feat: add option to select k8s nodes for topology config

Signed-off-by: Dmitry Shmulevich <[email protected]>

Author: dmitsh

charts/topograph/Chart.yaml

@@ -15,7 +15,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.1.0
+version: 0.2.0
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
@@ -25,8 +25,8 @@ appVersion: "1.16.0"
 
 dependencies:
   - name: node-data-broker
-    version: 0.1.0
+    version: 0.2.0
     repository: "file://charts/node-data-broker"
   - name: node-observer
-    version: 0.1.0
+    version: 0.2.0
     repository: "file://charts/node-observer"

charts/topograph/charts/node-data-broker/Chart.yaml

@@ -15,7 +15,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.1.0
+version: 0.2.0
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to

charts/topograph/charts/node-observer/Chart.yaml

@@ -15,7 +15,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.1.0
+version: 0.2.0
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to

charts/topograph/charts/node-observer/templates/configmap.yml

@@ -7,7 +7,9 @@ metadata:
 data:
   node-observer-config.yaml: |-
     generateTopologyUrl: "{{ include "topograph.url" $ }}/v1/generate"
-    params:
-      {{- toYaml .Values.global.engineParams | nindent 6 }}
+    provider:
+      {{- toYaml .Values.global.provider | nindent 6 }}
+    engine:
+      {{- toYaml .Values.global.engine | nindent 6 }}
     trigger:
       {{- toYaml .Values.topograph.trigger | nindent 6 }}

charts/topograph/templates/configmap.yml

@@ -9,8 +9,6 @@ data:
     http:
       port: {{ .Values.global.service.port }}
       ssl: false
-    provider: {{ .Values.global.provider }}
-    engine: {{ .Values.global.engine }}
     requestAggregationDelay: {{ .Values.config.requestAggregationDelay }}
     {{- if .Values.config.credentialsSecretName }}
     credentialsPath: /etc/topograph/credentials/credentials.yaml

charts/topograph/values-slinky-block-example.yaml

@@ -3,19 +3,25 @@
 # Declare variables to be passed into your templates.
 
 global:
-  # provider: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
-  provider: aws
-  #  engine: "k8s" or "slinky"
-  engine: slinky
-  engineParams:
-    namespace: slurm
-    podSelector:
-      matchLabels:
-        app.kubernetes.io/component: compute
-    plugin: topology/block
-    block_sizes: 4
-    topologyConfigPath: topology.conf
-    topologyConfigmapName: slurm-config
+  provider:
+    # name: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
+    name: aws
+    params:
+      nodeSelector:
+        slurmCluster: my-cluster
+  engine:
+    name: slinky
+    params:
+      namespace: slurm
+      nodeSelector:
+        slurmCluster: my-cluster
+      podSelector:
+        matchLabels:
+          app.kubernetes.io/component: compute
+      plugin: topology/block
+      block_sizes: 4
+      topologyConfigPath: topology.conf
+      topologyConfigmapName: slurm-config
 
 nodeSelector:
   dedicated: user-workload

charts/topograph/values-slinky-partition-example.yaml

@@ -3,29 +3,30 @@
 # Declare variables to be passed into your templates.
 
 global:
-  # provider: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
-  provider: aws
-  #  engine: "k8s" or "slinky"
-  engine: slinky
-  engineParams:
-    namespace: slurm
-    podSelector:
-      matchLabels:
-        app.kubernetes.io/component: compute
-    topologies:
-      topo1:
-        plugin: topology/block
-        blockSizes: [2,4]
-      topo2:
-        plugin: topology/block
-        blockSizes: [8,16]
-      topo3:
-        plugin: topology/tree
-      topo-default:
-        plugin: topology/flat
-        clusterDefault: true
-    topologyConfigPath: topology.conf
-    topologyConfigmapName: slurm-config
+  provider:
+    # name: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
+    name: aws
+  engine:
+    name: slinky
+    params:
+      namespace: slurm
+      podSelector:
+        matchLabels:
+          app.kubernetes.io/component: compute
+      topologies:
+        topo1:
+          plugin: topology/block
+          blockSizes: [2,4]
+        topo2:
+          plugin: topology/block
+          blockSizes: [8,16]
+        topo3:
+          plugin: topology/tree
+        topo-default:
+          plugin: topology/flat
+          clusterDefault: true
+      topologyConfigPath: topology.conf
+      topologyConfigmapName: slurm-config
 
 nodeSelector:
   dedicated: user-workload

charts/topograph/values-slinky-tree-example.yaml

@@ -3,18 +3,19 @@
 # Declare variables to be passed into your templates.
 
 global:
-  # provider: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
-  provider: aws
-  #  engine: "k8s" or "slinky"
-  engine: slinky
-  engineParams:
-    namespace: slurm
-    podSelector:
-      matchLabels:
-        app.kubernetes.io/component: compute
-    plugin: topology/tree
-    topologyConfigPath: topology.conf
-    topologyConfigmapName: slurm-config
+  provider:
+    # name: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test"
+    name: aws
+  engine:
+    name: slinky
+    params:
+      namespace: slurm
+      podSelector:
+        matchLabels:
+          app.kubernetes.io/component: compute
+      plugin: topology/tree
+      topologyConfigPath: topology.conf
+      topologyConfigmapName: slurm-config
 
 nodeSelector:
   dedicated: user-workload

charts/topograph/values.yaml

@@ -3,11 +3,12 @@
 # Declare variables to be passed into your templates.
 
 global:
-  # provider: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test".
-  provider: test
-  #  engine: "k8s" or "slinky"
-  engine: k8s
-  # engineParams:
+  provider:
+    # name: "aws", "oci", "gcp", "nebius", "netq", "infiniband-k8s", "dra" or "test".
+    name: test
+  engine:
+    # name: "k8s" or "slinky"
+    name: k8s
 
   service:
     type: ClusterIP

internal/k8s/utils.go

@@ -19,8 +19,12 @@ import (
 	"k8s.io/client-go/tools/remotecommand"
 )
 
-func GetNodes(ctx context.Context, client *kubernetes.Clientset) (*corev1.NodeList, error) {
-	nodes, err := client.CoreV1().Nodes().List(ctx, metav1.ListOptions{})
+func GetNodes(ctx context.Context, client *kubernetes.Clientset, opt *metav1.ListOptions) (*corev1.NodeList, error) {
+	if opt == nil {
+		opt = &metav1.ListOptions{}
+	}
+
+	nodes, err := client.CoreV1().Nodes().List(ctx, *opt)
 	if err != nil {
 		return nil, fmt.Errorf("failed to list node in the cluster: %v", err)
 	}