Skip to content

Commit 5446032

Browse files
adellapeadellape
authored
Merge pull request #85507 from StephenJamesSmith/TELCODOCS-1956
2 parents 97e2356 + 6c75b8b commit 5446032

20 files changed

+143
-20
lines changed

‎_topic_maps/_topic_map.yml

Lines changed: 2 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -90,9 +90,6 @@ Topics:
9090
- Name: Control plane architecture
9191
File: control-plane
9292
Distros: openshift-enterprise,openshift-origin,openshift-online
93-
- Name: NVIDIA GPU architecture overview
94-
File: nvidia-gpu-architecture-overview
95-
Distros: openshift-enterprise
9693
- Name: Understanding OpenShift development
9794
File: understanding-development
9895
Distros: openshift-enterprise
@@ -3456,6 +3453,8 @@ Distros: openshift-origin,openshift-enterprise
34563453
Topics:
34573454
- Name: About hardware accelerators
34583455
File: about-hardware-accelerators
3456+
- Name: NVIDIA GPU architecture
3457+
File: nvidia-gpu-architecture
34593458
---
34603459
Name: Backup and restore
34613460
Dir: backup_and_restore

‎hardware_accelerators/about-hardware-accelerators.adoc

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,7 +16,7 @@ Red{nbsp}Hat {product-title} provides support for cards and peripheral hardware
1616
* Data processing units (DPUs)
1717
1818
19-
image::OCP_HW_Accelerators_4.png[Supported hardware accelerators cards and peripherals]
19+
image::OCP_HW_Accelerators_5.png[Supported hardware accelerators cards and peripherals]
2020

2121
Specialized hardware accelerators provide a rich set of benefits for AI/ML development:
2222

@@ -44,3 +44,4 @@ NVIDIA GPU Operator on Red Hat OpenShift Container Platform]
4444
* link:https://www.amd.com/en/products/accelerators/instinct.html[AMD Instinct Accelerators]
4545
4646
* link:https://www.intel.com/content/www/us/en/products/details/processors/ai-accelerators/gaudi-overview.html[Intel Gaudi Al Accelerators]
47+

