OpsCanary
kubernetesschedulingPractitioner

Unlocking Performance: Kubernetes Pod-Level Resource Managers in Beta

5 min read Kubernetes BlogSep 15, 2026Reviewed for accuracy
Share
PractitionerHands-on experience recommended

Kubernetes v1.37 introduces Pod-Level Resource Managers, a significant enhancement aimed at improving resource allocation for latency-critical applications. This feature addresses the challenge of obtaining exclusive NUMA-aligned CPU cores or memory, which is crucial for performance-sensitive workloads. By enabling hybrid allocation models, it allows Kubelet to reserve exclusive NUMA-aligned resources for primary application containers while placing non-Guaranteed sidecars into a pod-isolated shared pool.

The Pod-Level Resource Managers build on the existing Pod-Level Resources framework and enhance the capabilities of Kubelet's Topology Manager, CPU Manager, and Memory Manager. With this feature, you can directly utilize pod-level resource declarations (.spec.resources) when making hardware placement decisions. However, keep in mind that this feature is disabled by default and requires the PodLevelResourceManagers feature gate to be enabled for use. It graduated to Beta status in v1.37 after being introduced as an Alpha feature in v1.36, indicating a maturity that makes it more reliable for production environments.

In production, you need to be aware of the implications of enabling this feature. While it offers significant advantages for resource allocation, careful consideration is required to ensure that your workloads are appropriately configured to take advantage of these capabilities. The transition from Alpha to Beta suggests that Kubernetes is confident in the stability and functionality of this feature, but as always, thorough testing in your environment is essential before rolling it out widely.

Key takeaways

  • Enable the PodLevelResourceManagers feature gate to utilize this capability.
  • Leverage pod-level resource declarations for more efficient hardware placement decisions.
  • Understand the hybrid allocation models for managing primary and sidecar containers effectively.

Why it matters

This feature allows for more efficient resource management, which can lead to improved application performance and reduced latency for critical workloads. By optimizing resource allocation, you can enhance the overall efficiency of your Kubernetes clusters.

When NOT to use this

The official docs don't call out specific anti-patterns here. Use your judgment based on your scale and requirements.

Want the complete reference?

Read official docs

Test what you just learned

Quiz questions written from this article

Take the quiz →
Linux Foundation🔥 SEPTEMBER PROMOSponsor

35% off certifications and e-learning with code SEPT26BTS35, or 40% off bundles and instructor-led training with SEPT26BTS40. New this month: the MCPA (Model Context Protocol Associate) certification.

Claim the discount →

Get the daily digest

One email. 5 articles. Every morning.

No spam. Unsubscribe anytime.