Resource Model

This documentation will explain the core resource model of KubeVela which is fully powered by Open Application Model (OAM).

The Application is the core API of KubeVela. It allows End User to work with a single artifact to capture the complete application deployment with simplified primitives.

This provides a simpler path for on-boarding End User to the platform without leaking low level details in runtime infrastructure. For example, they will be able to declare a “web service” without defining a detailed Kubernetes Deployment + Service combo each time, or claim the auto-scaling requirements without referring to the underlying KEDA ScaleObject. They can also declare a cloud database with same API if they want.

Every application is composed by multiple components with attachable operational behaviors (traits). For example:

The resource in KubeVela is a LEGO-style entity and does not even have fixed schema. Instead, it is assembled by below building block entities that are maintained by the platform-engineers. Though the application object doesn’t have fixed schema, it is a composition object assembled by several programmable building blocks as shown below.

The component model (ComponentDefinition API) is designed to allow component providers to encapsulate deployable/provisionable entities with a wide range of tools, and hence give a easier path to End User to deploy complicated microservices across hybrid environments at ease. A component normally carries its workload type description (i.e. WorkloadDefinition), a encapsulation module with a parameter list.

Components are shareable and reusable. For example, by referencing the same Alibaba Cloud RDS component and setting different parameter values, End User could easily provision Alibaba Cloud RDS instances of different sizes in different availability zones.

End User will use the Application entity to declare how they want to instantiate and deploy a group of certain components. In above example, it describes an application composed with Kubernetes stateless workload (component foo) and a Alibaba Cloud OSS bucket (component bar) alongside.

In above example, type: worker means the specification of this component (claimed in following properties section) will be enforced by a ComponentDefinition object named worker as below:

Hence, the properties section of only exposes two parameters to fill: image and cmd, this is enforced by the parameter list of the .spec.template field of the definition.

Traits (TraitDefinition API) are operational features provided by the platform. A trait augments the component instance with operational behaviors such as load balancing policy, network ingress routing, auto-scaling policies, or upgrade strategies, etc.

To attach a trait to component instance, the user will declare .type field to reference the specific TraitDefinition, and .properties field to set property values of the given trait. Similarly, TraitDefinition also allows you to define template for operational features.

The application also have a sidecar trait.

Please note that the End User do NOT need to know about definition objects, they learn how to use a given capability with visualized forms (or the JSON schema of parameters if they prefer). Please check the Generate Forms from Definitions section about how this is achieved.

Once the application is deployed, KubeVela will index and manage the underlying instances with name, revisions, labels and selector etc in automatic approach. These metadata are shown as below.

Consider these metadata as a standard contract for any “day 2” operation controller such as rollout controller to work on KubeVela deployed applications. This is the key to ensure the interoperability for KubeVela based platform as well.

Despite the efficiency and extensibility in abstracting application deployment, IaC (Infrastructure-as-Code) tools may lead to an issue called Infrastructure/Configuration Drift, i.e. the generated component instances are not in line with the expected configuration. This could be caused by incomplete coverage, less-than-perfect processes or emergency changes. This makes them can be barely used as a platform level building block.

Hence, KubeVela is designed to maintain all these programmable capabilities with and leverage Kubernetes control plane to eliminate the issue of configuration drifting, while still keeps the flexibly and velocity enabled by IaC.