Dask Kubernetes
Contents
Dask Kubernetes¶
Welcome to the documentation for dask-kubernetes.
Note
If you are looking for general documentation on deploying Dask on Kubernetes new users should head to the Dask documentation page on Kubernetes.
The package dask-kubernetes provides cluster managers for Kubernetes.
KubeCluster¶
KubeCluster deploys Dask clusters on Kubernetes clusters using native
Kubernetes APIs. It is designed to dynamically launch ad-hoc deployments.
from dask_kubernetes import KubeCluster, make_pod_spec
pod_spec = make_pod_spec(image='ghcr.io/dask/dask:latest')
cluster = KubeCluster(pod_spec)
cluster.scale(10)
HelmCluster¶
HelmCluster is for managing an existing Dask cluster which has been deployed using
Helm. You must have already installed the Dask Helm chart
and have the cluster running. You can then use it to manage scaling and retrieve logs.
from dask_kubernetes import HelmCluster
cluster = HelmCluster(release_name="myrelease")
cluster.scale(10)