Wednesday, January 22, 2020

There is also one more advantage to exposing REST API s from the store. The APIs can be deployed with the same deployment charts as that of the store
The deployment charts are a significant utility for any solutions involving streams of data. They make it easy to deploy the stream store and its components along with their APIs and services. Whether the deployment is over one container orchestration framework or another, the charts come helpful in declaring the requirements. These same charts also help with the performance and scalability of the store because they determine the container deployment strategies. The use of containers over standalone deployment enables resilience, fault tolerance, horizontal scaling and load balancing. All of the techniques for which the container orchestration frameworks are well-known for, are equally applicable to the stream store when deployed with its charts.
The number of resources for each component depends on the load. Some storage products can be deployed in high availability mode. This involves a cluster instead of a single host. The number of nodes in the cluster can be dynamically scaled based on load. The stream store is no different. Deployment charts provide an ability to scale the nodes of the store services. The deployment charts also define order of creation and destruction of pods with the help of a statefulset.  The pod deployment policies, their security and initialization routines are based on the application charts.
Another important aspect of APIs is the ability to include documentation. Swagger, for example, uses automatic documentation generation whenever the API descriptions change. The documentation is a critical aspect of the popularity of APIs.

Thus, a mix of conventional design and new APIs improves the audience for streaming data platform.

No comments:

Post a Comment