From 9dfd36b142f4b573b3d896e1a9109b122bacace6 Mon Sep 17 00:00:00 2001
From: Dmitry Shmulevich <dshmulevich@nvidia.com>
Date: Mon, 24 Mar 2025 12:25:35 -0700
Subject: [PATCH] update topology labels

Signed-off-by: Dmitry Shmulevich <dshmulevich@nvidia.com>
---
 .../nwtopo/templates/jobset/jobset-acc.yaml   |  2 +-
 .../nwtopo/templates/jobset/jobset.yaml       |  4 +-
 .../nwtopo/templates/runai/mpijob.yaml        |  4 +-
 .../nwtopo/workflows/config-nodes-acc.yaml    | 96 +++++++++----------
 .../nwtopo/workflows/config-nodes.yaml        | 72 +++++++-------
 5 files changed, 89 insertions(+), 89 deletions(-)

diff --git a/resources/benchmarks/nwtopo/templates/jobset/jobset-acc.yaml b/resources/benchmarks/nwtopo/templates/jobset/jobset-acc.yaml
index b6ec7cdc..6fa36fb0 100644
--- a/resources/benchmarks/nwtopo/templates/jobset/jobset-acc.yaml
+++ b/resources/benchmarks/nwtopo/templates/jobset/jobset-acc.yaml
@@ -51,7 +51,7 @@ spec:
                       operator: In
                       values:
                       - {{._NAME_}}
-                  topologyKey: network.topology.kubernetes.io/accelerator
+                  topologyKey: network.topology.nvidia.com/accelerator
             containers:
             - name: test
               image: ubuntu
diff --git a/resources/benchmarks/nwtopo/templates/jobset/jobset.yaml b/resources/benchmarks/nwtopo/templates/jobset/jobset.yaml
index 413e4ca7..4cfcbd0d 100644
--- a/resources/benchmarks/nwtopo/templates/jobset/jobset.yaml
+++ b/resources/benchmarks/nwtopo/templates/jobset/jobset.yaml
@@ -53,7 +53,7 @@ spec:
                             operator: In
                             values:
                               - {{._NAME_}}
-                      topologyKey: network.topology.kubernetes.io/spine
+                      topologyKey: network.topology.nvidia.com/spine
                   - weight: 90
                     podAffinityTerm:
                       labelSelector:
@@ -62,7 +62,7 @@ spec:
                             operator: In
                             values:
                               - {{._NAME_}}
-                      topologyKey: network.topology.kubernetes.io/block
+                      topologyKey: network.topology.nvidia.com/block
             containers:
             - name: test
               image: ubuntu
diff --git a/resources/benchmarks/nwtopo/templates/runai/mpijob.yaml b/resources/benchmarks/nwtopo/templates/runai/mpijob.yaml
index 0ac015a2..1285513e 100644
--- a/resources/benchmarks/nwtopo/templates/runai/mpijob.yaml
+++ b/resources/benchmarks/nwtopo/templates/runai/mpijob.yaml
@@ -51,7 +51,7 @@ spec:
                           operator: In
                           values:
                             - {{._NAME_}}
-                    topologyKey: network.topology.kubernetes.io/spine
+                    topologyKey: network.topology.nvidia.com/spine
                 - weight: 90
                   podAffinityTerm:
                     labelSelector:
@@ -60,7 +60,7 @@ spec:
                           operator: In
                           values:
                             - {{._NAME_}}
-                    topologyKey: network.topology.kubernetes.io/block
+                    topologyKey: network.topology.nvidia.com/block
           schedulerName: runai-scheduler
           containers:
           - image: runai/mpi-worker:latest
diff --git a/resources/benchmarks/nwtopo/workflows/config-nodes-acc.yaml b/resources/benchmarks/nwtopo/workflows/config-nodes-acc.yaml
index 632d501c..f6313ce6 100644
--- a/resources/benchmarks/nwtopo/workflows/config-nodes-acc.yaml
+++ b/resources/benchmarks/nwtopo/workflows/config-nodes-acc.yaml
@@ -38,108 +38,108 @@ tasks:
       count: 1
       labels:
         node-id: n1
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw11
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw11
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n2
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw11
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw11
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n3
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw11
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw11
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n4
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw12
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw12
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n5
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw12
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw12
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n6
-        network.topology.kubernetes.io/accelerator: nvl1
-        network.topology.kubernetes.io/block: sw12
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl1
+        network.topology.nvidia.com/block: sw12
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n7
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw13
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw13
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n8
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw13
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw13
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n9
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw13
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw13
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n10
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw14
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw14
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n11
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw14
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw14
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n12
-        network.topology.kubernetes.io/accelerator: nvl2
-        network.topology.kubernetes.io/block: sw14
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/accelerator: nvl2
+        network.topology.nvidia.com/block: sw14
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     timeout: 5m
diff --git a/resources/benchmarks/nwtopo/workflows/config-nodes.yaml b/resources/benchmarks/nwtopo/workflows/config-nodes.yaml
index 25735fc3..c715cb29 100644
--- a/resources/benchmarks/nwtopo/workflows/config-nodes.yaml
+++ b/resources/benchmarks/nwtopo/workflows/config-nodes.yaml
@@ -35,100 +35,100 @@ tasks:
       count: 1
       labels:
         node-id: n1
-        network.topology.kubernetes.io/block: sw11
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw11
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n2
-        network.topology.kubernetes.io/block: sw11
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw11
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n3
-        network.topology.kubernetes.io/block: sw12
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw12
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n4
-        network.topology.kubernetes.io/block: sw12
-        network.topology.kubernetes.io/spine: sw21
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw12
+        network.topology.nvidia.com/spine: sw21
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n5
-        network.topology.kubernetes.io/block: sw13
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw13
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         net-optimal: true
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n6
-        network.topology.kubernetes.io/block: sw13
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw13
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n7
-        network.topology.kubernetes.io/block: sw14
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw14
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         net-optimal: true
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n8
-        network.topology.kubernetes.io/block: sw14
-        network.topology.kubernetes.io/spine: sw22
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw14
+        network.topology.nvidia.com/spine: sw22
+        network.topology.nvidia.com/datacenter: sw31
         net-optimal: true
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n9
-        network.topology.kubernetes.io/block: sw15
-        network.topology.kubernetes.io/spine: sw23
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw15
+        network.topology.nvidia.com/spine: sw23
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n10
-        network.topology.kubernetes.io/block: sw15
-        network.topology.kubernetes.io/spine: sw23
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw15
+        network.topology.nvidia.com/spine: sw23
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n11
-        network.topology.kubernetes.io/block: sw16
-        network.topology.kubernetes.io/spine: sw23
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw16
+        network.topology.nvidia.com/spine: sw23
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     - type: dgxa100.80g
       count: 1
       labels:
         node-id: n12
-        network.topology.kubernetes.io/block: sw16
-        network.topology.kubernetes.io/spine: sw23
-        network.topology.kubernetes.io/datacenter: sw31
+        network.topology.nvidia.com/block: sw16
+        network.topology.nvidia.com/spine: sw23
+        network.topology.nvidia.com/datacenter: sw31
         nvidia.com/gpu.count: "8"
     timeout: 5m
 - id: update