Pablotrivino/karpentertest

pipelines/perf-eval/Autoscale Benchmark/node-auto-provisioning-benchmark.yml

@@ -125,6 +125,39 @@ stages:
           credential_type: service_connection
           ssh_key_enabled: false
 
+  - stage: azure_australiaeast_small_scale_on_demand
+    condition: eq(variables['Build.Reason'], 'Manual')
+    dependsOn: []
+    jobs:
+      - template: /jobs/competitive-test.yml
+        parameters:
+          cloud: azure
+          regions:
+            - australiaeast
+          engine: clusterloader2
+          engine_input:
+            image: "ghcr.io/azure/clusterloader2:v20250423"
+          topology: karpenter
+          matrix:
+            small-scale-on-demand:
+              cpu_per_node: 4
+              node_count: 10
+              pod_count: 100
+              scale_up_timeout: "15m"
+              scale_down_timeout: "15m"
+              node_label_selector: "karpenter.sh/nodepool = default"
+              node_selector: "{karpenter.sh/nodepool: default}"
+              loop_count: 5
+              karpenter_cooldown: 900 #seconds
+              warmup_deployment: true
+              warmup_deployment_template: warmup_deployment.yaml
+              vm_size: Standard_D4ds_v4
+              capacity_type: on-demand
+          max_parallel: 1
+          timeout_in_minutes: 560
+          credential_type: service_connection
+          ssh_key_enabled: false
+
   - stage: azure_australiaeast_small_scale
     condition: |
         or(

pipelines/system/new-pipeline-test.yml

@@ -1,25 +1,49 @@
 trigger: none
-
+schedules:
+  # canadacentral - daily
+  - cron: "0 21 * * *"
+    displayName: "Every day at 9:00 PM"
+    branches:
+      include:
+        - main
+    always: true
 variables:
-  SCENARIO_TYPE: <scenario-type>
-  SCENARIO_NAME: <scenario-name>
+  SCENARIO_TYPE: perf-eval
+  SCENARIO_NAME: nap-complex
 
 stages:
-  - stage: <stage-name> # format: <cloud>[_<region>]+ (e.g. azure_eastus2, aws_eastus_westus)
+  - stage: azure_canadacentral_ondemand_complex_config
     dependsOn: []
+    variables:
+      - group: NAP-MS-CanadaCentral
     jobs:
-      - template: /jobs/competitive-test.yml # must keep as is
+      - template: /jobs/competitive-test.yml
         parameters:
-          cloud: <cloud> # e.g. azure, aws
-          regions: # list of regions
-            - region1 # e.g. eastus2
-          topology: <topology> # e.g. cluster-autoscaler
-          engine: <engine> # e.g. clusterloader2
-          matrix: # list of test parameters to customize the provisioned resources
-            <case-name>:
-              <key1>: <value1>
-              <key2>: <value2>
-          max_parallel: <number of concurrent jobs> # required
-          credential_type: service_connection # required
-          ssh_key_enabled: false
-          timeout_in_minutes: 60 # if not specified, default is 60
+          cloud: azure
+          regions:
+            - canadacentral
+          engine: clusterloader2
+          engine_input:
+            image: "ghcr.io/azure/clusterloader2:v20250912"
+          topology: karpenter-private-cluster
+          matrix:
+            complex-nap:
+              pod_count: 5000
+              pod_cpu_request: 16
+              pod_memory_request: "60Gi"
+              scale_up_timeout: "900s" # 15 minutes
+              scale_down_timeout: "900s"
+              node_selector: "{karpenter.sh/nodepool: default}"
+              loop_count: 1
+              warmup_deployment: true
+              warmup_deployment_template: warmup_deployment.yaml
+              capacity_type: on-demand
+              cl2_config_file: "ms_complex_config.yaml"
+              enable_prometheus: True
+              scrape_kubelets: True
+              scrape_ksm: True
+              enable_apiserver_vnet_integration: "true"
+          max_parallel: 1
+          timeout_in_minutes: 120
+          credential_type: service_connection
+          ssh_key_enabled: true

steps/execute-tests.yml

@@ -13,6 +13,11 @@ parameters:
   default: {}
 
 steps:
+- script: |
+    echo "Waiting 20 minutes before starting test execution..."
+    sleep 1200  # 20 minutes = 1200 seconds
+    echo "Wait period completed, proceeding with tests"
+  displayName: "Pre-test wait period"
 - template: /steps/topology/${{ parameters.topology }}/execute-${{ parameters.engine }}.yml@self
   parameters:
     cloud: ${{ parameters.cloud }}