Merge pull request #1303 from NVIDIA/gb200-hgx-mig

[MIG] add support for HGX GB200

Author: tariq1890

assets/state-mig-manager/0400_configmap.yaml

@@ -97,6 +97,13 @@ data:
           mig-devices:
             "1g.23gb+me": 1
 
+      # GB200 HGX
+      all-1g.24gb.me:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "1g.24gb+me": 1
+
       all-2g.20gb:
         - devices: all
           mig-enabled: true
@@ -166,7 +173,16 @@ data:
             "1g.12gb+me": 1
 
       all-1g.24gb:
-        - devices: all
+        # GB200 HGX
+        - device-filter: ["0x294110DE"]
+          devices: all
+          mig-enabled: true
+          mig-devices:
+            "1g.24gb": 7
+
+        # H100 NVL 94GB
+        - device-filter: ["0x232110DE"]
+          devices: all
           mig-enabled: true
           mig-devices:
             "1g.24gb": 4
@@ -177,6 +193,12 @@ data:
           mig-devices:
             "1g.45gb": 4
 
+      all-1g.47gb:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "1g.47gb": 4
+
       all-2g.24gb:
         - devices: all
           mig-enabled: true
@@ -188,7 +210,13 @@ data:
           mig-enabled: true
           mig-devices:
             "2g.45gb": 3
-            
+
+      all-2g.47gb:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "2g.47gb": 3
+
       # H100 NVL, H800 NVL
       all-3g.47gb:
         - devices: all
@@ -221,12 +249,24 @@ data:
           mig-devices:
             "3g.90gb": 2
 
+      all-3g.95gb:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "3g.95gb": 2
+
       all-4g.48gb:
         - devices: all
           mig-enabled: true
           mig-devices:
             "4g.48gb": 1
 
+      all-4g.95gb:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "4g.95gb": 1
+
       all-7g.96gb:
         - devices: all
           mig-enabled: true
@@ -239,8 +279,23 @@ data:
           mig-devices:
             "7g.180gb": 1
 
-      # B200, GH200 144G HBM3e, H200-141GB, H200 NVL, H100-96GB, GH200, H100 NVL, H800 NVL, H100-80GB, H800-80GB, A800-40GB, A800-80GB, A100-40GB, A100-80GB, A30-24GB, PG506-96GB
+      all-7g.189gb:
+        - devices: all
+          mig-enabled: true
+          mig-devices:
+            "7g.189gb": 1
+
+      # GB200 HGX, B200, GH200 144G HBM3e, H200-141GB, H200 NVL, H100-96GB, GH200, H100 NVL, H800 NVL, H100-80GB, H800-80GB, A800-40GB, A800-80GB, A100-40GB, A100-80GB, A30-24GB, PG506-96GB
       all-balanced:
+        # GB200 HGX
+        - device-filter: ["0x294110DE"]
+          devices: all
+          mig-enabled: true
+          mig-devices:
+            "1g.24gb": 2
+            "2g.47gb": 1
+            "3g.95gb": 1
+
         # B200
         - device-filter: ["0x290110DE"]
           devices: all