@@ -8,22 +8,29 @@ weight = 1
8
8
9
9
This document provides information about setting up Kubeflow in various environments.
10
10
11
- Before you set up Kubeflow, it 's important to have some knowledge of the following systems and tools:
11
+ It 's important that you have some knowledge of the following systems and tools:
12
12
13
13
* [ Kubernetes] ( https://kubernetes.io/docs/tutorials/kubernetes-basics/ )
14
14
* [ TensorFlow] ( https://www.tensorflow.org/get_started/ )
15
15
* [ kustomize] ( https://kustomize.io/ )
16
16
17
17
### Kubeflow requirements
18
18
19
- * kustomize version {{% kustomize-min-version %}} or later. See the
20
- [ kustomize component guide] ( https://github.com/kubeflow/manifests#using-kustomize ) for details about
21
- installing kustomize.
22
- * An existing Kubernetes cluster using Kubernetes version
23
- {{% kubernetes-min-version %}} or later:
19
+ Before installing Kubeflow, ensure that your environment meets the
20
+ following requirements:
24
21
25
- * A minimum of 0.6 CPU in cluster (Reserved for 3 replicated ambassador pods and according to your need add additional CPUs)
26
- * Node with storage >= 10 GB (Due to the ML libraries and third party packages being bundled in Kubeflow Docker images)
22
+ * You need kustomize version {{% kustomize-min-version %}} or later. See the
23
+ [ kustomize component guide] ( /docs/components/misc/kustomize/ )
24
+ for details about installing kustomize.
25
+
26
+ * If you intend to install Kubeflow on an existing Kubernetes cluster, the
27
+ cluster must meet the following requirements:
28
+
29
+ * Kubernetes version {{% kubernetes-min-version %}} or later.
30
+ * A minimum of 0.6 CPU in cluster. (Reserved for 3 replicated ambassador pods
31
+ and according to your needs add additional CPUs.)
32
+ * Node with storage >= 10 GB. (Due to the ML libraries and third party packages
33
+ bundled in the Kubeflow Docker images.)
27
34
28
35
## Get your machine-learning workflow up and running on Kubeflow
29
36
@@ -88,12 +95,17 @@ to suit your environment (cloud, on premises (on prem), or local):
88
95
89
96
## Installing command line tools
90
97
91
- * Download the ` kfctl ` command line interface from the
92
- [ Kubeflow releases page ] ( https://github.com/kubeflow/kubeflow/releases/ ) .
98
+ The following information is useful if you need or prefer to use command line
99
+ tools for deploying and managing Kubeflow:
93
100
94
- * Follow the ` kubectl ` installation and setup from the [ Kubernetes documentation] ( https://kubernetes.io/docs/tasks/tools/install-kubectl/ ) .
101
+ * Download the ` kfctl ` binary from the
102
+ [ Kubeflow releases page] ( https://github.com/kubeflow/kubeflow/releases/ ) .
95
103
96
- * Download the ` kustomize ` command line interface from the [ kustomize release page] ( https://github.com/kubernetes-sigs/kustomize/releases/ ) .
104
+ * Follow the ` kubectl ` installation and setup from the [ Kubernetes
105
+ documentation] ( https://kubernetes.io/docs/tasks/tools/install-kubectl/ ) .
106
+ As described in the Kubernetes documentation, your kubectl
107
+ version must be within one minor version of the Kubernetes version that you
108
+ use in your Kubeflow cluster.
97
109
98
110
## Troubleshooting
99
111
0 commit comments