nodeutilization: separate code responsible for requested resource extraction into a dedicated usage client

Turning a usage client into an interface allows to implement other kinds
of usage clients like actual usage or prometheus based resource
collection.

pkg/framework/plugins/nodeutilization/highnodeutilization.go

@@ -44,6 +44,7 @@ type HighNodeUtilization struct {
 	underutilizationCriteria []interface{}
 	resourceNames            []v1.ResourceName
 	targetThresholds         api.ResourceThresholds
+	usageClient              usageClient
 }
 
 var _ frameworktypes.BalancePlugin = &HighNodeUtilization{}
@@ -84,6 +85,7 @@ func NewHighNodeUtilization(args runtime.Object, handle frameworktypes.Handle) (
 		targetThresholds:         targetThresholds,
 		underutilizationCriteria: underutilizationCriteria,
 		podFilter:                podFilter,
+		usageClient:              newRequestedUsageClient(resourceNames, handle.GetPodsAssignedToNodeFunc()),
 	}, nil
 }
 
@@ -94,22 +96,15 @@ func (h *HighNodeUtilization) Name() string {
 
 // Balance extension point implementation for the plugin
 func (h *HighNodeUtilization) Balance(ctx context.Context, nodes []*v1.Node) *frameworktypes.Status {
-	nodeUsage, err := getNodeUsage(nodes, h.resourceNames, h.handle.GetPodsAssignedToNodeFunc())
-	if err != nil {
+	if err := h.usageClient.sync(nodes); err != nil {
 		return &frameworktypes.Status{
 			Err: fmt.Errorf("error getting node usage: %v", err),
 		}
 	}
-	thresholds, err := getNodeThresholds(nodes, h.args.Thresholds, h.targetThresholds, h.resourceNames, h.handle.GetPodsAssignedToNodeFunc(), false)
-	if err != nil {
-		return &frameworktypes.Status{
-			Err: fmt.Errorf("error getting node thresholds: %v", err),
-		}
-	}
 
 	sourceNodes, highNodes := classifyNodes(
-		nodeUsage,
-		thresholds,
+		getNodeUsage(nodes, h.usageClient),
+		getNodeThresholds(nodes, h.args.Thresholds, h.targetThresholds, h.resourceNames, false, h.usageClient),
 		func(node *v1.Node, usage NodeUsage, threshold NodeThresholds) bool {
 			return isNodeWithLowUtilization(usage, threshold.lowResourceThreshold)
 		},
@@ -165,7 +160,9 @@ func (h *HighNodeUtilization) Balance(ctx context.Context, nodes []*v1.Node) *fr
 		evictions.EvictOptions{StrategyName: HighNodeUtilizationPluginName},
 		h.podFilter,
 		h.resourceNames,
-		continueEvictionCond)
+		continueEvictionCond,
+		h.usageClient,
+	)
 
 	return nil
 }

pkg/framework/plugins/nodeutilization/lownodeutilization.go

@@ -43,6 +43,7 @@ type LowNodeUtilization struct {
 	underutilizationCriteria []interface{}
 	overutilizationCriteria  []interface{}
 	resourceNames            []v1.ResourceName
+	usageClient              usageClient
 }
 
 var _ frameworktypes.BalancePlugin = &LowNodeUtilization{}
@@ -85,13 +86,16 @@ func NewLowNodeUtilization(args runtime.Object, handle frameworktypes.Handle) (f
 		return nil, fmt.Errorf("error initializing pod filter function: %v", err)
 	}
 
+	resourceNames := getResourceNames(lowNodeUtilizationArgsArgs.Thresholds)
+
 	return &LowNodeUtilization{
 		handle:                   handle,
 		args:                     lowNodeUtilizationArgsArgs,
 		underutilizationCriteria: underutilizationCriteria,
 		overutilizationCriteria:  overutilizationCriteria,
-		resourceNames:            getResourceNames(lowNodeUtilizationArgsArgs.Thresholds),
+		resourceNames:            resourceNames,
 		podFilter:                podFilter,
+		usageClient:              newRequestedUsageClient(resourceNames, handle.GetPodsAssignedToNodeFunc()),
 	}, nil
 }
 
