Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
dynamo
Commits
c8770464
Unverified
Commit
c8770464
authored
Jan 12, 2026
by
hhzhang16
Committed by
GitHub
Jan 12, 2026
Browse files
feat: normalize dynamo namespace computation (#5231)
Signed-off-by:
Hannah Zhang
<
hannahz@nvidia.com
>
parent
abd4b5d9
Changes
69
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
0 additions
and
55 deletions
+0
-55
examples/backends/sglang/deploy/disagg_planner.yaml
examples/backends/sglang/deploy/disagg_planner.yaml
+0
-4
examples/backends/trtllm/deploy/agg-with-config.yaml
examples/backends/trtllm/deploy/agg-with-config.yaml
+0
-2
examples/backends/trtllm/deploy/agg.yaml
examples/backends/trtllm/deploy/agg.yaml
+0
-2
examples/backends/trtllm/deploy/agg_router.yaml
examples/backends/trtllm/deploy/agg_router.yaml
+0
-2
examples/backends/trtllm/deploy/disagg-multinode.yaml
examples/backends/trtllm/deploy/disagg-multinode.yaml
+0
-3
examples/backends/trtllm/deploy/disagg.yaml
examples/backends/trtllm/deploy/disagg.yaml
+0
-3
examples/backends/trtllm/deploy/disagg_planner.yaml
examples/backends/trtllm/deploy/disagg_planner.yaml
+0
-4
examples/backends/trtllm/deploy/disagg_router.yaml
examples/backends/trtllm/deploy/disagg_router.yaml
+0
-3
examples/backends/vllm/deploy/agg.yaml
examples/backends/vllm/deploy/agg.yaml
+0
-2
examples/backends/vllm/deploy/agg_kvbm.yaml
examples/backends/vllm/deploy/agg_kvbm.yaml
+0
-2
examples/backends/vllm/deploy/agg_router.yaml
examples/backends/vllm/deploy/agg_router.yaml
+0
-2
examples/backends/vllm/deploy/agg_router_kv_approx.yaml
examples/backends/vllm/deploy/agg_router_kv_approx.yaml
+0
-2
examples/backends/vllm/deploy/disagg-multinode.yaml
examples/backends/vllm/deploy/disagg-multinode.yaml
+0
-3
examples/backends/vllm/deploy/disagg.yaml
examples/backends/vllm/deploy/disagg.yaml
+0
-3
examples/backends/vllm/deploy/disagg_kvbm.yaml
examples/backends/vllm/deploy/disagg_kvbm.yaml
+0
-3
examples/backends/vllm/deploy/disagg_kvbm_2p2d.yaml
examples/backends/vllm/deploy/disagg_kvbm_2p2d.yaml
+0
-3
examples/backends/vllm/deploy/disagg_kvbm_tp2.yaml
examples/backends/vllm/deploy/disagg_kvbm_tp2.yaml
+0
-3
examples/backends/vllm/deploy/disagg_planner.yaml
examples/backends/vllm/deploy/disagg_planner.yaml
+0
-4
examples/backends/vllm/deploy/disagg_router.yaml
examples/backends/vllm/deploy/disagg_router.yaml
+0
-3
examples/backends/vllm/deploy/lora/agg_lora.yaml
examples/backends/vllm/deploy/lora/agg_lora.yaml
+0
-2
No files found.
examples/backends/sglang/deploy/disagg_planner.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
dynamo
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
my-registry/sglang-runtime:my-tag
image
:
my-registry/sglang-runtime:my-tag
Planner
:
Planner
:
dynamoNamespace
:
dynamo
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
planner
componentType
:
planner
replicas
:
1
replicas
:
1
...
@@ -43,7 +41,6 @@ spec:
...
@@ -43,7 +41,6 @@ spec:
# See docs/planner/sla_planner_quickstart.md for more details
# See docs/planner/sla_planner_quickstart.md for more details
name
:
planner-profile-data
name
:
planner-profile-data
decode
:
decode
:
dynamoNamespace
:
dynamo
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
...
@@ -79,7 +76,6 @@ spec:
...
@@ -79,7 +76,6 @@ spec:
-
--host
-
--host
-
"
0.0.0.0"
-
"
0.0.0.0"
prefill
:
prefill
:
dynamoNamespace
:
dynamo
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
prefill
subComponentType
:
prefill
...
...
examples/backends/trtllm/deploy/agg-with-config.yaml
View file @
c8770464
...
@@ -29,7 +29,6 @@ metadata:
...
@@ -29,7 +29,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-agg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -37,7 +36,6 @@ spec:
...
@@ -37,7 +36,6 @@ spec:
image
:
nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:my-tag
TRTLLMWorker
:
TRTLLMWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
trtllm-agg
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
resources
:
resources
:
...
...
examples/backends/trtllm/deploy/agg.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-agg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -16,7 +15,6 @@ spec:
...
@@ -16,7 +15,6 @@ spec:
image
:
my-registry/tensorrtllm-runtime:my-tag
image
:
my-registry/tensorrtllm-runtime:my-tag
TRTLLMWorker
:
TRTLLMWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
trtllm-agg
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
resources
:
resources
:
...
...
examples/backends/trtllm/deploy/agg_router.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-agg-router
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -19,7 +18,6 @@ spec:
...
@@ -19,7 +18,6 @@ spec:
value
:
kv
value
:
kv
TRTLLMWorker
:
TRTLLMWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
trtllm-agg-router
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
resources
:
resources
:
...
...
examples/backends/trtllm/deploy/disagg-multinode.yaml
View file @
c8770464
...
@@ -93,7 +93,6 @@ spec:
...
@@ -93,7 +93,6 @@ spec:
volumeMounts
:
volumeMounts
:
-
name
:
models
-
name
:
models
mountPoint
:
/models
mountPoint
:
/models
dynamoNamespace
:
trtllm-disagg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -111,7 +110,6 @@ spec:
...
@@ -111,7 +110,6 @@ spec:
volumeMounts
:
volumeMounts
:
-
name
:
models
-
name
:
models
mountPoint
:
/models
mountPoint
:
/models
dynamoNamespace
:
trtllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
@@ -149,7 +147,6 @@ spec:
...
@@ -149,7 +147,6 @@ spec:
volumeMounts
:
volumeMounts
:
-
name
:
models
-
name
:
models
mountPoint
:
/models
mountPoint
:
/models
dynamoNamespace
:
trtllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
...
examples/backends/trtllm/deploy/disagg.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-disagg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
my-registry/tensorrtllm-runtime:my-tag
image
:
my-registry/tensorrtllm-runtime:my-tag
TRTLLMPrefillWorker
:
TRTLLMPrefillWorker
:
dynamoNamespace
:
trtllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
prefill
subComponentType
:
prefill
...
@@ -41,7 +39,6 @@ spec:
...
@@ -41,7 +39,6 @@ spec:
-
--disaggregation-mode
-
--disaggregation-mode
-
prefill
-
prefill
TRTLLMDecodeWorker
:
TRTLLMDecodeWorker
:
dynamoNamespace
:
trtllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
...
...
examples/backends/trtllm/deploy/disagg_planner.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-disagg-planner
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -32,7 +31,6 @@ spec:
...
@@ -32,7 +31,6 @@ spec:
-
"
0.0"
-
"
0.0"
-
--no-kv-events
-
--no-kv-events
Planner
:
Planner
:
dynamoNamespace
:
trtllm-disagg-planner
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
planner
componentType
:
planner
replicas
:
1
replicas
:
1
...
@@ -63,7 +61,6 @@ spec:
...
@@ -63,7 +61,6 @@ spec:
# See docs/planner/sla_planner_quickstart.md for more details
# See docs/planner/sla_planner_quickstart.md for more details
name
:
planner-profile-data
name
:
planner-profile-data
TRTLLMDecodeWorker
:
TRTLLMDecodeWorker
:
dynamoNamespace
:
trtllm-disagg-planner
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
...