‎hardware_accelerators/nvidia-gpu-architecture.adoc

Lines changed: 106 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,106 @@
1+
:_mod-docs-content-type: ASSEMBLY
2+
[id="nvidia-gpu-architecture"]
3+
= NVIDIA GPU architecture
4+
include::_attributes/common-attributes.adoc[]
5+
:context: nvidia-gpu-architecture
6+
7+
toc::[]
8+
9+
NVIDIA supports the use of graphics processing unit (GPU) resources on {product-title}. {product-title} is a security-focused and hardened Kubernetes platform developed and supported by Red Hat for deploying and managing Kubernetes clusters at scale. {product-title} includes enhancements to Kubernetes so that users can easily configure and use NVIDIA GPU resources to accelerate workloads.
10+
11+
The NVIDIA GPU Operator uses the Operator framework within {product-title} to manage the full lifecycle of NVIDIA software components required to run GPU-accelerated workloads.
12+
13+
These components include the NVIDIA drivers (to enable CUDA), the Kubernetes device plugin for GPUs, the NVIDIA Container Toolkit, automatic node tagging using GPU feature discovery (GFD), DCGM-based monitoring, and others.
14+
15+
[NOTE]
16+
====
17+
The NVIDIA GPU Operator is only supported by NVIDIA. For more information about obtaining support from NVIDIA, see link:https://access.redhat.com/solutions/5174941[Obtaining Support from NVIDIA].
18+
====
19+
20+
//TELCODOCS-1956 - Moved to Hardware accelerators section
21+
22+
include::modules/nvidia-gpu-prerequisites.adoc[leveloffset=+1]
23+
24+
// New enablement modules
25+
ifndef::openshift-dedicated,openshift-rosa[]
26+
include::modules/nvidia-gpu-enablement.adoc[leveloffset=+1]
27+
28+
include::modules/nvidia-gpu-bare-metal.adoc[leveloffset=+2]
29+
[role="_additional-resources"]
30+
.Additional resources
31+
* link:https://docs.nvidia.com/ai-enterprise/deployment-guide-openshift-on-bare-metal/0.1.0/on-bare-metal.html[Red Hat OpenShift on Bare Metal Stack]
32+
33+
include::modules/nvidia-gpu-virtualization.adoc[leveloffset=+2]
34+
[role="_additional-resources"]
35+
.Additional resources
36+
* link:https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/openshift/openshift-virtualization.html[NVIDIA GPU Operator with OpenShift Virtualization]
37+
38+
include::modules/nvidia-gpu-vsphere.adoc[leveloffset=+2]
39+
[role="_additional-resources"]
40+
.Additional resources
41+
* link:https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/openshift/nvaie-with-ocp.html#openshift-container-platform-on-vmware-vsphere-with-nvidia-vgpus[OpenShift Container Platform on VMware vSphere with NVIDIA vGPUs]
42+
43+
include::modules/nvidia-gpu-kvm.adoc[leveloffset=+2]
44+
[role="_additional-resources"]
45+
.Additional resources
46+
* link:https://computingforgeeks.com/how-to-deploy-openshift-container-platform-on-kvm/[How To Deploy OpenShift Container Platform 4.13 on KVM]
47+
48+
include::modules/nvidia-gpu-csps.adoc[leveloffset=+2]
49+
[role="_additional-resources"]
50+
.Additional resources
51+
* link:https://docs.nvidia.com/ai-enterprise/deployment-guide-cloud/0.1.0/aws-redhat-openshift.html[Red Hat Openshift in the Cloud]
52+
endif::openshift-dedicated,openshift-rosa[]
53+
54+
// Include this module at a higher leveloffset for OSD/ROSA.
55+
ifdef::openshift-dedicated,openshift-rosa[]
56+
include::modules/nvidia-gpu-csps.adoc[leveloffset=+1]
57+
[role="_additional-resources"]
58+
.Additional resources
59+
* link:https://docs.nvidia.com/ai-enterprise/deployment-guide-cloud/0.1.0/aws-redhat-openshift.html[Red Hat Openshift in the Cloud]
60+
endif::openshift-dedicated,openshift-rosa[]
61+
62+
ifndef::openshift-dedicated,openshift-rosa[]
63+
include::modules/nvidia-gpu-red-hat-device-edge.adoc[leveloffset=+2]
64+
[role="_additional-resources"]
65+
.Additional resources
66+
* link:https://cloud.redhat.com/blog/how-to-accelerate-workloads-with-nvidia-gpus-on-red-hat-device-edge[How to accelerate workloads with NVIDIA GPUs on Red Hat Device Edge]
67+
endif::openshift-dedicated,openshift-rosa[]
68+
69+
// TELCODOCS-1092 GPU sharing methods
70+
include::modules/nvidia-gpu-sharing-methods.adoc[leveloffset=+1]
71+
.Additional resources
72+
* link:https://developer.nvidia.com/blog/improving-gpu-utilization-in-kubernetes/[Improving GPU Utilization]
73+
74+
include::modules/nvidia-gpu-cuda-streams.adoc[leveloffset=+2]
75+
.Additional resources
76+
* link:https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#asynchronous-concurrent-execution[Asynchronous Concurrent Execution]
77+
78+
include::modules/nvidia-gpu-time-slicing.adoc[leveloffset=+2]
79+
80+
include::modules/nvidia-gpu-cuda-mps.adoc[leveloffset=+2]
81+
.Additional resources
82+
* link:https://docs.nvidia.com/deploy/mps/index.html[CUDA MPS]
83+
84+
include::modules/nvidia-gpu-mig-gpu.adoc[leveloffset=+2]
85+
.Additional resources
86+
* link:https://docs.nvidia.com/datacenter/tesla/mig-user-guide/[NVIDIA Multi-Instance GPU User Guide]
87+
88+
include::modules/nvidia-gpu-virtualization-with-gpu.adoc[leveloffset=+2]
89+
.Additional resources
90+
* link:https://www.nvidia.com/en-us/data-center/virtual-solutions/[Virtual GPUs]
91+
92+
include::modules/nvidia-gpu-features.adoc[leveloffset=+1]
93+
[role="_additional-resources"]
94+
.Additional resources
95+
96+
* link:https://docs.nvidia.com/ngc/ngc-deploy-on-premises/nvidia-certified-systems/index.html[NVIDIA-Certified Systems]
97+
* link:https://docs.nvidia.com/ai-enterprise/index.html#deployment-guides[NVIDIA AI Enterprise]
98+
* link:https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/overview.html#[NVIDIA Container Toolkit]
99+
* link:https://docs.nvidia.com/datacenter/cloud-native/openshift/latest/enable-gpu-monitoring-dashboard.html[Enabling the GPU Monitoring Dashboard]
100+
* link:https://docs.nvidia.com/datacenter/cloud-native/openshift/latest/mig-ocp.html[MIG Support in OpenShift Container Platform]
101+
* link:https://docs.nvidia.com/datacenter/cloud-native/openshift/latest/time-slicing-gpus-in-openshift.html[Time-slicing NVIDIA GPUs in OpenShift]
102+
* link:https://docs.nvidia.com/datacenter/cloud-native/openshift/latest/mirror-gpu-ocp-disconnected.html[Deploy GPU Operators in a disconnected or airgapped environment]
103+
// Topic not available in OSD/ROSA
104+
ifndef::openshift-dedicated,openshift-rosa[]
105+
* xref:../hardware_enablement/psap-node-feature-discovery-operator.html[Node Feature Discovery Operator]
106+
endif::openshift-dedicated,openshift-rosa[]

