Harshal Patil
unread,Aug 3, 2022, 6:07:36 AM8/3/22Sign in to reply to author
Sign in to forward
You do not have permission to delete messages in this group
Sign in to report message as abuse
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to kubernete...@googlegroups.com, Mrunal Patel, Derek Carr, releas...@kubernetes.io, kubernetes-...@googlegroups.com, Swarup Ghosh, Sai Ramesh Vanka
Enhancement name: Kubelet Evented PLEG for Better Performance
Enhancement status (alpha/beta/stable): alpha
SIG: sig-node
k/enhancements repo issue #:
https://github.com/kubernetes/enhancements/issues/3386
PR #’s:
https://github.com/kubernetes/kubernetes/pull/111642 and
https://github.com/kubernetes/kubernetes/pull/111384
Additional time needed (in days): 6
Reason this enhancement is critical for this milestone: This
enhancement implements critical performance improvement that is
essential for achieving high pod density on the node.
Risks from adding code late: (to k8s stability, testing, etc.): None.
The implementation is feature gated, so there is no risk for the
existing Generic PLEG implementation.
Risks from cutting enhancement: (partial implementation, critical
customer usecase, etc.): Users with high resource usage due to high
pod density would stand to benefit with this feature. We have the CRI
API changes already merged,
https://github.com/kubernetes/kubernetes/pull/111642. The only
standing issue was the non-blocking job
`pull-kubernetes-e2e-gce-alpha-features` failing, but that has been
fixed now and all CI jobs on the main PR,
https://github.com/kubernetes/kubernetes/pull/111384, are green now.
Thanks,
Harshal Patil