dbaplus Community
Oct 29, 2019 · Cloud Native
How Meituan‑Dianping Scaled Kubernetes to 100k+ Nodes with HULK2.0
Meituan‑Dianping describes its evolution from a custom Docker‑based scheduler (HULK1.0) to an open‑source Kubernetes‑based platform (HULK2.0), detailing architecture, resource‑management strategies, scheduler optimizations, Kubelet enhancements, and online‑cluster tuning that together enable stable, cost‑effective operation of a 100k+ node fleet.
Cloud NativeCluster ManagementKubernetes
0 likes · 19 min read