@@ -104,7 +101,6 @@ spec:
...
@@ -104,7 +101,6 @@ spec:
-
--disaggregation-mode
-
--disaggregation-mode
-
decode
-
decode
TRTLLMPrefillWorker
:
TRTLLMPrefillWorker
:
dynamoNamespace
:
trtllm-disagg-planner
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
prefill
subComponentType
:
prefill
...
...
examples/backends/trtllm/deploy/disagg_router.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
trtllm-v1-disagg-router
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -18,7 +17,6 @@ spec:
...
@@ -18,7 +17,6 @@ spec:
-
name
:
DYN_ROUTER_MODE
-
name
:
DYN_ROUTER_MODE
value
:
kv
value
:
kv
TRTLLMPrefillWorker
:
TRTLLMPrefillWorker
:
dynamoNamespace
:
trtllm-v1-disagg-router
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
@@ -44,7 +42,6 @@ spec:
...
@@ -44,7 +42,6 @@ spec:
-
prefill
-
prefill
-
--publish-events-and-metrics
-
--publish-events-and-metrics
TRTLLMDecodeWorker
:
TRTLLMDecodeWorker
:
dynamoNamespace
:
trtllm-v1-disagg-router
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
...
examples/backends/vllm/deploy/agg.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-agg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -16,7 +15,6 @@ spec:
...
@@ -16,7 +15,6 @@ spec:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
resources
:
resources
:
...
...
examples/backends/vllm/deploy/agg_kvbm.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-agg-kvbm
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -16,7 +15,6 @@ spec:
...
@@ -16,7 +15,6 @@ spec:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg-kvbm
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
resources
:
resources
:
...
...
examples/backends/vllm/deploy/agg_router.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-agg-router
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -19,7 +18,6 @@ spec:
...
@@ -19,7 +18,6 @@ spec:
value
:
kv
value
:
kv
VllmDecodeWorker
:
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg-router
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
resources
:
resources
:
...
...
examples/backends/vllm/deploy/agg_router_kv_approx.yaml
View file @
c8770464
...
@@ -12,7 +12,6 @@ metadata:
...
@@ -12,7 +12,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-agg-router-kv-approx
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -31,7 +30,6 @@ spec:
...
@@ -31,7 +30,6 @@ spec:
value
:
kv
value
:
kv
VllmDecodeWorker
:
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg-router-kv-approx
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
resources
:
resources
:
...
...
examples/backends/vllm/deploy/disagg-multinode.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -23,7 +22,6 @@ spec:
...
@@ -23,7 +22,6 @@ spec:
-
--http-port
-
--http-port
-
"
8000"
-
"
8000"
decode
:
decode
:
dynamoNamespace
:
vllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
@@ -46,7 +44,6 @@ spec:
...
@@ -46,7 +44,6 @@ spec:
-
--tensor-parallel-size
-
--tensor-parallel-size
-
"
2"
-
"
2"
prefill
:
prefill
:
dynamoNamespace
:
vllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
...
examples/backends/vllm/deploy/disagg.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
...
@@ -36,7 +34,6 @@ spec:
...
@@ -36,7 +34,6 @@ spec:
-
Qwen/Qwen3-0.6B
-
Qwen/Qwen3-0.6B
-
--is-decode-worker
-
--is-decode-worker
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-disagg
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
prefill
subComponentType
:
prefill
...
...
examples/backends/vllm/deploy/disagg_kvbm.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg-kvbm
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-disagg-kvbm
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
@@ -37,7 +35,6 @@ spec:
...
@@ -37,7 +35,6 @@ spec:
-
"
32000"
-
"
32000"
-
--enforce-eager
-
--enforce-eager
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-disagg-kvbm
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
...
examples/backends/vllm/deploy/disagg_kvbm_2p2d.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg-kvbm-2p2d
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-disagg-kvbm-2p2d
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
@@ -37,7 +35,6 @@ spec:
...
