Request whole numbers of CPU cores (e.g., 1000m, 4000m) for containers that need exclusive cores. Then configure your pod to be in the Guaranteed Quality of Service (QoS) class. First, enable CPU manager with the Static policy in the Kubelet running on the compute nodes of your cluster. Sensitive or requires hyperthreads from the same physical CPU core.Sensitive to cross-socket memory traffic.Benefits from sharing a processor resources (e.g., data and instruction caches).If your workload is sensitive to such scenarios, then CPU Manager can be enabled to provide better performance isolation by allocating exclusive CPUs for your workload.ĬPU manager might help workloads with the following characteristics: In all the above scenarios, the performance of the workload might be affected. There might also be cases where the workload could be sensitive to context switches. When this contention intensifies, the workload can move to different CPUs depending on whether the pod is throttled and the availability of CPUs at scheduling time. In such a scenario, the pods might contend for the CPU resources available in that compute node. A single compute node in a Kubernetes cluster can run many pods and some of these pods could be running CPU-intensive workloads. Sounds Good! But Does the CPU Manager Help Me? The CPU manager feature enables better placement of workloads in the Kubelet, the Kubernetes node agent, by allocating exclusive CPUs to certain pod containers. This blog post describes the CPU Manager, a beta feature in Kubernetes. Authors: Balaji Subramaniam ( Intel), Connor Doyle ( Intel)
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |