mirror of
https://github.com/openshift/openshift-docs.git
synced 2026-02-05 12:46:18 +01:00
35 lines
2.3 KiB
Plaintext
35 lines
2.3 KiB
Plaintext
// Module included in the following assemblies:
|
|
//
|
|
// * hardware_accelerators/about-hardware-accelerators.adoc
|
|
|
|
:_mod-docs-content-type: CONCEPT
|
|
[id="nvidia-gpu-csps_{context}"]
|
|
ifndef::openshift-dedicated,openshift-rosa[]
|
|
= GPUs and CSPs
|
|
endif::openshift-dedicated,openshift-rosa[]
|
|
ifdef::openshift-rosa,openshift-dedicated[]
|
|
= GPUs and {product-title}
|
|
endif::openshift-rosa,openshift-dedicated[]
|
|
|
|
ifndef::openshift-dedicated,openshift-rosa[]
|
|
You can deploy {product-title} to one of the major cloud service providers (CSPs): {aws-first}, {gcp-full}, or Microsoft Azure.
|
|
|
|
Two modes of operation are available: a fully managed deployment and a self-managed deployment.
|
|
|
|
* In a fully managed deployment, everything is automated by Red{nbsp}Hat in collaboration with CSP. You can request an OpenShift instance through the CSP web console, and the cluster is automatically created and fully managed by Red{nbsp}Hat. You do not have to worry about node failures or errors in the environment. Red{nbsp}Hat is fully responsible for maintaining the uptime of the cluster. The fully managed services are available on {aws-short}, {azure-short}, and {gcp-short}. For {aws-short}, the OpenShift service is called (Red{nbsp}Hat OpenShift Service on AWS). For Azure, the service is called Azure Red{nbsp}Hat OpenShift. For {gcp-short}, the service is called OpenShift Dedicated on {gcp-short}.
|
|
|
|
* In a self-managed deployment, you are responsible for instantiating and maintaining the OpenShift cluster. Red{nbsp}Hat provides the OpenShift-install utility to support the deployment of the OpenShift cluster in this case. The self-managed services are available globally to all CSPs.
|
|
endif::openshift-dedicated,openshift-rosa[]
|
|
|
|
ifdef::openshift-dedicated,openshift-rosa[]
|
|
You can deploy {product-title} on NVIDIA GPU instance types.
|
|
endif::openshift-dedicated,openshift-rosa[]
|
|
|
|
It is important that this compute instance is a GPU-accelerated compute instance and that the GPU type matches the list of supported GPUs from NVIDIA AI Enterprise. For example, T4, V100, and A100 are part of this list.
|
|
|
|
You can choose one of the following methods to access the containerized GPUs:
|
|
|
|
* GPU passthrough to access and use GPU hardware within a virtual machine (VM).
|
|
|
|
* GPU (vGPU) time slicing when the entire GPU is not required.
|