Microsoft Makes Serverless Frameworks More Accessible

At the KubeCon + CloudNativeCon 2017 conference this week, Microsoft updated its Kubernetes connector dubbed Virtual Kubelet, which provides a simpler means to invoke serverless computing frameworks without having to master a functional programming language.

In addition, Microsoft announced it is making available as open source code an instance of the Open Service Broker API for the Microsoft Azure public cloud. That software makes it possible to integrate container applications with legacy applications using the Kubernetes API. The core Open Service Broker API was developed by The Cloud Foundry Foundation (CFF).

Finally, Microsoft unveiled a Kashti dashboard for the Brigade continuous integration/continuous (CI/CD) pipeline for Kubernetes environments, which it is pushing as a modern alternative to Jenkins.

Gabe Monroy, lead program manager for Azure Containers, says Virtual Kubelet can be employed against Azure Container Instances or any runtime that supports the Microsoft interface. In theory, developers should be able to invoke multiple serverless computing frameworks running on Azure regardless of whether they were created by Microsoft or a third-party provider.

Monroy says that unlike the Lambda serverless computing framework developed by Amazon Web Services (AWS), Microsoft’s approach to serverless computing is designed to be accessible to developers using container technologies and programming tools they already know. Those serverless frameworks would make it easier to invoke an event-driven computing architecture to run batch processes in parallel to their primary workload, which in turn will make it easier to dynamically burst workloads using cloud resources as part of a hybrid cloud computing environment. In fact, Monroy says hybrid clouds will eventually become the “killer application” for serverless computing frameworks.

In general, Monroy says Microsoft is moving down a path to make container orchestration engines and serverless computing frameworks completely transparent to developers. Otherwise, the number of developers who can build applications using microservices architectures will be too constrained, says Monroy.

Microsoft is committed to exposing containers as compute-primitive versus simply providing virtual machines on which they get deployed, he says. To accomplish that goal. Microsoft has been taking advantage of lighter-weight hypervisors to provide isolation between containers.

At this point it’s not clear if a few frameworks will emerge as de facto standards, or whether there will be a multitude of serverless computing frameworks strewn across multiple clouds. Each developer, for example, may decide to deploy their own favorite serverless framework in the cloud or on-premises.

What is clear is that another tier of computing is starting to manifest itself. Initially, containers were used primarily to build stateless applications. But serverless frameworks may become a more efficient means for processing those types of applications. Containers, meanwhile, increasingly are being used to build stateful applications that should be able to invoke serverless frameworks dynamically whenever needed.

It may take a while for developers to master all these various options. But if Microsoft has its way, many of them won’t even realize what type of service is being invoked when.

Mike Vizard

Mike Vizard is a seasoned IT journalist with over 25 years of experience. He also contributed to IT Business Edge, Channel Insider, Baseline and a variety of other IT titles. Previously, Vizard was the editorial director for Ziff-Davis Enterprise as well as Editor-in-Chief for CRN and InfoWorld.

Mike Vizard has 1615 posts and counting. See all posts by Mike Vizard