As enterprises get larger, their adoption of Lambda serverless architectures grows due to both developers and managers wanting to move faster, ship faster and have less infrastructure to manage, according to the recently published State of Serverless report published by Datadog.
“I think we found that very surprising. You typically see large organizations that are moving slowly, but when we looked at the data, we found that really wasn’t true at all,” Stephen Pinkerton, product manager at Datadog told SD Times.
Among the companies with the largest infrastructure footprints, more than three quarters have adopted Lambda. In fact, half of AWS users have adopted Lambda, showing that Lambda is no longer limited to cloud-native early adopters or for niche use cases, the report said.
RELATED CONTENT: 10 use cases for serverless
In addition, container users have flocked to Lambda. 80% of organizations in AWS that are running containers have adopted Lambda.
“As soon as people are going into the cloud, they’re moving quickly to containers,” Pinkerton said. “It really validates what we hear when we talk to customers that Lambda is being used in addition to existing infrastructure. And so it’s less common to see a lift and shift, but more companies are running hybrid environments and infrastructure.”
A variety of different types of companies are embracing serverless technologies, whether that’s Silicon Valley tech companies or companies for whom technology is not their core competency and they don’t need to manage servers and use serverless as a way to express business logic, Pinkerton continued.
Also, Node.js and Python dominate among Lambda users with 47% of all deployed Lambdas currently running Python and another 39% running Node.js applications.
AWS first launched Lambda in preview in 2014 with Node.js as the first supported runtime, before adding Java and Python support in 2015.
Other findings in the report include: the median Lambda function runs for 800 milliseconds, half of Lambda functions have the minimum memory allocation, two-thirds of defined timeouts are under 1 minute, and only 4% of functions have a defined concurrency limit.
“I think we’re going to continue to see more of these sort of event driven architectures. We’re seeing it a lot with our customers now where they’re adopting serverless for a new product. And so they’ll build an isolated piece of infrastructure and serverless and some companies we’ve even seen have said, okay, we’re going to keep all of our existing containerized infrastructure, but all new projects are going to be serverless or are going to be built on Lambda,” Pinkerton said.
