AWS Batch 为 SageMaker 训练任务推出配额管理，支持跨团队借还 GPU 资源
原标题：Lend and borrow resources across teams with AWS Batch’s quota management for SageMaker Training jobs
AI 摘要
AWS 为 SageMaker Training jobs 的 AWS Batch 支持推出配额管理（quota management）能力，管理员可按团队或项目划分配额份额，并配置 RESERVE、LEND、LEND_AND_BORROW 三种资源共享策略。该机制支持跨份额抢占以回收借出容量，以及份额内基于优先级的抢占，从而在保障各团队分配额度的同时提升 GPU 利用率。文章通过 SageMaker Python SDK v3 示例演示了 Scheduling Policy、Service Environment、Job Queue 与三个配额份额的配置方式。
正文节选
Lend and borrow resources across teams with AWS Batch’s quota management for SageMaker Training jobs With the launch of AWS Batch support for Amazon SageMaker Training jobs, customers gained a new fully managed way to queue, schedule, and retry machine learning (ML) training workloads without building their own infrastructure. Instead, they use the queuing and scheduling features of AWS Batch. As organizations scale their artificial intelligence and machine learning (AI/ML) operations and onboar