Serverless computing and functions as a service are both growing in appeal for enterprises, but there are still many questions about how they fit into a standard business environment. For a start, how do you manage your dependencies in a serverless environment? How are applications, themselves, tracked and managed over time? And how do the various serverless and functions-as-a-service offerings handle in real-world enterprise use cases?
To find the answers to these questions and more, we sat down with Mike Roberts, a partner at Symphonia.io and Austen Collins, founder and CEO of Serverless, Inc. Together, they’ve got front row seats into how business application developers and administrators are utilizing serverless technologies today. One of the big topics of discussion, however, still focuses on where these functions will live: in the cloud, or in Kubernetes?
“Kubernetes is so huge right now. While I’m most interested in directly cloud managed function-as-a-service platforms, there’s definitely a lot of interest there in function-as-a-service platforms running on Kubernetes. Similarly, I’m super interested in how hosted Kubernetes is developing. EKS, Amazon’s Kubernetes managed service was launched a couple of weeks ago, and once that comes out with Amazon Fargate support, which is their fully managed, don’t even have to think about it cluster with Kubernetes support, what gets super interesting then is when do people use functions-as-a-service, and when do people use a fully managed container platform? I really do think there’s a lot of scope to use both of those things,” said Roberts.
One way to answer some of the questions being asked of serverless compute applications is to create new standards, specifications and projects to support functions. The Cloud Native Computing Foundation’s Serverless Working Group has created a new project called CloudEvents to address the problem of efficiently and quickly distributing incoming data to the proper function, reliably. The CloudEvents project offers a single way to describe event data, allowing multiple platforms to support the same stream of events data.
Collins, a member of the CNCF’s Serverless Working Group, invites interested parties to join the group and contribute to the various projects they’re fomenting. “We started to investigate standards, although we can’t say ‘standards’ within the CNCF… different ways of harmonizing that are important to serverless architectures. And there are a few areas our site is aimed at: events–this is a big concept in serverless architecture… It’s bigger than serverless. Event-driven design and event-driven IT has been around for a long time. In this future that we’re heading into with serverless compute, AKA event-driven-computing, IoT, and building systems of intelligence that can make decisions in real-time, we think events are going to be a bigger concept than ever. So we’re working on that first in the Serverless Working Group,” said Collins.
In this Edition:
2:17: How is the serverless compute market shaping up halfway into 2018?
7:09: What is Amazon Fargate?
9:55: The Serverless Working Group at the Cloud Native Computing Foundation.
13:18: Can you explain CloudEvents?
18:26: Exploring the definition surrounding “event gateways.”
23:06: Data gravity and the state of the market in a serverless context.
The Cloud Native Computing Foundation is a sponsor of The New Stack.
Feature image via Pixabay.