How to Install/Deploy Red Hat OpenShift AI 3.3 # RHOAI
How to Install/Deploy Red Hat OpenShift AI 3.3 # RHOAI
Multi Language around, feel free to check all the available, and pick the one which is better for you!
Chapters
0:00 - Intro
0:19 - Official Documentation
12:15 - Logging Into the Cluster via CLI / OC command
13:46 - Checking some Information from our Current Cluster
16:33 - Checking the Cluster via WebUI
17:42 - Some Additional Commands to Check our System during the RHOAI Installation
18:22 - Let's Install the Operator LVM Storage, which will manager the StorageClass for Us
22:56 - Bonus Time! Can I see all Events that are not Normal?
24:44 - Additional Commands that will be Useful to see under the hood
29:40 - Installing the Operators Required for Red Hat Openshift AI
30:46 - Job Set Operator Installation
34:06 - Custom Metrics Autoscaler Installation
36:12 - cert-manager Operator for Red Hat OpenShift Installation
37:04 - Leader Worker Set Installation
37:26 - Red Hat Connectivity Link Installation
37:53 - Red Hat build of Kueue Installation
38:24 - SR-IOV Network Operator Installation
39:52 - Red Hat build of OpenTelemetry Installation
40:11 - Tempo Operator Installation
40:56 - NVIDIA GPU Operator Installation
42:23 - Node Feature Discovery Installation
43:38 - Kernel Module Management Installation
47:15 - Red Hat OpenShift AI Installation
51:42 - Describing the Nodes to check the Nvidia Information
53:33 - Checking the Instances Configuration via WebUI and CLI
1:00:03 - Gold Information, Checking the GPU State
1:03:16 - Accessing the Red Hat Openshift AI WebUI
1:07:32 - Closing
Operators
---
Job Set Operator (*add instance JobSetOperators)
Custom Metrics Autoscaler
cert-manager Operator for Red Hat OpenShift
Leader Worker Set
Red Hat Connectivity Link
Red Hat build of Kueue (*add instance Kueue)
SR-IOV Network Operator (*add instance SriovOperatorConfig)
Red Hat build of OpenTelemetry
Tempo Operator
Cluster Observability Operator
GPU (*add instance ClusterPolicy)
Node Feature Discovery (*add instance NodeFeatureDiscovery)
Kernel Module Management
Openshift AI (*add instance DataScienceCluster)
---
Commands used in this video
---
oc login -u admin -p Secret123 https://api.cluster02.king.lab:6443
oc get co
oc get nodes
oc get clusterversion
ssh -i .ssh/ocp [email protected]
sudo -i
lspci -nnk | grep NVIDIA
lsblk
watch -t -n1 "lsblk"
watch -t -n1 "oc get sc -A"
oc whoami --show-console
oc get LVMCluster -A
oc get LVMCluster -n openshift-lvm-storage
oc get sc
top -d1 -c
oc events -A -w
watch -t -n1 "oc get pods -A | grep -v -E '( Completed | Running )'"
watch -t -n1 "oc get pods --no-headers -A | grep -v -E '( Completed | Running )' | wc -l"
watch -t -n1 "oc describe nodes | grep -E '(nvidia.com/gpu:|nvidia.com/gpu.present)'"
oc get jobsetoperator cluster
oc get jobsetoperator cluster -o yaml
oc get ClusterPolicy gpu-cluster-policy
oc get ClusterPolicy gpu-cluster-policy -o yaml
oc get NodeFeatureDiscovery -A
oc get NodeFeatureDiscovery nfd-instance -n openshift-nfd
oc get NodeFeatureDiscovery nfd-instance -n openshift-nfd -o yaml
oc get DataScienceCluster
oc get pods -n nvidia-gpu-operator -l app.kubernetes.io/component=nvidia-driver
oc exec -it nvidia-driver-daemonset-9.6.20260303-1-p9tlk -n nvidia-gpu-operator -- nvidia-smi
watch -t -n1 "oc exec -it nvidia-driver-daemonset-9.6.20260414-0-czdbm -n nvidia-gpu-operator -- nvidia-smi"
oc get operators
oc get clusterserviceversion
---
Links
https://docs.redhat.com/en/documentat...
Would you Like to Collab?
- / @waldirio
Contact
LinkedIn - / waldirio
Instagram - / waldirio
Would you like to discuss this subject and much more? join the discord channel - / discord
Gostaria de falar sobre esse topico, e muito mais? Entre no canal do discord - / discord
Editor
LinkedIn - / narutocanada