@@ -102,22 +106,15 @@ func (l *LowNodeUtilization) Name() string {
 
 // Balance extension point implementation for the plugin
 func (l *LowNodeUtilization) Balance(ctx context.Context, nodes []*v1.Node) *frameworktypes.Status {
-	nodeUsage, err := getNodeUsage(nodes, l.resourceNames, l.handle.GetPodsAssignedToNodeFunc())
-	if err != nil {
+	if err := l.usageClient.sync(nodes); err != nil {
 		return &frameworktypes.Status{
 			Err: fmt.Errorf("error getting node usage: %v", err),
 		}
 	}
-	thresholds, err := getNodeThresholds(nodes, l.args.Thresholds, l.args.TargetThresholds, l.resourceNames, l.handle.GetPodsAssignedToNodeFunc(), l.args.UseDeviationThresholds)
-	if err != nil {
-		return &frameworktypes.Status{
-			Err: fmt.Errorf("error getting node thresholds: %v", err),
-		}
-	}
 
 	lowNodes, sourceNodes := classifyNodes(
-		nodeUsage,
-		thresholds,
+		getNodeUsage(nodes, l.usageClient),
+		getNodeThresholds(nodes, l.args.Thresholds, l.args.TargetThresholds, l.resourceNames, l.args.UseDeviationThresholds, l.usageClient),
 		// The node has to be schedulable (to be able to move workload there)
 		func(node *v1.Node, usage NodeUsage, threshold NodeThresholds) bool {
 			if nodeutil.IsNodeUnschedulable(node) {
@@ -185,7 +182,9 @@ func (l *LowNodeUtilization) Balance(ctx context.Context, nodes []*v1.Node) *fra
 		evictions.EvictOptions{StrategyName: LowNodeUtilizationPluginName},
 		l.podFilter,
 		l.resourceNames,
-		continueEvictionCond)
+		continueEvictionCond,
+		l.usageClient,
+	)
 
 	return nil
 }

pkg/framework/plugins/nodeutilization/nodeutilization.go

@@ -28,7 +28,6 @@ import (
 	"k8s.io/apimachinery/pkg/util/sets"
 	"k8s.io/klog/v2"
 	"sigs.k8s.io/descheduler/pkg/descheduler/evictions"
-	"sigs.k8s.io/descheduler/pkg/descheduler/node"
 	nodeutil "sigs.k8s.io/descheduler/pkg/descheduler/node"
 	podutil "sigs.k8s.io/descheduler/pkg/descheduler/pod"
 	frameworktypes "sigs.k8s.io/descheduler/pkg/framework/types"
@@ -78,18 +77,14 @@ func getNodeThresholds(
 	nodes []*v1.Node,
 	lowThreshold, highThreshold api.ResourceThresholds,
 	resourceNames []v1.ResourceName,
-	getPodsAssignedToNode podutil.GetPodsAssignedToNodeFunc,
 	useDeviationThresholds bool,
-) (map[string]NodeThresholds, error) {
+	usageClient usageClient,
+) map[string]NodeThresholds {
 	nodeThresholdsMap := map[string]NodeThresholds{}
 
 	averageResourceUsagePercent := api.ResourceThresholds{}
 	if useDeviationThresholds {
-		usage, err := averageNodeBasicresources(nodes, getPodsAssignedToNode, resourceNames)
-		if err != nil {
-			return nil, err
-		}
-		averageResourceUsagePercent = usage
+		averageResourceUsagePercent = averageNodeBasicresources(nodes, usageClient)
 	}
 
 	for _, node := range nodes {
@@ -120,39 +115,24 @@ func getNodeThresholds(
 		}
 
 	}
-	return nodeThresholdsMap, nil
+	return nodeThresholdsMap
 }
 
 func getNodeUsage(
 	nodes []*v1.Node,
-	resourceNames []v1.ResourceName,
-	getPodsAssignedToNode podutil.GetPodsAssignedToNodeFunc,
-) ([]NodeUsage, error) {
+	usageClient usageClient,
+) []NodeUsage {
 	var nodeUsageList []NodeUsage
 
 	for _, node := range nodes {
-		pods, err := podutil.ListPodsOnANode(node.Name, getPodsAssignedToNode, nil)
-		if err != nil {
-			klog.V(2).InfoS("Node will not be processed, error accessing its pods", "node", klog.KObj(node), "err", err)
-			continue
-		}
-
-		nodeUsage, err := nodeutil.NodeUtilization(pods, resourceNames, func(pod *v1.Pod) (v1.ResourceList, error) {
-			req, _ := utils.PodRequestsAndLimits(pod)
-			return req, nil
-		})
-		if err != nil {
-			return nil, err
-		}
-
 		nodeUsageList = append(nodeUsageList, NodeUsage{
 			node:    node,
-			usage:   nodeUsage,
-			allPods: pods,
+			usage:   usageClient.nodeUtilization(node.Name),
+			allPods: usageClient.pods(node.Name),
 		})
 	}
 
