When the DefaultAutoscalerV2 decides not to scale up (e.g., utilization is low), it currently sends an empty resource request ([]) to the autoscaling coordinator. This should be changed to request the previous allocation instead, preserving the dataset's current resource footprint.
Motivation
- Semantic correctness: An empty request tells the coordinator "this dataset doesn't need any resources," which is misleading. A running dataset usually needs its current allocation.
- Future-proofing: If we implement downscaling support in the future, an empty request could be misinterpreted as a signal to reclaim all resources from this dataset, causing premature downscaling.
- Consistency: Requesting the current allocation maintains a consistent view of resource needs across all autoscaler states.