Kubernetes has released version 1.37, codenamed Garhwal, delivering 67 enhancements including 16 stable features, 23 beta capabilities, 27 alpha additions, and one deprecation, according to a report published this week by The New Stack. The release arrives as the Cloud Native Computing Foundation announced graduation status for three major projects—Kubeflow, Karmada, and Cloud Native Buildpacks—signaling their maturity and long-term sustainability. The report tracks recent developments across the Kubernetes ecosystem ahead of KubeCon + CloudNativeCon North America, scheduled for November 9-12 in Salt Lake City.
The Kubernetes v1.37 update introduces several significant features, including completed resilient watch cache initialization designed to improve stability for large clusters and prevent control plane failures. KYAML has achieved stable status, offering a solution to whitespace sensitivity and the "Norway Problem" in YAML configuration. HorizontalPodAutoscaler scale-to-zero has graduated to beta and is now enabled by default, allowing pods using object or external metrics to scale down to zero when not in use. The release also brings beta support for manifest-based admission control and alpha support for pod-level checkpoint and restore capabilities. Kubeflow, the platform for AI and machine learning training on Kubernetes, has recorded 260 million PyPI downloads to date. Karmada's latest version, v1.19, advances multi-component scheduling for distributed AI training jobs, while Cloud Native Buildpacks joined CNCF as a sandbox project in 2018 before its recent graduation.
The report highlights that graduation status indicates a project is highly mature, has undergone security reviews, and operates under a vendor-neutral governance structure. "Graduation marks a critical milestone, cementing Kubeflow as a mature option for enterprise AI workloads on Kubernetes," says CNCF CTO Chris Aniszczyk in the graduation announcement. The report emphasizes that access control often doesn't make day-zero checklists alongside networking and storage for most on-premises clusters. Self-hosted Kubernetes includes authentication and authorization mechanisms, but teams must configure integration with an external identity provider—without it, operators may depend on static client certificates or long-lived tokens that create security vulnerabilities when they remain valid longer than intended.
VMware Cloud Foundation 9.1.1 adds enhanced observability into real-time Kubernetes operations, reducing standard five-minute polling intervals to two-second metric streaming to help operators detect short-lived pods, memory spikes, and transient performance issues. The update includes a new AI Assistant for VCF that provides a conversational interface for troubleshooting, diagnostics, and health checks. Azure Kubernetes Service has made autoscaling for virtual machine node pools generally available in its latest release notes, with Kubernetes v1.37 preview now rolling out and patches for previous versions available. HPE has converged the Morpheus Terraform Provider functionality into the HPE Terraform provider and released tfmigrator, a tool that automates migration from the standalone Morpheus provider to the unified HPE provider. The report notes that KubeCon offers a scholarship program for underrepresented groups and those without funding means, with travel funding requests closing Sunday, September 13, 11:59 p.m. Mountain Daylight Time. The convergence of AI capabilities into Kubernetes operations and the maturation of multi-cloud orchestration tools suggest the platform is evolving beyond container management into a broader infrastructure control plane. Organizations will need to weigh the operational overhead of these new features against their actual deployment needs before adopting them at scale.