-	return nodeUsageList, nil
+	return nodeUsageList
 }
 
 func resourceThreshold(nodeCapacity v1.ResourceList, resourceName v1.ResourceName, threshold api.Percentage) *resource.Quantity {
@@ -239,7 +219,7 @@ func usageToKeysAndValues(usage map[v1.ResourceName]*resource.Quantity) []interf
 		keysAndValues = append(keysAndValues, "Pods", quantity.Value())
 	}
 	for name := range usage {
-		if !node.IsBasicResource(name) {
+		if !nodeutil.IsBasicResource(name) {
 			keysAndValues = append(keysAndValues, string(name), usage[name].Value())
 		}
 	}
@@ -258,6 +238,7 @@ func evictPodsFromSourceNodes(
 	podFilter func(pod *v1.Pod) bool,
 	resourceNames []v1.ResourceName,
 	continueEviction continueEvictionCond,
+	usageClient usageClient,
 ) {
 	// upper bound on total number of pods/cpu/memory and optional extended resources to be moved
 	totalAvailableUsage := map[v1.ResourceName]*resource.Quantity{}
@@ -270,6 +251,10 @@ func evictPodsFromSourceNodes(
 		taintsOfDestinationNodes[node.node.Name] = node.node.Spec.Taints
 
 		for _, name := range resourceNames {
+			if _, exists := node.usage[name]; !exists {
+				klog.Errorf("unable to find %q resource in node's %q usage, terminating eviction", name, node.node.Name)
+				return
+			}
 			if _, ok := totalAvailableUsage[name]; !ok {
 				totalAvailableUsage[name] = resource.NewQuantity(0, resource.DecimalSI)
 			}
@@ -295,7 +280,7 @@ func evictPodsFromSourceNodes(
 		klog.V(1).InfoS("Evicting pods based on priority, if they have same priority, they'll be evicted based on QoS tiers")
 		// sort the evictable Pods based on priority. This also sorts them based on QoS. If there are multiple pods with same priority, they are sorted based on QoS tiers.
 		podutil.SortPodsBasedOnPriorityLowToHigh(removablePods)
-		err := evictPods(ctx, evictableNamespaces, removablePods, node, totalAvailableUsage, taintsOfDestinationNodes, podEvictor, evictOptions, continueEviction)
+		err := evictPods(ctx, evictableNamespaces, removablePods, node, totalAvailableUsage, taintsOfDestinationNodes, podEvictor, evictOptions, continueEviction, usageClient)
 		if err != nil {
 			switch err.(type) {
 			case *evictions.EvictionTotalLimitError:
@@ -316,6 +301,7 @@ func evictPods(
 	podEvictor frameworktypes.Evictor,
 	evictOptions evictions.EvictOptions,
 	continueEviction continueEvictionCond,
+	usageClient usageClient,
 ) error {
 	var excludedNamespaces sets.Set[string]
 	if evictableNamespaces != nil {
@@ -341,6 +327,11 @@ func evictPods(
 			if !preEvictionFilterWithOptions(pod) {
 				continue
 			}
+			podUsage, err := usageClient.podUsage(pod)
+			if err != nil {
+				klog.Errorf("unable to get pod usage for %v/%v: %v", pod.Namespace, pod.Name, err)
+				continue
+			}
 			err = podEvictor.Evict(ctx, pod, evictOptions)
 			if err == nil {
 				klog.V(3).InfoS("Evicted pods", "pod", klog.KObj(pod))
@@ -350,9 +341,8 @@ func evictPods(
 						nodeInfo.usage[name].Sub(*resource.NewQuantity(1, resource.DecimalSI))
 						totalAvailableUsage[name].Sub(*resource.NewQuantity(1, resource.DecimalSI))
 					} else {
-						quantity := utils.GetResourceRequestQuantity(pod, name)
-						nodeInfo.usage[name].Sub(quantity)
-						totalAvailableUsage[name].Sub(quantity)
+						nodeInfo.usage[name].Sub(*podUsage[name])
+						totalAvailableUsage[name].Sub(*podUsage[name])
 					}
 				}
 
@@ -456,24 +446,12 @@ func classifyPods(pods []*v1.Pod, filter func(pod *v1.Pod) bool) ([]*v1.Pod, []*
 	return nonRemovablePods, removablePods
 }
 