@@ -37,7 +35,6 @@ spec:
-
"
32000"
-
"
32000"
-
--enforce-eager
-
--enforce-eager
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-disagg-kvbm-2p2d
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
...
examples/backends/vllm/deploy/disagg_kvbm_tp2.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg-kvbm-tp2
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-disagg-kvbm-tp2
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
@@ -43,7 +41,6 @@ spec:
...
@@ -43,7 +41,6 @@ spec:
-
--tensor-parallel-size
-
--tensor-parallel-size
-
"
2"
-
"
2"
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-disagg-kvbm-tp2
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
1
replicas
:
1
...
...
examples/backends/vllm/deploy/disagg_planner.yaml
View file @
c8770464
...
@@ -8,14 +8,12 @@ metadata:
...
@@ -8,14 +8,12 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-disagg-planner
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
mainContainer
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
Planner
:
Planner
:
dynamoNamespace
:
vllm-disagg-planner
componentType
:
planner
componentType
:
planner
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -42,7 +40,6 @@ spec:
...
@@ -42,7 +40,6 @@ spec:
# See docs/planner/sla_planner_quickstart.md for more details
# See docs/planner/sla_planner_quickstart.md for more details
name
:
planner-profile-data
name
:
planner-profile-data
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-disagg-planner
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
...
@@ -62,7 +59,6 @@ spec:
...
@@ -62,7 +59,6 @@ spec:
-
--model
-
--model
-
Qwen/Qwen3-0.6B
-
Qwen/Qwen3-0.6B
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-disagg-planner
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
subComponentType
:
prefill
subComponentType
:
prefill
...
...
examples/backends/vllm/deploy/disagg_router.yaml
View file @
c8770464
...
@@ -8,7 +8,6 @@ metadata:
...
@@ -8,7 +8,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-v1-disagg-router
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -18,7 +17,6 @@ spec:
...
@@ -18,7 +17,6 @@ spec:
-
name
:
DYN_ROUTER_MODE
-
name
:
DYN_ROUTER_MODE
value
:
kv
value
:
kv
VllmDecodeWorker
:
VllmDecodeWorker
:
dynamoNamespace
:
vllm-v1-disagg-router
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
@@ -38,7 +36,6 @@ spec:
...
@@ -38,7 +36,6 @@ spec:
-
Qwen/Qwen3-0.6B
-
Qwen/Qwen3-0.6B
-
--is-decode-worker
-
--is-decode-worker
VllmPrefillWorker
:
VllmPrefillWorker
:
dynamoNamespace
:
vllm-v1-disagg-router
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
componentType
:
worker
componentType
:
worker
replicas
:
2
replicas
:
2
...
...
examples/backends/vllm/deploy/lora/agg_lora.yaml
View file @
c8770464
...
@@ -7,7 +7,6 @@ metadata:
...
@@ -7,7 +7,6 @@ metadata:
spec
:
spec
:
services
:
services
:
Frontend
:
Frontend
:
dynamoNamespace
:
vllm-agg-lora
componentType
:
frontend
componentType
:
frontend
replicas
:
1
replicas
:
1
extraPodSpec
:
extraPodSpec
:
...
@@ -15,7 +14,6 @@ spec:
...
@@ -15,7 +14,6 @@ spec:
image
:
nvcr.io/nvidian/dynamo-dev/biswa:7e499b5c460f1883a9945d221123e0760051210f-39500608-vllm-amd64
image
:
nvcr.io/nvidian/dynamo-dev/biswa:7e499b5c460f1883a9945d221123e0760051210f-39500608-vllm-amd64
VllmDecodeWorker
:
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg-lora
componentType
:
worker
componentType
:
worker
subComponentType
:
decode
subComponentType
:
decode
replicas
:
1
replicas
:
1
...
...
Prev
1
2
3
4
Next
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment