Skip to content

Commit 19a0808

Browse files
dmitshdmitsh
authored
update topology labels (#119)
Signed-off-by: Dmitry Shmulevich <[email protected]>
1 parent 138d600 commit 19a0808

File tree

5 files changed

+89
-89
lines changed

5 files changed

+89
-89
lines changed

‎resources/benchmarks/nwtopo/templates/jobset/jobset-acc.yaml‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -51,7 +51,7 @@ spec:
5151
operator: In
5252
values:
5353
- {{._NAME_}}
54-
topologyKey: network.topology.kubernetes.io/accelerator
54+
topologyKey: network.topology.nvidia.com/accelerator
5555
containers:
5656
- name: test
5757
image: ubuntu

‎resources/benchmarks/nwtopo/templates/jobset/jobset.yaml‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -53,7 +53,7 @@ spec:
5353
operator: In
5454
values:
5555
- {{._NAME_}}
56-
topologyKey: network.topology.kubernetes.io/spine
56+
topologyKey: network.topology.nvidia.com/spine
5757
- weight: 90
5858
podAffinityTerm:
5959
labelSelector:
@@ -62,7 +62,7 @@ spec:
6262
operator: In
6363
values:
6464
- {{._NAME_}}
65-
topologyKey: network.topology.kubernetes.io/block
65+
topologyKey: network.topology.nvidia.com/block
6666
containers:
6767
- name: test
6868
image: ubuntu

‎resources/benchmarks/nwtopo/templates/runai/mpijob.yaml‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -51,7 +51,7 @@ spec:
5151
operator: In
5252
values:
5353
- {{._NAME_}}
54-
topologyKey: network.topology.kubernetes.io/spine
54+
topologyKey: network.topology.nvidia.com/spine
5555
- weight: 90
5656
podAffinityTerm:
5757
labelSelector:
@@ -60,7 +60,7 @@ spec:
6060
operator: In
6161
values:
6262
- {{._NAME_}}
63-
topologyKey: network.topology.kubernetes.io/block
63+
topologyKey: network.topology.nvidia.com/block
6464
schedulerName: runai-scheduler
6565
containers:
6666
- image: runai/mpi-worker:latest

‎resources/benchmarks/nwtopo/workflows/config-nodes-acc.yaml‎

Lines changed: 48 additions & 48 deletions
Original file line numberDiff line numberDiff line change
@@ -38,108 +38,108 @@ tasks:
3838
count: 1
3939
labels:
4040
node-id: n1
41-
network.topology.kubernetes.io/accelerator: nvl1
42-
network.topology.kubernetes.io/block: sw11
43-
network.topology.kubernetes.io/spine: sw21
44-
network.topology.kubernetes.io/datacenter: sw31
41+
network.topology.nvidia.com/accelerator: nvl1
42+
network.topology.nvidia.com/block: sw11
43+
network.topology.nvidia.com/spine: sw21
44+
network.topology.nvidia.com/datacenter: sw31
4545
nvidia.com/gpu.count: "8"
4646
- type: dgxa100.80g
4747
count: 1
4848
labels:
4949
node-id: n2
50-
network.topology.kubernetes.io/accelerator: nvl1
51-
network.topology.kubernetes.io/block: sw11
52-
network.topology.kubernetes.io/spine: sw21
53-
network.topology.kubernetes.io/datacenter: sw31
50+
network.topology.nvidia.com/accelerator: nvl1
51+
network.topology.nvidia.com/block: sw11
52+
network.topology.nvidia.com/spine: sw21
53+
network.topology.nvidia.com/datacenter: sw31
5454
nvidia.com/gpu.count: "8"
5555
- type: dgxa100.80g
5656
count: 1
5757
labels:
5858
node-id: n3
59-
network.topology.kubernetes.io/accelerator: nvl1
60-
network.topology.kubernetes.io/block: sw11
61-
network.topology.kubernetes.io/spine: sw21
62-
network.topology.kubernetes.io/datacenter: sw31
59+
network.topology.nvidia.com/accelerator: nvl1
60+
network.topology.nvidia.com/block: sw11
61+
network.topology.nvidia.com/spine: sw21
62+
network.topology.nvidia.com/datacenter: sw31
6363
nvidia.com/gpu.count: "8"
6464
- type: dgxa100.80g
6565
count: 1
6666
labels:
6767
node-id: n4
68-
network.topology.kubernetes.io/accelerator: nvl1
69-
network.topology.kubernetes.io/block: sw12
70-
network.topology.kubernetes.io/spine: sw21
71-
network.topology.kubernetes.io/datacenter: sw31
68+
network.topology.nvidia.com/accelerator: nvl1
69+
network.topology.nvidia.com/block: sw12
70+
network.topology.nvidia.com/spine: sw21
71+
network.topology.nvidia.com/datacenter: sw31
7272
nvidia.com/gpu.count: "8"
7373
- type: dgxa100.80g
7474
count: 1
7575
labels:
7676
node-id: n5
77-
network.topology.kubernetes.io/accelerator: nvl1
78-
network.topology.kubernetes.io/block: sw12
79-
network.topology.kubernetes.io/spine: sw21
80-
network.topology.kubernetes.io/datacenter: sw31
77+
network.topology.nvidia.com/accelerator: nvl1
78+
network.topology.nvidia.com/block: sw12
79+
network.topology.nvidia.com/spine: sw21
80+
network.topology.nvidia.com/datacenter: sw31
8181
nvidia.com/gpu.count: "8"
8282
- type: dgxa100.80g
8383
count: 1
8484
labels:
8585
node-id: n6
86-
network.topology.kubernetes.io/accelerator: nvl1
87-
network.topology.kubernetes.io/block: sw12
88-
network.topology.kubernetes.io/spine: sw21
89-
network.topology.kubernetes.io/datacenter: sw31
86+
network.topology.nvidia.com/accelerator: nvl1
87+
network.topology.nvidia.com/block: sw12
88+
network.topology.nvidia.com/spine: sw21
89+
network.topology.nvidia.com/datacenter: sw31
9090