-func averageNodeBasicresources(nodes []*v1.Node, getPodsAssignedToNode podutil.GetPodsAssignedToNodeFunc, resourceNames []v1.ResourceName) (api.ResourceThresholds, error) {
+func averageNodeBasicresources(nodes []*v1.Node, usageClient usageClient) api.ResourceThresholds {
 	total := api.ResourceThresholds{}
 	average := api.ResourceThresholds{}
 	numberOfNodes := len(nodes)
 	for _, node := range nodes {
-		pods, err := podutil.ListPodsOnANode(node.Name, getPodsAssignedToNode, nil)
-		if err != nil {
-			numberOfNodes--
-			continue
-		}
-		usage, err := nodeutil.NodeUtilization(pods, resourceNames, func(pod *v1.Pod) (v1.ResourceList, error) {
-			req, _ := utils.PodRequestsAndLimits(pod)
-			return req, nil
-		})
-		if err != nil {
-			return nil, err
-		}
-
+		usage := usageClient.nodeUtilization(node.Name)
 		nodeCapacity := node.Status.Capacity
 		if len(node.Status.Allocatable) > 0 {
 			nodeCapacity = node.Status.Allocatable
@@ -490,5 +468,5 @@ func averageNodeBasicresources(nodes []*v1.Node, getPodsAssignedToNode podutil.G
 	for resource, value := range total {
 		average[resource] = value / api.Percentage(numberOfNodes)
 	}
-	return average, nil
+	return average
 }

pkg/framework/plugins/nodeutilization/usageclients.go

@@ -0,0 +1,102 @@
+/*
+Copyright 2024 The Kubernetes Authors.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package nodeutilization
+
+import (
+	"fmt"
+
+	v1 "k8s.io/api/core/v1"
+	"k8s.io/apimachinery/pkg/api/resource"
+	"k8s.io/klog/v2"
+	utilptr "k8s.io/utils/ptr"
+	nodeutil "sigs.k8s.io/descheduler/pkg/descheduler/node"
+	podutil "sigs.k8s.io/descheduler/pkg/descheduler/pod"
+	"sigs.k8s.io/descheduler/pkg/utils"
+)
+
+type usageClient interface {
+	// Both low/high node utilization plugins are expected to invoke sync right
+	// after Balance method is invoked. There's no cache invalidation so each
+	// Balance is expected to get the latest data by invoking sync.
+	sync(nodes []*v1.Node) error
+	nodeUtilization(node string) map[v1.ResourceName]*resource.Quantity
+	pods(node string) []*v1.Pod
+	podUsage(pod *v1.Pod) (map[v1.ResourceName]*resource.Quantity, error)
+}
+
+type requestedUsageClient struct {
+	resourceNames         []v1.ResourceName
+	getPodsAssignedToNode podutil.GetPodsAssignedToNodeFunc
+
+	_pods            map[string][]*v1.Pod
+	_nodeUtilization map[string]map[v1.ResourceName]*resource.Quantity
+}
+
+var _ usageClient = &requestedUsageClient{}
+
+func newRequestedUsageClient(
+	resourceNames []v1.ResourceName,
+	getPodsAssignedToNode podutil.GetPodsAssignedToNodeFunc,
+) *requestedUsageClient {
+	return &requestedUsageClient{
+		resourceNames:         resourceNames,
+		getPodsAssignedToNode: getPodsAssignedToNode,
+	}
+}
+
+func (s *requestedUsageClient) nodeUtilization(node string) map[v1.ResourceName]*resource.Quantity {
+	return s._nodeUtilization[node]
+}
+
+func (s *requestedUsageClient) pods(node string) []*v1.Pod {
+	return s._pods[node]
+}
+
+func (s *requestedUsageClient) podUsage(pod *v1.Pod) (map[v1.ResourceName]*resource.Quantity, error) {
+	usage := make(map[v1.ResourceName]*resource.Quantity)
+	for _, resourceName := range s.resourceNames {
+		usage[resourceName] = utilptr.To[resource.Quantity](utils.GetResourceRequestQuantity(pod, resourceName).DeepCopy())
+	}
+	return usage, nil
+}
+
+func (s *requestedUsageClient) sync(nodes []*v1.Node) error {
+	s._nodeUtilization = make(map[string]map[v1.ResourceName]*resource.Quantity)
+	s._pods = make(map[string][]*v1.Pod)
+
+	for _, node := range nodes {
+		pods, err := podutil.ListPodsOnANode(node.Name, s.getPodsAssignedToNode, nil)
+		if err != nil {
+			klog.V(2).InfoS("Node will not be processed, error accessing its pods", "node", klog.KObj(node), "err", err)
+			return fmt.Errorf("error accessing %q node's pods: %v", node.Name, err)
+		}
+
+		nodeUsage, err := nodeutil.NodeUtilization(pods, s.resourceNames, func(pod *v1.Pod) (v1.ResourceList, error) {
+			req, _ := utils.PodRequestsAndLimits(pod)
+			return req, nil
+		})
+		if err != nil {
+			return err
+		}
+
+		// store the snapshot of pods from the same (or the closest) node utilization computation
+		s._pods[node.Name] = pods
+		s._nodeUtilization[node.Name] = nodeUsage
+	}
+
+	return nil
+}