The Node Readiness Controller enhances node readiness management in Kubernetes by allowing operators to define custom readiness conditions tailored to their infrastructure needs. It automates the application of node taints based on specific health signals, ensuring workloads are only scheduled on fully operational nodes. This controller supports flexible enforcement modes and integrates with existing health reporting tools.

Kubernetes v1.35 introduces workload aware scheduling, enhancing how multiple Pods are scheduled together. It features a new Workload API for defining scheduling requirements and supports gang scheduling to optimize resource use for large workloads. The update also includes opportunistic batching to speed up scheduling for identical Pods.

This article outlines the ten new Alpha features in Kubernetes 1.35, focusing on enhancements aimed at improving AI workload orchestration and resource management. Key features include Device Binding Conditions, Mixed Version Proxy, and support for gang scheduling, which collectively enhance scheduling reliability and efficiency.

Kubernetes v1.35 introduces Extended Toleration Operators, allowing numeric comparisons for scheduling. This lets users set specific thresholds for tolerations, improving workload placement based on metrics like failure probability and performance. The new Gt and Lt operators enhance flexibility in managing workloads across on-demand and spot nodes.