‎images/OCP_HW_Accelerators_5.png

23.1 KB
Loading

‎modules/nvidia-gpu-architecture.adoc

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
// Module included in the following assemblies:
2+
//
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
4+
5+
:_mod-docs-content-type: CONCEPT
6+
[id="nvidia-gpu-architecture_{context}"]
7+
= NVIDIA GPU architecture
8+
9+
NVIDIA supports the use of graphics processing unit (GPU) resources on {product-title}. {product-title} is a security-focused and hardened Kubernetes platform developed and supported by Red Hat for deploying and managing Kubernetes clusters at scale. {product-title} includes enhancements to Kubernetes so that users can easily configure and use NVIDIA GPU resources to accelerate workloads.
10+
11+
The NVIDIA GPU Operator leverages the Operator framework within {product-title} to manage the full lifecycle of NVIDIA software components required to run GPU-accelerated workloads.
12+
13+
These components include the NVIDIA drivers (to enable CUDA), the Kubernetes device plugin for GPUs, the NVIDIA Container Toolkit, automatic node tagging using GPU feature discovery (GFD), DCGM-based monitoring, and others.
14+
15+
[NOTE]
16+
====
17+
The NVIDIA GPU Operator is only supported by NVIDIA. For more information about obtaining support from NVIDIA, see link:https://access.redhat.com/solutions/5174941[Obtaining Support from NVIDIA].
18+
====

‎modules/nvidia-gpu-bare-metal.adoc

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
// Module included in the following assemblies:
22
//
3-
// * architecture/nvidia-gpu-architecture-overview.adoc
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
44

55
:_mod-docs-content-type: CONCEPT
66
[id="nvidia-gpu-bare-metal_{context}"]

‎modules/nvidia-gpu-csps.adoc

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
// Module included in the following assemblies:
22
//
3-
// * architecture/nvidia-gpu-architecture-overview.adoc
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
44

55
:_mod-docs-content-type: CONCEPT
66
[id="nvidia-gpu-csps_{context}"]

‎modules/nvidia-gpu-cuda-mps.adoc

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
// Module included in the following assemblies:
22
//
3-
// * architecture/nvidia-gpu-architecture-overview.adoc
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
44

55
:_mod-docs-content-type: CONCEPT
66
[id="nvidia-gpu-cuda-mps_{context}"]

‎modules/nvidia-gpu-cuda-streams.adoc

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
// Module included in the following assemblies:
22
//
3-
// * architecture/nvidia-gpu-architecture-overview.adoc
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
44

55
:_mod-docs-content-type: CONCEPT
66
[id="nvidia-gpu-cuda-streams_{context}"]

‎modules/nvidia-gpu-enablement.adoc

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
// Module included in the following assemblies:
22
//
3-
// * architecture/nvidia-gpu-architecture-overview.adoc
3+
// * hardware_accelerators/about-hardware-accelerators.adoc
44

55
:_mod-docs-content-type: CONCEPT
66
[id="nvidia-gpu-enablement_{context}"]

0 commit comments

Comments
 (0)