This topic describes the role of TaskExecutors in the startup process of a Realtime Compute for Apache Flink cluster and explains the TaskExecutor tab.
TaskExecutors are an indispensable part in starting a Realtime Compute for Apache Flink cluster. TaskExecutors receive tasks from and return execution results to the JobManager. The number of slots is specified when a TaskExecutor is started. Only one task thread can be executed in each slot. A TaskExecutor receives tasks from the JobManager, and then builds a Netty connection with its upstream to receive and process data.
Go to the TaskExecutor tab
- Log on to the Realtime Compute development platform.
- In the top navigation bar, click Administration.
- In the Jobs section, click the name of the required job under the Job Name field.
- On the Job Administration page, click the TaskExecutor tab.
Role of TaskExecutors in cluster startupTaskExecutors are an indispensable part in starting a Realtime Compute for Apache Flink cluster. The following items describe the startup process of a Realtime Compute for Apache Flink cluster:
- When a Realtime Compute for Apache Flink cluster is started, one JobManager and several TaskExecutors are started at the same time.
- The client submits tasks to the JobManager.
- The JobManager assigns tasks to TaskExecutors.
- The TaskExecutors report the heartbeat and statistical information to the JobManager.