Adds a setting to cmd/autoscale that configures an upper and lower threshold for triggering Kinesis scaling events
Adds a tag to any scaled Kinesis streams that is used to disrupt runaway scaling events
Updates the Kinesis scaling strategy to be dynamic (more shards at the low end and less shards at the high end)
Updates Kinesis datapoints to support scaling up and down across any time range
Motivation and Context
These changes make the Kinesis autoscale app closer to the deprecated sample solution provided by AWS and adds some new features (like dynamic scaling). The threshold setting is partially based on DynamoDB's auto scaling solution, but operating around the limitations of the Kinesis UpdateShardCount API.
In the future we may adjust the dynamic scaling pattern, which would result in a patch version bump.
How Has This Been Tested?
This has been tested end-to-end on several production data pipelines.
Types of changes
[ ] Bug fix (non-breaking change which fixes an issue)
[x] New feature (non-breaking change which adds functionality)
[ ] Breaking change (fix or feature that would cause existing functionality to change)
Checklist:
[x] My code follows the code style of this project.
[x] My change requires a change to the documentation.
Description
cmd/autoscalethat configures an upper and lower threshold for triggering Kinesis scaling eventsMotivation and Context
These changes make the Kinesis autoscale app closer to the deprecated sample solution provided by AWS and adds some new features (like dynamic scaling). The threshold setting is partially based on DynamoDB's auto scaling solution, but operating around the limitations of the Kinesis
UpdateShardCountAPI.In the future we may adjust the dynamic scaling pattern, which would result in a patch version bump.
How Has This Been Tested?
This has been tested end-to-end on several production data pipelines.
Types of changes
Checklist: