Merge pull request #81068 from Huang-Wei/eps-structure-optimize

scheduler: internal data structure optimization

pkg/scheduler/algorithm/predicates/metadata.go

@@ -39,7 +39,7 @@ import (
 type PredicateMetadata interface {
 	ShallowCopy() PredicateMetadata
 	AddPod(addedPod *v1.Pod, nodeInfo *schedulernodeinfo.NodeInfo) error
-	RemovePod(deletedPod *v1.Pod) error
+	RemovePod(deletedPod *v1.Pod, node *v1.Node) error
 }
 
 // PredicateMetadataProducer is a function that computes predicate metadata for a given pod.
@@ -67,17 +67,67 @@ type topologyPairsMaps struct {
 	podToTopologyPairs map[string]topologyPairSet
 }
 
-// topologyPairsPodSpreadMap combines topologyKeyToMinPodsMap and topologyPairsMaps
+type criticalPath struct {
+	// topologyValue denotes the topology value mapping to topology key.
+	topologyValue string
+	// matchNum denotes the number of matching pods.
+	matchNum int32
+}
+
+// CAVEAT: the reason that `[2]criticalPath` can work is based on the implementation of current
+// preemption algorithm, in particular the following 2 facts:
+// Fact 1: we only preempt pods on the same node, instead of pods on multiple nodes.
+// Fact 2: each node is evaluated on a separate copy of the metadata during its preemption cycle.
+// If we plan to turn to a more complex algorithm like "arbitrary pods on multiple nodes", this
+// structure needs to be revisited.
+type criticalPaths [2]criticalPath
+
+func newCriticalPaths() *criticalPaths {
+	return &criticalPaths{{matchNum: math.MaxInt32}, {matchNum: math.MaxInt32}}
+}
+
+func (paths *criticalPaths) update(tpVal string, num int32) {
+	// first verify if `tpVal` exists or not
+	i := -1
+	if tpVal == paths[0].topologyValue {
+		i = 0
+	} else if tpVal == paths[1].topologyValue {
+		i = 1
+	}
+
+	if i >= 0 {
+		// `tpVal` exists
+		paths[i].matchNum = num
+		if paths[0].matchNum > paths[1].matchNum {
+			// swap paths[0] and paths[1]
+			paths[0], paths[1] = paths[1], paths[0]
+		}
+	} else {
+		// `tpVal` doesn't exist
+		if num < paths[0].matchNum {
+			// update paths[1] with paths[0]
+			paths[1] = paths[0]
+			// update paths[0]
+			paths[0].topologyValue, paths[0].matchNum = tpVal, num
+		} else if num < paths[1].matchNum {
+			// update paths[1]
+			paths[1].topologyValue, paths[1].matchNum = tpVal, num
+		}
+	}
+}
+
+// podSpreadCache combines tpKeyToCriticalPaths and tpPairToMatchNum
 // to represent:
-// (1) minimum number of pods matched on the spread constraints.
-// (2) how existing pods match incoming pod on its spread constraints.
-type topologyPairsPodSpreadMap struct {
-	// This map is keyed with a topology key, and valued with minimum number
-	// of pods matched on that topology domain.
-	// TODO(Huang-Wei): refactor to {tpKey->tpValSet(or tpValSlice)}
-	topologyKeyToMinPodsMap map[string]int32
-	// TODO(Huang-Wei): refactor to {tpPair->count, podName->tpPairSet(optional)}
-	*topologyPairsMaps
+// (1) critical paths where the least pods are matched on each spread constraint.
+// (2) number of pods matched on each spread constraint.
+type podSpreadCache struct {
+	// We record 2 critical paths instead of all critical paths here.
+	// criticalPaths[0].matchNum always holds the minimum matching number.
+	// criticalPaths[1].matchNum is always greater or equal to criticalPaths[0].matchNum, but
+	// it's not guaranteed to be the 2nd minimum match number.
+	tpKeyToCriticalPaths map[string]*criticalPaths
+	// tpPairToMatchNum is keyed with topologyPair, and valued with the number of matching pods.
+	tpPairToMatchNum map[topologyPair]int32
 }
 
 // NOTE: When new fields are added/removed or logic is changed, please make sure that
@@ -105,9 +155,9 @@ type predicateMetadata struct {
 	// which should be accounted only by the extenders. This set is synthesized
 	// from scheduler extender configuration and does not change per pod.
 	ignoredExtendedResources sets.String
-	// Similar to the map for pod (anti-)affinity, but imposes additional min matches info
-	// to describe minimum match number on each topology spread constraint.
-	topologyPairsPodSpreadMap *topologyPairsPodSpreadMap
+	// podSpreadCache holds info of the minimum match number on each topology spread constraint,
+	// and the match number of all valid topology pairs.
+	podSpreadCache *podSpreadCache
 }
 
 // Ensure that predicateMetadata implements algorithm.PredicateMetadata.
@@ -154,9 +204,9 @@ func (pfactory *PredicateMetadataFactory) GetMetadata(pod *v1.Pod, nodeNameToInf
 	if pod == nil {
 		return nil
 	}
-	// existingPodSpreadConstraintsMap represents how existing pods match "pod"
+	// existingPodSpreadCache represents how existing pods match "pod"
 	// on its spread constraints
-	existingPodSpreadConstraintsMap, err := getTPMapMatchingSpreadConstraints(pod, nodeNameToInfoMap)
+	existingPodSpreadCache, err := getExistingPodSpreadCache(pod, nodeNameToInfoMap)
 	if err != nil {
 		klog.Errorf("Error calculating spreadConstraintsMap: %v", err)
 		return nil
@@ -182,7 +232,7 @@ func (pfactory *PredicateMetadataFactory) GetMetadata(pod *v1.Pod, nodeNameToInf
 		topologyPairsPotentialAffinityPods:     incomingPodAffinityMap,
 		topologyPairsPotentialAntiAffinityPods: incomingPodAntiAffinityMap,
 		topologyPairsAntiAffinityPodsMap:       existingPodAntiAffinityMap,
-		topologyPairsPodSpreadMap:              existingPodSpreadConstraintsMap,
+		podSpreadCache:                         existingPodSpreadCache,
 	}
 	for predicateName, precomputeFunc := range predicateMetadataProducers {
 		klog.V(10).Infof("Precompute: %v", predicateName)
@@ -191,7 +241,7 @@ func (pfactory *PredicateMetadataFactory) GetMetadata(pod *v1.Pod, nodeNameToInf
 	return predicateMetadata
 }
 
-func getTPMapMatchingSpreadConstraints(pod *v1.Pod, nodeInfoMap map[string]*schedulernodeinfo.NodeInfo) (*topologyPairsPodSpreadMap, error) {
+func getExistingPodSpreadCache(pod *v1.Pod, nodeInfoMap map[string]*schedulernodeinfo.NodeInfo) (*podSpreadCache, error) {
 	// We have feature gating in APIServer to strip the spec
 	// so don't need to re-check feature gate, just check length of constraints.
 	constraints := getHardTopologySpreadConstraints(pod)
@@ -207,14 +257,15 @@ func getTPMapMatchingSpreadConstraints(pod *v1.Pod, nodeInfoMap map[string]*sche
 	errCh := schedutil.NewErrorChannel()
 	var lock sync.Mutex
 
-	topologyPairsPodSpreadMap := &topologyPairsPodSpreadMap{
-		// topologyKeyToMinPodsMap will be initialized with proper size later.
-		topologyPairsMaps: newTopologyPairsMaps(),
+	// TODO(Huang-Wei): It might be possible to use "make(map[topologyPair]*int32)".
+	// In that case, need to consider how to init each tpPairToCount[pair] in an atomic fashion.
+	m := podSpreadCache{
+		tpKeyToCriticalPaths: make(map[string]*criticalPaths, len(constraints)),
+		tpPairToMatchNum:     make(map[topologyPair]int32),
 	}
-
-	appendTopologyPairsMaps := func(toAppend *topologyPairsMaps) {
+	addTopologyPairMatchNum := func(pair topologyPair, num int32) {
 		lock.Lock()
-		topologyPairsPodSpreadMap.appendMaps(toAppend)
+		m.tpPairToMatchNum[pair] += num
 		lock.Unlock()
 	}
 
@@ -237,9 +288,8 @@ func getTPMapMatchingSpreadConstraints(pod *v1.Pod, nodeInfoMap map[string]*sche
 		if !NodeLabelsMatchSpreadConstraints(node.Labels, constraints) {
 			return
 		}
-		nodeTopologyMaps := newTopologyPairsMaps()
 		for _, constraint := range constraints {
-			pairAdded := false
+			matchTotal := int32(0)
 			// nodeInfo.Pods() can be empty; or all pods don't fit
 			for _, existingPod := range nodeInfo.Pods() {
 				if existingPod.Namespace != pod.Namespace {
@@ -251,26 +301,12 @@ func getTPMapMatchingSpreadConstraints(pod *v1.Pod, nodeInfoMap map[string]*sche
 					return
 				}
 				if ok {
-					// constraint.TopologyKey is already guaranteed to be present
-					pair := topologyPair{key: constraint.TopologyKey, value: node.Labels[constraint.TopologyKey]}
-					nodeTopologyMaps.addTopologyPair(pair, existingPod)
-					pairAdded = true
+					matchTotal++
 				}
 			}
-			// If needed, append topology pair without entry of pods.
-			// For example, on node-x, there is no pod matching spread constraints,
-			// but node-x should be also considered as a match (with match number 0)
-			// i.e. <node: node-x>: {}
-			if !pairAdded {
-				pair := topologyPair{
-					key:   constraint.TopologyKey,
-					value: node.Labels[constraint.TopologyKey],
-				}
-				nodeTopologyMaps.addTopologyPairWithoutPods(pair)
-			}
+			pair := topologyPair{key: constraint.TopologyKey, value: node.Labels[constraint.TopologyKey]}
+			addTopologyPairMatchNum(pair, matchTotal)
 		}
-
-		appendTopologyPairsMaps(nodeTopologyMaps)
 	}
 	workqueue.ParallelizeUntil(ctx, 16, len(allNodeNames), processNode)
 
@@ -279,18 +315,15 @@ func getTPMapMatchingSpreadConstraints(pod *v1.Pod, nodeInfoMap map[string]*sche
 	}
 
 	// calculate min match for each topology pair
-	topologyPairsPodSpreadMap.topologyKeyToMinPodsMap = make(map[string]int32, len(constraints))
-	for _, constraint := range constraints {
-		topologyPairsPodSpreadMap.topologyKeyToMinPodsMap[constraint.TopologyKey] = math.MaxInt32
+	for i := 0; i < len(constraints); i++ {
+		key := constraints[i].TopologyKey
+		m.tpKeyToCriticalPaths[key] = newCriticalPaths()
 	}
-	for pair, podSet := range topologyPairsPodSpreadMap.topologyPairToPods {
-		// TODO(Huang-Wei): short circuit unvisited portions of <topologyKey: any value>
-		// if we already see 0 as min match of that topologyKey.
-		if l := int32(len(podSet)); l < topologyPairsPodSpreadMap.topologyKeyToMinPodsMap[pair.key] {
-			topologyPairsPodSpreadMap.topologyKeyToMinPodsMap[pair.key] = l
-		}
+	for pair, num := range m.tpPairToMatchNum {
+		m.tpKeyToCriticalPaths[pair.key].update(pair.value, num)
 	}
-	return topologyPairsPodSpreadMap, nil
+
+	return &m, nil
 }
 
 func getHardTopologySpreadConstraints(pod *v1.Pod) (constraints []v1.TopologySpreadConstraint) {
@@ -337,7 +370,9 @@ func newTopologyPairsMaps() *topologyPairsMaps {
 
 func (m *topologyPairsMaps) addTopologyPair(pair topologyPair, pod *v1.Pod) {
 	podFullName := schedutil.GetPodFullName(pod)
-	m.addTopologyPairWithoutPods(pair)
+	if m.topologyPairToPods[pair] == nil {
+		m.topologyPairToPods[pair] = make(map[*v1.Pod]struct{})
+	}
 	m.topologyPairToPods[pair][pod] = struct{}{}
 	if m.podToTopologyPairs[podFullName] == nil {
 		m.podToTopologyPairs[podFullName] = make(map[topologyPair]struct{})
@@ -345,13 +380,6 @@ func (m *topologyPairsMaps) addTopologyPair(pair topologyPair, pod *v1.Pod) {
 	m.podToTopologyPairs[podFullName][pair] = struct{}{}
 }
 
-// add a topology pair holder if needed
-func (m *topologyPairsMaps) addTopologyPairWithoutPods(pair topologyPair) {
-	if m.topologyPairToPods[pair] == nil {
-		m.topologyPairToPods[pair] = make(map[*v1.Pod]struct{})
-	}
-}
-
 func (m *topologyPairsMaps) removePod(deletedPod *v1.Pod) {
 	deletedPodFullName := schedutil.GetPodFullName(deletedPod)
 	for pair := range m.podToTopologyPairs[deletedPodFullName] {
@@ -368,12 +396,8 @@ func (m *topologyPairsMaps) appendMaps(toAppend *topologyPairsMaps) {
 		return
 	}
 	for pair := range toAppend.topologyPairToPods {
-		if podSet := toAppend.topologyPairToPods[pair]; len(podSet) == 0 {
-			m.addTopologyPairWithoutPods(pair)
-		} else {
-			for pod := range podSet {
-				m.addTopologyPair(pair, pod)
-			}
+		for pod := range toAppend.topologyPairToPods[pair] {
+			m.addTopologyPair(pair, pod)
 		}
 	}
 }
@@ -384,8 +408,16 @@ func (m *topologyPairsMaps) clone() *topologyPairsMaps {
 	return copy
 }
 
-func (m *topologyPairsPodSpreadMap) addPod(addedPod, preemptorPod *v1.Pod, node *v1.Node) error {
-	if addedPod.Namespace != preemptorPod.Namespace {
+func (c *podSpreadCache) addPod(addedPod, preemptorPod *v1.Pod, node *v1.Node) error {
+	return c.updatePod(addedPod, preemptorPod, node, 1)
+}
+
+func (c *podSpreadCache) removePod(deletedPod, preemptorPod *v1.Pod, node *v1.Node) {
+	c.updatePod(deletedPod, preemptorPod, node, -1)
+}
+
+func (c *podSpreadCache) updatePod(updatedPod, preemptorPod *v1.Pod, node *v1.Node, delta int32) error {
+	if updatedPod.Namespace != preemptorPod.Namespace || node == nil {
 		return nil
 	}
 	constraints := getHardTopologySpreadConstraints(preemptorPod)
@@ -393,98 +425,45 @@ func (m *topologyPairsPodSpreadMap) addPod(addedPod, preemptorPod *v1.Pod, node
 		return nil
 	}
 
-	// records which topology key(s) needs to be updated
-	minMatchNeedingUpdate := make(map[string]struct{})
-	podLabelSet := labels.Set(addedPod.Labels)
+	podLabelSet := labels.Set(updatedPod.Labels)
 	for _, constraint := range constraints {
 		if match, err := PodMatchesSpreadConstraint(podLabelSet, constraint); err != nil {
 			return err
 		} else if !match {
 			continue
 		}
-		pair := topologyPair{
-			key:   constraint.TopologyKey,
-			value: node.Labels[constraint.TopologyKey],
-		}
-		// it means current node is one of the critical paths of topologyKeyToMinPodsMap[TopologyKey]
-		if int32(len(m.topologyPairToPods[pair])) == m.topologyKeyToMinPodsMap[pair.key] {
-			minMatchNeedingUpdate[pair.key] = struct{}{}
-		}
-		m.addTopologyPair(pair, addedPod)
-	}
-	// no need to addTopologyPairWithoutPods b/c if a pair without pods must be present,
-	// it should have already been created earlier in removePod() phase
 
