WG-Batch Updates: What’s New and What Is Next? - Yuki Iwai, CyberAgent, Inc. & Kevin Hannon, Red Hat

This talk presents the latest updates from the Kubernetes Working Group Batch, led by Yuki Iwai from CyberAgent, Inc. and Kevin Hannon from Red Hat. The speakers explore how Kubernetes is adapting to meet the evolving demands of artificial intelligence and high-performance computing, focusing on upgrades to the Kubernetes Job API, including the managedBy and Mutable Container Resource features for enhanced elasticity and efficiency. They introduce the InPlaceRestart strategy that accelerates recovery for large-scale distributed training while minimizing overhead. Furthermore, the session discusses Kueue as a platform for AI and machine learning, emphasizing capabilities like Multi-Cluster dispatching and Fair Sharing, and concludes with insights into workload-aware scheduling in collaboration with the SIG Scheduling community.