Ververica operationalized years of experience working with Apache Flink users to provide a turnkey solution to operate and manage stateful stream processing applications. The result, Ververica Platform, reduces the infrastructure requirements to move your applications to production and allows development teams to focus on core business logic.
The underlying resource manager (Kubernetes), integrations, and default configurations are fully managed by the platform, giving enterprises the flexibility to achieve standardized deployments across different teams and avoid “reinventing the wheel” when it comes to infrastructure management and integration work.
Ververica Platform provides a secure RESTful API and a web-based user interface for performing stateful operations (e.g. upgrades, migrations, roll-backs) and other common operational tasks.
Application Manager orchestrates Apache Flink jobs in Kubernetes.
It provides a secured, RESTful API and a user interface for specifying Flink applications.
The main abstraction for specifying Flink applications is a declarative specification, including the Flink jar, parallelism, configuration parameters, and resource requirements.
Application Manager ensures that all the managed Flink applications are healthy and operating as expected.
If Application Manager is unavailable due to a failure, Flink jobs will keep running. When the Application Manager recovers, the state of all Flink applications under its control will be recovered, and any pending operations, such as migrations or deployment changes, will be completed.
Ververica Artifact Service manages application artifacts on top of a distributed file system or blob storage.
Artifacts can be either uploaded via the Ververica Platform web interface or the file system/blob storage directly. Once uploaded these artifacts can be used in Application Manager deployments.
Ververica Platform supports HDFS, NFS and the blob storage services of major cloud providers (S3, GCS, ABS, OSS).