All Products
Search

This Product

    Realtime Compute for Apache Flink:Flink JAR jobs

    Document Center

    Realtime Compute for Apache Flink:Flink JAR jobs

    Last Updated:Jun 20, 2026

    This guide shows you how to create, deploy, and start streaming and batch Flink JAR jobs to demonstrate the workflow in Realtime Compute for Apache Flink.

    Prerequisites

    Step 1: Develop the JAR package

    The Development Console of Realtime Compute for Apache Flink does not provide an integrated development environment (IDE) for JAR packages. You must develop, compile, and package your job locally. For more information about how to configure dependencies, use connectors, and read dependent files from Object Storage Service (OSS), see Develop a Flink JAR job.

    Important

    Ensure that the Flink version used for local development matches the engine version that you select in Step 3: Deploy the JAR job. Also, be mindful of the dependency package scope.

    To help you get started quickly with Flink JAR jobs, this guide provides a sample word-counting JAR package and a data file. You can download these files to complete the following steps.

    Step 2: Upload the JAR package and data file

    1. Log on to the Realtime Compute console.

    2. Find the target Flink workspace and click Console in the Actions column.

    3. In the left-side navigation pane, click Artifacts.

    4. Click Upload Artifact to upload the JAR package and data file.

      In this tutorial, upload the FlinkQuickStart-1.0-SNAPSHOT.jar and Shakespeare files that you downloaded in Step 1. For more information about file storage paths, see Artifacts.

    Step 3: Deploy the JAR job

    Streaming job

    1. On the O&M > Deployments page, click Create Deployment and select JAR Deployment.

    2. Configure the deployment parameters.

      Parameter

      Description

      Example

      Deployment mode

      Select Stream Mode as the deployment mode.

      Stream Mode

      Deployment name

      Enter a name for the JAR job.

      flink-streaming-test-jar

      Engine version

      The Flink engine version for the job.

      We recommend that you use a version with the RECOMMENDED or STABLE tag for better reliability and performance. For more information, see Release notes and Engine versions.

      vvr-8.0.9-flink-1.17

      JAR URI

      Select the FlinkQuickStart-1.0-SNAPSHOT.jar file that you uploaded in Step 2. You can also click the 上传 icon to upload your own JAR package.

      If the file already exists on the Artifacts page, you can select it directly.

      Note

      Realtime Compute for Apache Flink engine VVR 8.0.6 and later can only access the bucket bound to the workspace.

      -

      Entry point class

      The program's entry point. If the JAR package does not specify a main class, you must enter its fully qualified class name.

      The test JAR package provided in this document contains code for both a streaming job and a batch job. Therefore, you must specify the entry point for the streaming job.

      org.example.WordCountStreaming

      Entry point main arguments

      The arguments to pass to the main method.

      For this tutorial, enter the storage path of the Shakespeare input data file.

      --input oss://<Your-OSS-Bucket-Name>/artifacts/namespaces/<Your-Namespace>/Shakespeare

      You can copy the full path of the Shakespeare file from the Artifacts page.

      Deployment target

      From the drop-down list, select a target resource queue or session cluster. For more information, see Manage resource queues and Step 1: Create a session cluster.

      Important

      Jobs that are deployed to a session cluster do not support alert monitoring, alert configuration, or auto tuning. Do not use session clusters in production environments. Session clusters are intended for development and testing. For more information, see Debug a job.

      default-queue

      For more information about configuration parameters, see Deploy a job.

    3. Click Deploy.

    Batch job

    1. On the O&M > Deployments page, click Create Deployment and select JAR Deployment.

    2. Configure the deployment parameters.

      Parameter

      Description

      Example

      Deployment mode

      Select Batch Mode as the deployment mode.

      Batch Mode

      Deployment name

      Enter a name for the JAR job.

      flink-batch-test-jar

      Engine version

      The Flink engine version for the job.

      We recommend that you use a version with the RECOMMENDED or STABLE tag for better reliability and performance. For more information, see Release notes and Engine versions.

      vvr-8.0.9-flink-1.17

      JAR URI

      Select the FlinkQuickStart-1.0-SNAPSHOT.jar file that you uploaded in Step 2. You can also click the 上传 icon to upload your own JAR package.

      -

      Entry point class

      The entry class of the program. If your JAR package does not specify a main class, enter the fully qualified name of your Endpoint Class here.

      The test JAR package provided in this document contains code for both a streaming job and a batch job. Therefore, you must specify the entry point for the batch job.

      org.example.WordCountBatch

      Entry point main arguments

      The arguments to pass to the main method.

      For this tutorial, enter the storage paths for the Shakespeare input data file and the batch-quickstart-test-output.txt output data file.

      Note

      Specify the full path for the output file. The system creates this file automatically. In this tutorial, the output path places the file in the same directory as the input file.

      --input oss://<Your-OSS-Bucket-Name>/artifacts/namespaces/<Your-Namespace>/Shakespeare

      --output oss://<Your-OSS-Bucket-Name>/artifacts/namespaces/<Your-Namespace>/batch-quickstart-test-output.txt

      You can copy the full path of the Shakespeare file from the Artifacts page.

      Deployment target

      From the drop-down list, select a target resource queue or session cluster. For more information, see Manage resource queues and Step 1: Create a session cluster.

      Important

      Jobs that are deployed to a session cluster do not support alert monitoring, alert configuration, or auto tuning. Do not use session clusters in production environments. Session clusters are intended for development and testing. For more information, see Debug a job.

      default-queue

      For more information about configuration parameters, see Deploy a job.

    3. Click Deploy.

    Step 4: Start the job and view the results

    Streaming job

    1. On the O&M > Deployments page, find the target job and click Start in the Actions column.

    2. Select Stateless Start and click Start. For more information about how to start a job, see Start a job.

    3. After the job status changes to RUNNING, view the results of the streaming job.

      In the TaskManager log file that ends with .out, search for shakespeare to view the Flink results.

      To find the log file, go to the Logs tab, select a running TaskManager under Running Task Managers, and then click the Log List tab. Open the flink.out file and search for shakespeare to locate the results, such as (shakespeare,1).

    Batch job

    1. On the O&M > Deployments page, find the target job and click Start in the Actions column.

    2. In the Start Job dialog box, click Start. For more information about how to start a job, see Start a job.

    3. After the job status changes to FINISHED, view the results of the batch job.

      Log on to the OSS console and view the results in the oss://<Your-OSS-Bucket-Name>/artifacts/namespaces/<Your-Namespace>/batch-quickstart-test-output.txt file.

      The following is an example of the results from the batch-quickstart-test-output.txt file.

      a 164
abhor 2
abide 2
able 1
about 1
above 4
absence 5
absent 4
abundance 4
abundant 1
abuse 3
abused 1
abuses 1
abysm 1
accents 1
acceptable 1
acceptance 1
    Note

    The number of result entries for the streaming job and the batch job may differ because the TaskManager.out log displays a maximum of 2,000 entries. For more information about this limitation, see Print.

    (Optional) Step 5: Stop the job

    To apply changes to a job (such as code modifications, WITH parameter updates, or version changes), you must redeploy, stop, and then restart it. A restart is also required for a stateless start or to apply non-dynamic configuration changes. For more information about stopping a job, see Stop a job.

    Related documents