buildkit: deploy on meta-prod-aws-ue1 (us-east-1)

Same buildkit config as arc-cbr-production (ue2): m6id.24xlarge amd64 (2/node)
+ m7gd.16xlarge arm64 (4/node), autoscaling min 2/4, max 360/30, fallback 32/8.
Adds keda + buildkit to the cluster's module list (keda provides the CRDs the
autoscaling needs).

ghstack-source-id: 47a94e8
Pull-Request: #728

Author: huydhn

osdc/clusters.yaml

@@ -255,6 +255,19 @@ clusters:
       central_memory_limit: "50Gi"
     node_compactor:
       min_node_age_seconds: 900
+    buildkit:
+      amd64_instance_type: m6id.24xlarge
+      amd64_pods_per_node: 2
+      arm64_instance_type: m7gd.16xlarge
+      arm64_pods_per_node: 4
+      autoscaling:
+        enabled: true
+        amd64_min: 2   # 1x m6id.24xlarge (2 pods/node)
+        amd64_max: 360  # ~90d peak ≈180, x2 for headroom
+        arm64_min: 4   # 1x m7gd.16xlarge (4 pods/node)
+        arm64_max: 30  # ~90d peak ≈15, x2 for headroom
+        amd64_fallback: 32  # if KEDA can't read metrics, hold the proven fixed pool
+        arm64_fallback: 8
     arc-runners:
       github_config_url: "https://github.com/pytorch"
       github_secret_name: meta-prod-aws-ue1
@@ -267,6 +280,8 @@ clusters:
       - arc
       - nodepools
       - arc-runners
+      - keda
+      - buildkit
       - pypi-cache
       - cache-enforcer
       - zombie-cleanup
@@ -299,9 +314,9 @@ clusters:
       arm64_pods_per_node: 4
       autoscaling:
         enabled: true
-        amd64_min: 32  # warm baseline = proven fixed pool (16x m6id.24xlarge)
+        amd64_min: 2   # 1x m6id.24xlarge (2 pods/node)
         amd64_max: 360  # ~90d peak ≈180, x2 for headroom
-        arm64_min: 8   # warm baseline = proven fixed pool (2x m7gd.16xlarge)
+        arm64_min: 4   # 1x m7gd.16xlarge (4 pods/node)
         arm64_max: 30  # ~90d peak ≈15, x2 for headroom
         amd64_fallback: 32  # if KEDA can't read metrics, hold the proven fixed pool
         arm64_fallback: 8