Conversation
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
| apiVersion: metallb.io/v1beta1 | ||
| kind: IPAddressPool | ||
| metadata: | ||
| name: ingress-pool | ||
| namespace: metallb-system | ||
| spec: | ||
| addresses: | ||
| - <RESERVED IP>/32 | ||
| autoAssign: false | ||
| serviceAllocation: | ||
| priority: 50 | ||
| namespaces: | ||
| - ingress-nginx |
There was a problem hiding this comment.
| apiVersion: metallb.io/v1beta1 | |
| kind: IPAddressPool | |
| metadata: | |
| name: ingress-pool | |
| namespace: metallb-system | |
| spec: | |
| addresses: | |
| - <RESERVED IP>/32 | |
| autoAssign: false | |
| serviceAllocation: | |
| priority: 50 | |
| namespaces: | |
| - ingress-nginx | |
| apiVersion: metallb.io/v1beta1 | |
| kind: IPAddressPool | |
| metadata: | |
| name: ingress-pool | |
| namespace: metallb-system | |
| spec: | |
| addresses: | |
| - 192.168.0.250-192.168.0.251 # Example of two ip address - | |
| autoAssign: false | |
| serviceAllocation: | |
| priority: 61 | |
| namespaces: | |
| - ingress-nginx | |
| - knative-serving |
There was a problem hiding this comment.
You can select the EXACT IP you want for each service -
kubectl -n kourier-system patch svc kourier
--type='merge'
-p '{"spec": {"type": "LoadBalancer", "loadBalancerIP": "192.168.0.250"}}'
kubectl -n ingress-nginx patch svc ingress-nginx-controller --type='merge' -p '{"spec": {"type": "LoadBalancer", "loadBalancerIP": "192.168.0.251"}}'
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
| | Component | Required Capacity | | ||
| | ---------- | ----------------- | | ||
| | CPU | 2 cores | | ||
| | Memory | 16GB | | ||
| | Disk space | 100GB | | ||
|
|
||
| ### NVIDIA Run:ai - System Nodes | ||
|
|
||
| This configuration is the minimum requirement you need to install and use NVIDIA Run:ai. | ||
|
|
||
| | Component | Required Capacity | | ||
| | ---------- | ----------------- | | ||
| | CPU | 20 cores | | ||
| | Memory | 42GB | | ||
| | Disk space | 160GB | | ||
|
|
||
|
|
||
| To designate nodes to NVIDIA Run:ai system services, follow the instructions as described in [Label the NVIDIA Run:ai System Nodes](#label-the-nvidia-runai-system-nodes). | ||
|
|
||
|
|
||
| ### NVIDIA Run:ai - Worker Nodes | ||
|
|
||
| NVIDIA Run:ai supports NVIDIA SuperPods built on the A100, H100, H200, and B200 GPU architectures. These systems are optimized for high-performance AI workloads at scale. | ||
|
|
||
| The following configuration represents the minimum hardware requirements for installing and operating NVIDIA Run:ai on worker nodes. Each node must meet these specifications: | ||
|
|
||
| | Component | Required Capacity | | ||
| | --------- | ----------------- | | ||
| | CPU | 2 cores | | ||
| | Memory | 4GB | | ||
|
|
||
|
|
There was a problem hiding this comment.
imo this section is confusing.
there are 3 tables that tell different numbers.
i'm a simple customer, let me know what i need to do with only a single table and not 3
| * An nvIPAM IP address pool | ||
| * SR-IOV InfiniBand networks | ||
|
|
||
| The above CRD YAML specs can be downloaded from the following Gitlab repo: https://gitlab-master.nvidia.com/kuberpod/runai-deployment-assets. TBD: Should we add these yaml files |
There was a problem hiding this comment.
Files should be shared with external link and not gitlab repo link.
Partners/etc will not have access to Nvidia gitlab.
docs/admin/runai-setup/self-hosted/bcm/install-control-plane.md
Outdated
Show resolved
Hide resolved
| Run the following command. Replace `global.domain=<DOMAIN>` with the one obtained [here](./system-requirements.md#fully-qualified-domain-name-fqdn) | ||
|
|
||
| ```bash | ||
| helm upgrade -i runai-backend -n runai-backend runai-backend/control-plane \ |
There was a problem hiding this comment.
This will fail for airgapped superpods as there is no access to external-public chart repositories.
Consider adding Connected/airgapped options, same options we have for self-hosted control-plane installation.
If superpod is connected, we need to pull the chart first with the following:
helm repo add runai-backend https://runai.jfrog.io/artifactory/cp-charts-prod
helm repo updateCo-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
5 similar comments
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
Co-authored-by: Ofir Eldar <ofir.eldar.github@gmail.com>
|
Preview environment URL: https://d161wck8lc3ih2.cloudfront.net/PR-1456/ |
No description provided.