Back to main page

Cloudwatch

Abstracted Boto3 Methods:

• put_metric_data()

Usage

Batch put metric data to Cloudwatch.

The Cloudwatch client is intialised with the Namespace required.

To submit a custom Cloudwatch metric, provide the following information:

• metric_name: The name of the metric which the data will be associated with. See Metrics for more information.
• value: The value of the metric.
• timestamp: (Optional) Defaults to now(). The timestamp for when the metric is pertinent.
• dimensions: (Optional) See Dimensions for more information. Provide either a cloudwatch_dimension or a list of multiple cloudwatch_dimension.
• unit: (Optional) Defaults to Count. For more information see Metric Datum. Any of the following values may be provided: Seconds | Microseconds | Milliseconds | Bytes | Kilobytes | Megabytes | Gigabytes | Terabytes | Bits | Kilobits | Megabits | Gigabits | Terabits | Percent | Count | Bytes/Second | Kilobytes/Second | Megabytes/Second | Gigabytes/Second | Terabytes/Second | Bits/Second | Kilobits/Second | Megabits/Second | Gigabits/Second | Terabits/Second | Count/Second | None

from boto3_batch_utils import CloudwatchBatchDispatcher


cw = CloudwatchBatchDispatcher('TestService')

cw.submit_metric('DoingACountMetric', value=555)
cw.submit_metric('DoingACountMetric', value=4, unit='Seconds')

unprocessed_records = cw.flush_payloads()

Cloudwatch comes with a handy dimension builder function cloudwatch_dimension to help you construct dimensions (see below).

Advanced Usage

The Cloudwatch client can be used very simply. However, you can gain additional control with the following advanced features:

Batch Size

For information about batch sizes click here.

The Cloudwatch client has the following maximum batch limitations:

Dimensions

Cloudwatch Metrics Dimensions are optional. If required between one and ten dimensions may be used. Boto3 Batch Utils provides a helper function cloudwatch_dimension to correctly format dimensions. The inputs to this helper function are:

• name: The name of the dimension.
• value: The value of the dimension.

from boto3_batch_utils import CloudwatchBatchDispatcher, cloudwatch_dimension as cd


cw = CloudwatchBatchDispatcher('TestService')

cw.submit_metric('DoingACountMetric', value=1234, dimensions=cd('dimA', '12345'))
cw.submit_metric('DoingASecondsMetric', value=978, dimensions=[cd('timeA', '11'), cd('timeB', '4')])

cw.flush_payloads()

Uniqueness

When a record is submitted to the Cloudwatch client using submit_metric it is NOT checked for uniqueness. When sending data to Cloudwatch Metrics Stream, it is both likely and valid to send duplicate metric payloads.

Back to main page