The committee that oversees the development of Knative, open source middleware for integrating Kubernetes clusters with serverless computing frameworks, announced today that a 1.0 release is now available.
Omer Bensaadon, a senior product manager for VMware and a maintainer of the Knative project, says the 1.0 designation signals a level of maturity has been achieved that should encourage every organization to deploy Knative on top of Kubernetes as part of a broader effort to make the platform more accessible to developers and IT operations teams.
Originally developed at Google, Knative has two main components. Knative Serving enables IT teams to deploy one of several serverless computing frameworks on top of Kubernetes, while Knative Eventing enables developers to build event-driven applications. Collectively, Knative makes it possible to provision secure, stateless services that scale in seconds either on a public cloud or in an on-premises IT environment, says Bensaadon.
The level of abstraction Knative provides should increase the overall appeal of Kubernetes itself, he adds. In fact, Bensaadon says every organization should install Knative to provide a layer of abstraction that makes it easier to invoke Kubernetes capabilities without having to master a range of lower-level application programming interfaces (APIs) known as custom resource definitions. The Knative project itself, however, is not governed by any open source consortium, like the Cloud Native Computing Foundation (CNCF) that oversees the development of Kubernetes.
Serverless computing frameworks are starting to play a crucial role in enabling developers to build event-driven applications. While applications based on event-driven architectures have been around for decades, interest in building and deploying them has increased sharply as organizations look to drive a wider variety of near-real-time applications. End users are becoming accustomed to applications that are updated immediately to reflect new data versus waiting for IT teams to update applications using traditional batch-mode processes.
Amazon Web Services (AWS) makes available a proprietary AWS Lambda serverless computing platform to drive those classes of applications. However, there are also several open source frameworks that other cloud service providers are embracing to provide a similar capability.
It may be a while before event-driven applications spanning multiple cloud platforms are deployed widely. However, in the wake of the economic downturn brought on by the COVID-19 pandemic, many organizations are either launching digital business transformation initiatives or accelerating ones already in place that are based on event-driven application architectures.
Developers, meanwhile, are embracing functional programming tools and serverless computing frameworks to run, for example, analytics processes external to their applications. That approach enables them to reduce the amount of code that needs to run inside their own applications. Arguably, the less code developers have to write the more secure the overall application environment becomes.
It’s not likely most developers will build an entire enterprise-class application using functions and serverless computing frameworks. There are plenty of longer-running applications that don’t lend themselves to serverless computing services that are invoked using functions. However, it’s clear serverless computing frameworks are here to stay. The challenge now is determining how best to incorporate them within a larger DevOps workflow that is already being used to build a wide range of classes of applications.