Job APIs and scheduling services (preview): The Job API is a scheduler for scheduling future tasks that can be executed at specific times or at specific intervals. This applies to a variety of scenarios such as automated database backups, periodic data processing and ETL, email notifications, maintenance tasks and system updates, and batch processing.Dapr's Work API ensures that tasks in these scenarios are executed consistently and reliably, increasing efficiency and reducing the risk of errors. In addition, Dapr introduces a new scheduling service, a control plane service for scheduling actor alerts.
API Updates: Dapr v1.14 has updates to the API to support new features and components. These updates include the introduction of the Job API, which is used to schedule future tasks, either at a specific time or at a specific interval.Performance improvements to increase throughput and reduce latency when using Actor and Workflow.Actor multi-tenancy uses namespaces to isolate identical Actor types and prevent name conflicts. Streaming subscriptions for dynamic topic subscriptions without restarting sidecar. improved HTTP metrics filtering with path matching to prevent high base counts from leading to excessive CPU and memory usage. Outbound message projection with support for transaction commits across multiple publish/subscribe agents and state stores.
Dapr Shared (preview):By default, Dapr is injected as a sidecar to enable the Dapr API for your application for optimal availability and reliability.Dapr Shared supports two alternative deployment strategies for creating Dapr applications, per-node deployment using Kubernetes Daemonset or per-cluster deployment using Deployment. Deployment for per-node deployment using Kubernetes Daemonset or per-cluster deployment using Deployment.
Building Blocks for Dapr: Dapr provides a set of building blocks for distributed systems to build microservice applications in a standard way and deploy them to any environment. These building block APIs are independent, meaning that any number of them can be used in an application.
platform independence: Dapr is platform-agnostic and can run applications locally, in any Kubernetes cluster, on virtual or physical machines, and in other Dapr-integrated hosting environments. This makes it possible to run microservice applications in the cloud and at the edge.
Upgrade Notes: It is important to note that this release contains some disruptive changes. For information on upgrading to Dapr v1.14, please refer to the relevant section in the official documentation.