nvidia.com/gpu.count: "8"
9191
- type: dgxa100.80g
9292
count: 1
9393
labels:
9494
node-id: n7
95-
network.topology.kubernetes.io/accelerator: nvl2
96-
network.topology.kubernetes.io/block: sw13
97-
network.topology.kubernetes.io/spine: sw22
98-
network.topology.kubernetes.io/datacenter: sw31
95+
network.topology.nvidia.com/accelerator: nvl2
96+
network.topology.nvidia.com/block: sw13
97+
network.topology.nvidia.com/spine: sw22
98+
network.topology.nvidia.com/datacenter: sw31
9999
nvidia.com/gpu.count: "8"
100100
- type: dgxa100.80g
101101
count: 1
102102
labels:
103103
node-id: n8
104-
network.topology.kubernetes.io/accelerator: nvl2
105-
network.topology.kubernetes.io/block: sw13
106-
network.topology.kubernetes.io/spine: sw22
107-
network.topology.kubernetes.io/datacenter: sw31
104+
network.topology.nvidia.com/accelerator: nvl2
105+
network.topology.nvidia.com/block: sw13
106+
network.topology.nvidia.com/spine: sw22
107+
network.topology.nvidia.com/datacenter: sw31
108108
nvidia.com/gpu.count: "8"
109109
- type: dgxa100.80g
110110
count: 1
111111
labels:
112112
node-id: n9
113-
network.topology.kubernetes.io/accelerator: nvl2
114-
network.topology.kubernetes.io/block: sw13
115-
network.topology.kubernetes.io/spine: sw22
116-
network.topology.kubernetes.io/datacenter: sw31
113+
network.topology.nvidia.com/accelerator: nvl2
114+
network.topology.nvidia.com/block: sw13
115+
network.topology.nvidia.com/spine: sw22
116+
network.topology.nvidia.com/datacenter: sw31
117117
nvidia.com/gpu.count: "8"
118118
- type: dgxa100.80g
119119
count: 1
120120
labels:
121121
node-id: n10
122-
network.topology.kubernetes.io/accelerator: nvl2
123-
network.topology.kubernetes.io/block: sw14
124-
network.topology.kubernetes.io/spine: sw22
125-
network.topology.kubernetes.io/datacenter: sw31
122+
network.topology.nvidia.com/accelerator: nvl2
123+
network.topology.nvidia.com/block: sw14
124+
network.topology.nvidia.com/spine: sw22
125+
network.topology.nvidia.com/datacenter: sw31
126126
nvidia.com/gpu.count: "8"
127127
- type: dgxa100.80g
128128
count: 1
129129
labels:
130130
node-id: n11
131-
network.topology.kubernetes.io/accelerator: nvl2
132-
network.topology.kubernetes.io/block: sw14
133-
network.topology.kubernetes.io/spine: sw22
134-
network.topology.kubernetes.io/datacenter: sw31
131+
network.topology.nvidia.com/accelerator: nvl2
132+
network.topology.nvidia.com/block: sw14
133+
network.topology.nvidia.com/spine: sw22
134+
network.topology.nvidia.com/datacenter: sw31
135135
nvidia.com/gpu.count: "8"
136136
- type: dgxa100.80g
137137
count: 1
138138
labels:
139139
node-id: n12
140-
network.topology.kubernetes.io/accelerator: nvl2
141-
network.topology.kubernetes.io/block: sw14
142-
network.topology.kubernetes.io/spine: sw22
143-
network.topology.kubernetes.io/datacenter: sw31
140+
network.topology.nvidia.com/accelerator: nvl2
141+
network.topology.nvidia.com/block: sw14
142+
network.topology.nvidia.com/spine: sw22
143+
network.topology.nvidia.com/datacenter: sw31
144144
nvidia.com/gpu.count: "8"
145145
timeout: 5m

