripled the memory it was willing to put behind Lambda functions, lifting the ceiling to 10GB of RAM and allowing half a dozen vCPUs. That spec remains an option with Graviton2, so it's not as if AWS is suggesting you run functions on just one of the Graviton2's 64 cores.
Nor is AWS restricting Graviton2 from jobs that at first blush seem best-suited to a Xeon or EPYC: the biz lets it handle heavyweight workloads including databases that need 64 cores and a terabyte of RAM.
Lambda is the 25th service AWS runs on the silicon.
Amazon.com's side hustle is not alone in using Arm-powered servers to offer cheaper cloud services – Oracle has done the same. Among hyperscalers, Microsoft has also shown signs of being Arm-curious. ®