MetricsLevel

class aws_cdk.aws_kinesisanalytics_flink_alpha.MetricsLevel(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)

Bases: Enum

(experimental) Granularity of metrics sent to CloudWatch.

Stability:: experimental
ExampleMetadata:: infused

Example:

# bucket: s3.Bucket

flink_app = flink.Application(self, "Application",
    code=flink.ApplicationCode.from_bucket(bucket, "my-app.jar"),
    runtime=flink.Runtime.FLINK_1_20,
    checkpointing_enabled=True,  # default is true
    checkpoint_interval=Duration.seconds(30),  # default is 1 minute
    min_pause_between_checkpoints=Duration.seconds(10),  # default is 5 seconds
    log_level=flink.LogLevel.ERROR,  # default is INFO
    metrics_level=flink.MetricsLevel.PARALLELISM,  # default is APPLICATION
    auto_scaling_enabled=False,  # default is true
    parallelism=32,  # default is 1
    parallelism_per_kpu=2,  # default is 1
    snapshots_enabled=False,  # default is true
    log_group=logs.LogGroup(self, "LogGroup")
)

Attributes

APPLICATION

(experimental) Application sends the least metrics to CloudWatch.

Stability:: experimental

OPERATOR

(experimental) Operator includes task-level and operator-level metrics sent to CloudWatch.

Stability:: experimental

PARALLELISM

(experimental) Send all metrics including metrics per task thread.

Stability:: experimental

TASK

(experimental) Task includes task-level metrics sent to CloudWatch.

Stability:: experimental