‎resources/benchmarks/nwtopo/workflows/config-nodes.yaml‎

Lines changed: 36 additions & 36 deletions
Original file line numberDiff line numberDiff line change
@@ -35,100 +35,100 @@ tasks:
3535
count: 1
3636
labels:
3737
node-id: n1
38-
network.topology.kubernetes.io/block: sw11
39-
network.topology.kubernetes.io/spine: sw21
40-
network.topology.kubernetes.io/datacenter: sw31
38+
network.topology.nvidia.com/block: sw11
39+
network.topology.nvidia.com/spine: sw21
40+
network.topology.nvidia.com/datacenter: sw31
4141
nvidia.com/gpu.count: "8"
4242
- type: dgxa100.80g
4343
count: 1
4444
labels:
4545
node-id: n2
46-
network.topology.kubernetes.io/block: sw11
47-
network.topology.kubernetes.io/spine: sw21
48-
network.topology.kubernetes.io/datacenter: sw31
46+
network.topology.nvidia.com/block: sw11
47+
network.topology.nvidia.com/spine: sw21
48+
network.topology.nvidia.com/datacenter: sw31
4949
nvidia.com/gpu.count: "8"
5050
- type: dgxa100.80g
5151
count: 1
5252
labels:
5353
node-id: n3
54-
network.topology.kubernetes.io/block: sw12
55-
network.topology.kubernetes.io/spine: sw21
56-
network.topology.kubernetes.io/datacenter: sw31
54+
network.topology.nvidia.com/block: sw12
55+
network.topology.nvidia.com/spine: sw21
56+
network.topology.nvidia.com/datacenter: sw31
5757
nvidia.com/gpu.count: "8"
5858
- type: dgxa100.80g
5959
count: 1
6060
labels:
6161
node-id: n4
62-
network.topology.kubernetes.io/block: sw12
63-
network.topology.kubernetes.io/spine: sw21
64-
network.topology.kubernetes.io/datacenter: sw31
62+
network.topology.nvidia.com/block: sw12
63+
network.topology.nvidia.com/spine: sw21
64+
network.topology.nvidia.com/datacenter: sw31
6565
nvidia.com/gpu.count: "8"
6666
- type: dgxa100.80g
6767
count: 1
6868
labels:
6969
node-id: n5
70-
network.topology.kubernetes.io/block: sw13
71-
network.topology.kubernetes.io/spine: sw22
72-
network.topology.kubernetes.io/datacenter: sw31
70+
network.topology.nvidia.com/block: sw13
71+
network.topology.nvidia.com/spine: sw22
72+
network.topology.nvidia.com/datacenter: sw31
7373
net-optimal: true
7474
nvidia.com/gpu.count: "8"
7575
- type: dgxa100.80g
7676
count: 1
7777
labels:
7878
node-id: n6
79-
network.topology.kubernetes.io/block: sw13
80-
network.topology.kubernetes.io/spine: sw22
81-
network.topology.kubernetes.io/datacenter: sw31
79+
network.topology.nvidia.com/block: sw13
80+
network.topology.nvidia.com/spine: sw22
81+
network.topology.nvidia.com/datacenter: sw31
8282
nvidia.com/gpu.count: "8"
8383
- type: dgxa100.80g
8484
count: 1
8585
labels:
8686
node-id: n7
87-
network.topology.kubernetes.io/block: sw14
88-
network.topology.kubernetes.io/spine: sw22
89-
network.topology.kubernetes.io/datacenter: sw31
87+
network.topology.nvidia.com/block: sw14
88+
network.topology.nvidia.com/spine: sw22
89+
network.topology.nvidia.com/datacenter: sw31
9090
net-optimal: true
9191
nvidia.com/gpu.count: "8"
9292
- type: dgxa100.80g
9393
count: 1
9494
labels:
9595
node-id: n8
96-
network.topology.kubernetes.io/block: sw14
97-
network.topology.kubernetes.io/spine: sw22
98-
network.topology.kubernetes.io/datacenter: sw31
96+
network.topology.nvidia.com/block: sw14
97+
network.topology.nvidia.com/spine: sw22
98+
network.topology.nvidia.com/datacenter: sw31
9999
net-optimal: true
100100
nvidia.com/gpu.count: "8"
101101
- type: dgxa100.80g
102102
count: 1
103103
labels:
104104
node-id: n9
105-
network.topology.kubernetes.io/block: sw15
106-
network.topology.kubernetes.io/spine: sw23
107-
network.topology.kubernetes.io/datacenter: sw31
105+
network.topology.nvidia.com/block: sw15
106+
network.topology.nvidia.com/spine: sw23
107+
network.topology.nvidia.com/datacenter: sw31
108108
nvidia.com/gpu.count: "8"
109109
- type: dgxa100.80g
110110
count: 1
111111
labels:
112112
node-id: n10
113-
network.topology.kubernetes.io/block: sw15
114-
network.topology.kubernetes.io/spine: sw23
115-
network.topology.kubernetes.io/datacenter: sw31
113+
network.topology.nvidia.com/block: sw15
114+
network.topology.nvidia.com/spine: sw23
115+
network.topology.nvidia.com/datacenter: sw31
116116
nvidia.com/gpu.count: "8"
117117
- type: dgxa100.80g
118118
count: 1
119119
labels:
120120
node-id: n11
121-
network.topology.kubernetes.io/block: sw16
122-
network.topology.kubernetes.io/spine: sw23
123-
network.topology.kubernetes.io/datacenter: sw31
121+
network.topology.nvidia.com/block: sw16
122+
network.topology.nvidia.com/spine: sw23
123+
network.topology.nvidia.com/datacenter: sw31
124124
nvidia.com/gpu.count: "8"
125125
- type: dgxa100.80g
126126
count: 1
127127
labels:
128128
node-id: n12
129-
network.topology.kubernetes.io/block: sw16
130-
network.topology.kubernetes.io/spine: sw23
131-
network.topology.kubernetes.io/datacenter: sw31
129+
network.topology.nvidia.com/block: sw16
130+
network.topology.nvidia.com/spine: sw23
131+
network.topology.nvidia.com/datacenter: sw31
132132
nvidia.com/gpu.count: "8"
133133
timeout: 5m
134134
- id: update

0 commit comments

Comments
 (0)