Deployment Templates

Deployment Templates are specified as part of Deployment resources in spec.template. Templates are translated into Job resources when you set the desired state of a Deployment to running.

You can think of the relation between the overall Deployment specification and its template as follows:

  • The template specifies which Flink job is executed and how to execute it, including its configuration.
  • The Deployment specification defines how job instances are managed over time, for instance how to perform upgrades or which Savepoint to restore from.

The template has two parts, metadata and spec.

  • The metadata section only accepts optional annotations. Annotations are key/value pairs used to provide additional information or configuration options.
  • The spec specifies which artifact to execute and how to execute it.
kind: Deployment

At the end of this page, you will find a full example that creates a Deployment for a running Flink job. The following sections will break down each part of the template in more detail.

The following sections are not valid on their own and should be pasted in under the spec section of the template Deployment.spec.template.spec.


The artifact section of the template specifies which artifact to execute. Currently, there is a single kind of supported artifact, jar.

JAR Artifacts

JAR artifacts must package regular Flink programs that are executed via the main method of their entry class.

  kind: jar
  jarUri: https://artifacts/flink-job.jar
  entryClass: com.ververica.myjob.EntryPoint    # Optional, if no Manifest entry
  mainArgs: --arg1=1 --arg2=2                   # Optional

A JAR is referenced by a URI. Ververica Platform supports http(s) as well as multiple blob storage services. Please refer to Artifact Management for more details on artifact storage and retrieval.

spec.template.spec.artifact.mainArgs are the positional parameters fed into the main method of the entry point class. These parameters are split by whitespace. Single parameters containing whitespace can be represented by wrapping them in single or double quotes. Double quoted values are escaped by means of backslashes: --json "{\"key\": \"value\"}" results in two parameters: --json and {"key": "value"}.

Parallelism, Number of Taskmanagers, and Slots

You can specify the parallelism of your jobs via the parallelism key. By default, there will be as many taskmanager instances created as the specified parallelism. You can overwrite this behavior via the numberOfTaskManagers key.

parallelism: 1
numberOfTaskManagers: 1 # Optional, defaults to parallelism

Each taskmanager will have Flink’s default setting for number of task slots (currently a single slot per taskmanager). Combining the configuration of parallelism and numberOfTaskManagers with the taskmanager.numberOfTaskSlots option of Flink gives you full flexibility in how to execute your jobs.

In the following snippet, we specify two taskmanager instances with 4 slots each and a parallelism of 8 for our jobs.

parallelism: 8
numberOfTaskManagers: 2
  taskmanager.numberOfTaskSlots: 4

For more details on Flink task slots, consult the official Flink documentation.

Compute Resources

You can configure requested compute resources for CPU and memory via the resources:

    cpu: 1
    memory: 1G
    cpu: 1
    memory: 2G

The keys jobmanager and taskmanager configure the respective Flink components. By default, the above values are used as defaults. You can overwrite each value selectively.

Note that resources are configured per component instance, e.g. if you use 10 taskmanager instances, Ververica Platform will in total try to acquire 10 times the configured taskmanager resources.


You can specify CPU as a decimal number, e.g. 1, 1.0, or 0.5.


You can specify memory as a decimal number with an optional memory unit as indicated by the suffix (ignoring case):

  • Gigabytes G, e.g. 1G or 1.5G
  • Megabytes M, e.g. 1000M or 1500M

If no unit is specified, the provided number is interpreted as bytes. Each memory unit is interpreted as a power of ten, e.g. 1K equals 1000 bytes.


The CPU resources you configure for your Deployments are counted against your licensed resource quota (CPU Quota).

JVM Heap Size

Application Manager will respect the container memory cut-offs specified in flinkConfiguration similar to Apache Flink’s behavior.

The following table lists the default values of the minimum memory cutoff and the cutoff ratio for jobmanager and taskmanager instances, respectively. Addtionally, you will find the configuration keys to configure the values.

  Minimum cutoff Cutoff ratio
Jobmanager 400 M 25 %
jobmanager.heap-cutoff-min: 400 jobmanager.heap-cutoff-ratio: 0.25
Taskmanager 600 M 25 %
containerized.heap-cutoff-min: 600 containerized.heap-cutoff-ratio: 0.25

As an example, take a memory request of 2G for a taskmanager. This means that 600 M will be cut off from the configured heap (as 25% of 2G is less than 600 M).


You can customize the default log4j logger configuration by setting log levels for your desired loggers.

    "": INFO            # Root log level DEBUG  # Log level of

The default log level for org.apache.flink is INFO.

Full Example

kind: Deployment
apiVersion: v1
  name: TopSpeedWindowing Example
    env: testing
  state: running
  deploymentTargetId: 57b4c290-73ad-11e7-8cf7-a6006ad3dba0
    kind: latest_savepoint
    kind: stateless
    annotations: false
        kind: jar
        mainArgs: --windowSize 10 --windowUnit minutes
        entryClass: org.apache.flink.streaming.examples.windowing.TopSpeedWindowing
      parallelism: 1
      numberOfTaskManagers: 2
          memory: 1.5g
        taskmanager.numberOfTaskSlots: 1
        state.savepoints.dir: s3://flink/savepoints
          org.apache.flink.streaming.examples: DEBUG