AWS Batch introduced the minScaleDownDelayMinutes parameter, allowing users to set how long managed compute environment instances remain active after finishing jobs, from 20 minutes up to one week. This prevents frequent instance termination and relaunching, mitigating delays in handling intermittent workloads.
By configuring this delay via API or console, users can better align resource availability with job patterns, improving throughput and scheduling predictability without incurring unnecessary costs. The feature is globally available across AWS regions supporting Batch services.
Source & further reading
AWS Batch now supports configurable scale down delay ↗Source published 2026-03-02. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.