# K8s Performance Optimization

**URL:** <https://discuss.kubernetes.io/t/k8s-performance-optimization/35226>\
**Category:** General Discussions\
**Tags:** development, network\
**Created:** [September 15, 2026, 4:07pm UTC](https://discuss.kubernetes.io/t/k8s-performance-optimization/35226 "2026-09-15T16:07:34Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![liuningle6](https://sea2.discourse-cdn.com/flex016/user_avatar/discuss.kubernetes.io/liuningle6/32/18492_2.png) [@liuningle6](https://discuss.kubernetes.io/u/liuningle6)\
**Post date:** [September 15, 2026, 4:07pm UTC](https://discuss.kubernetes.io/t/k8s-performance-optimization/35226/1 "2026-09-15T16:07:35Z")

</div>

I have a problem. My environment is Kubernetes v1.34.10, using Volcano to submit large batches of multi-parameter training tasks. At one time, thousands of pod groups may be spawned for inference. Currently, both scheduling and task execution are relatively slow. Are there any optimization approaches to improve speed?