Azure Features continuously innovates so to obtain extra with serverless functions, enabling builders to beat widespread serverless challenges by a productive, event-driven programming mannequin. Some releases we made in the previous couple of weeks are good examples of this, together with:
The brand new releases and enhancements don’t cease there, and right this moment we’re happy to current a number of developments supposed to offer a greater end-to-end expertise when constructing serverless functions. Preserve studying under to study extra concerning the following:
- A brand new method to host Azure Features in Kubernetes environments
- Stateful entities with Sturdy Features (in preview)
- Much less cluttered .NET functions with dependency injection
- Streamlined deployment with Azure DevOps
- Improved integration with Azure API Administration (in preview)
Convey Azure Features to Kubernetes with KEDA
There’s no higher method to leverage the serverless benefits than utilizing a totally managed service within the cloud like Azure Features. However some functions would possibly have to run on disconnected environments, or they require customized {hardware} and dependencies. Prospects normally take a containerized strategy for these situations, through which Kubernetes is the de facto trade customary. Managing application-aware, event-driven scale in these environments is non-trivial and normally inadequate, because it’s based mostly solely on useful resource utilization, corresponding to CPU or reminiscence.
Microsoft and Pink Hat partnered to construct Kubernetes-based event-driven auto scaling (KEDA). KEDA is an open supply part for Kubernetes that gives event-driven scale for any container workload enabling containers to scale from zero to 1000’s of situations based mostly on occasion metrics, such because the size of an Azure Queue or Kafka stream, and again to zero once more when accomplished processing.
Since Azure Features might be containerized, now you can deploy a Operate App to any Kubernetes cluster, retaining the identical scaling conduct you’ll have on the Azure Features service.
This can be a important milestone for the open supply ecosystem round Kubernetes, so we’re sharing far more element in a separate weblog publish titled, “Saying KEDA: bringing event-driven containers and features to Kubernetes.” If you wish to study extra about it, register right this moment for the Azure webinar collection scheduled for later in Might. On this webinar we are going to go extra in depth on this thrilling subject.
Sturdy Features stateful patterns
We have now been thrilled with the joy and power from the group round Sturdy Features, and our extension to the Features runtime that unlocks new stateful and workflow patterns for serverless workflows. At this time we’re releasing some new capabilities in a preview bundle of Sturdy Features.
For stateful features that map to an entity like an IoT gadget or a gaming session, you should utilize the brand new stateful entity set off for actor-like capabilities in Azure Features. We’re additionally making the state administration of your stateful features extra versatile with preview help for Redis cache because the state supplier for Sturdy Features, enabling situations the place functions might run in a disconnected or edge atmosphere.
You possibly can study extra concerning the new sturdy options in our documentation, “Sturdy Features 2.zero preview (Azure Features).”
Dependency injection for .NET functions
We’re continuously striving so as to add new patterns and capabilities that make features simpler to code, take a look at, and handle. .NET builders have been making the most of dependency injection (DI) to raised architect their functions, and right this moment we’re excited to help DI in Azure Features written in .NET. This allows simplified administration of connections plus dependent companies, and unlocks simpler testability for features that you just writer.
Be taught extra about dependency injection in our documentation, “Use dependency injection in .NET Azure Features.”
Streamlined Azure DevOps expertise
With new construct templates in Azure Pipelines, you should have the flexibility to rapidly configure your Azure Pipeline with function-optimized duties to construct your .NET, Node.js, and Python functions. We’re additionally saying the final availability of the Azure Features deployment job, which is optimized to work with the very best deployment choice on your operate app.
Moreover, with the newest Azure CLI launch we launched a brand new command that may mechanically create and configure an Azure DevOps pipeline on your operate app. The DevOps definition now lives together with your code, which permits to fantastic tune construct and deployment duties.
For extra detailed info please test our documentation, “Steady supply utilizing Azure DevOps.”
Defining and managing your Features APIs with serverless API Administration
We have now additionally simplified how one can expose and handle APIs constructed with Azure Features by API Administration. With this improved integration, the Operate Apps blade within the Azure portal presents an choice to reveal your HTTP-triggered features by a brand new or an current API in API Administration.
As soon as the Operate App is linked with API Administration, you possibly can handle API operations, apply insurance policies, edit and obtain OpenAPI specification information, or navigate to your API Administration occasion for a full-featured expertise.
Be taught extra about easy methods to expose your Operate Apps with API Administration in our documentation.
Sharing is caring
We have now additionally included a set of enhancements to the Azure Serverless Neighborhood Library, together with an up to date look, a streamlined pattern submission course of, and extra detailed details about every pattern. Take a look at the Serverless Neighborhood Library to achieve inspiration on your subsequent serverless mission, and share one thing cool when you’ve constructed it.
Get began right this moment
With Features choices increasing and rapidly bettering, we’d sincerely love to listen to your suggestions. You possibly can attain the group on Twitter and on GitHub, and we additionally actively monitor StackOverflow and UserVoice. For the newest updates, please subscribe to our month-to-month dwell webcast.
Inform us what you like about Azure Features, and begin studying extra about all the brand new capabilities we’re presenting right this moment:
