https://oguzpastirmaci.com/tags/dra/Recent content in Dra on Under the ModelHugoen-USMon, 26 Oct 2026 00:00:00 +0000https://oguzpastirmaci.com/posts/gang-scheduling-and-topology-aware-scheduling-in-upstream-kubernetes/Mon, 26 Oct 2026 00:00:00 +0000https://oguzpastirmaci.com/posts/gang-scheduling-and-topology-aware-scheduling-in-upstream-kubernetes/<p>In my <a href="https://oguzpastirmaci.com/posts/rdma-network-locality-for-multi-node-jobs-with-kueue/">Kueue post</a>, Kueue did all the placement work. It kept the pods of a job in one RDMA local block or network block. It started a job only when all of its pods fit. The default scheduler only bound each pod to the node that Kueue picked.</p> <p>This is changing. In v1.35, kube-scheduler got an alpha feature that schedules a group of pods as one unit. In v1.36, another alpha feature started to keep a group in one topology domain. The work comes from the <a href="https://github.com/kubernetes/community/blob/c189d23894d0eb65b92853052e72f7b242d62516/wg-workload-aware-scheduling/charter.md">Workload-Aware Scheduling working group</a> of SIG Scheduling.</p>