-	// In most cases, min match map doesn't need to be updated.
-	// But it's required to be updated when current node is the ONLY critical path which impacts
-	// the min match. With that said, in this case min match needs to be updated to min match + 1
-	if len(minMatchNeedingUpdate) != 0 {
-		// TODO(Huang-Wei): performance can be optimized.
-		// A possible solution is to record number of critical paths which co-impact the min match.
-		tempMinMatchMap := make(map[string]int32, len(minMatchNeedingUpdate))
-		for key := range minMatchNeedingUpdate {
-			tempMinMatchMap[key] = math.MaxInt32
-		}
-		for pair, podSet := range m.topologyPairToPods {
-			if _, ok := minMatchNeedingUpdate[pair.key]; !ok {
-				continue
-			}
-			if l := int32(len(podSet)); l < tempMinMatchMap[pair.key] {
-				tempMinMatchMap[pair.key] = l
-			}
-		}
-		for key, tempMin := range tempMinMatchMap {
-			if tempMin == m.topologyKeyToMinPodsMap[key]+1 {
-				m.topologyKeyToMinPodsMap[key] = tempMin
-			}
-		}
+		k, v := constraint.TopologyKey, node.Labels[constraint.TopologyKey]
+		pair := topologyPair{key: k, value: v}
+		c.tpPairToMatchNum[pair] = c.tpPairToMatchNum[pair] + delta
+
+		c.tpKeyToCriticalPaths[k].update(v, c.tpPairToMatchNum[pair])
 	}
 	return nil
 }
 
-func (m *topologyPairsPodSpreadMap) removePod(deletedPod *v1.Pod) {
-	if m == nil || deletedPod == nil {
-		return
-	}
-
-	deletedPodFullName := schedutil.GetPodFullName(deletedPod)
-	pairSet, ok := m.podToTopologyPairs[deletedPodFullName]
-	if !ok {
-		return
-	}
-	topologyPairToPods := m.topologyPairToPods
-	for pair := range pairSet {
-		delete(topologyPairToPods[pair], deletedPod)
-		// if topologyPairToPods[pair] is empty after deletion
-		// don't clean it up as that topology counts as a match now
-
-		// removal of the deletedPod would probably genereate a smaller matching number
-		// so re-calculate minMatch to a smaller value if possible
-		if l := int32(len(topologyPairToPods[pair])); l < m.topologyKeyToMinPodsMap[pair.key] {
-			m.topologyKeyToMinPodsMap[pair.key] = l
-		}
-	}
-	delete(m.podToTopologyPairs, deletedPodFullName)
-}
-
-func (m *topologyPairsPodSpreadMap) clone() *topologyPairsPodSpreadMap {
-	// m could be nil when EvenPodsSpread feature is disabled
-	if m == nil {
+func (c *podSpreadCache) clone() *podSpreadCache {
+	// c could be nil when EvenPodsSpread feature is disabled
+	if c == nil {
 		return nil
 	}
-	copy := &topologyPairsPodSpreadMap{
-		topologyKeyToMinPodsMap: make(map[string]int32),
-		topologyPairsMaps:       m.topologyPairsMaps.clone(),
+	copy := podSpreadCache{
+		tpKeyToCriticalPaths: make(map[string]*criticalPaths),
+		tpPairToMatchNum:     make(map[topologyPair]int32),
 	}
-	for key, minMatched := range m.topologyKeyToMinPodsMap {
-		copy.topologyKeyToMinPodsMap[key] = minMatched
+	for tpKey, paths := range c.tpKeyToCriticalPaths {
+		copy.tpKeyToCriticalPaths[tpKey] = &criticalPaths{paths[0], paths[1]}
 	}
-	return copy
+	for tpPair, matchNum := range c.tpPairToMatchNum {
+		copyPair := topologyPair{key: tpPair.key, value: tpPair.value}
+		copy.tpPairToMatchNum[copyPair] = matchNum
+	}
+	return &copy
 }
 
 // RemovePod changes predicateMetadata assuming that the given `deletedPod` is
 // deleted from the system.
-func (meta *predicateMetadata) RemovePod(deletedPod *v1.Pod) error {
+func (meta *predicateMetadata) RemovePod(deletedPod *v1.Pod, node *v1.Node) error {
 	deletedPodFullName := schedutil.GetPodFullName(deletedPod)
 	if deletedPodFullName == schedutil.GetPodFullName(meta.pod) {
 		return fmt.Errorf("deletedPod and meta.pod must not be the same")
@@ -494,7 +473,7 @@ func (meta *predicateMetadata) RemovePod(deletedPod *v1.Pod) error {
 	meta.topologyPairsPotentialAffinityPods.removePod(deletedPod)
 	meta.topologyPairsPotentialAntiAffinityPods.removePod(deletedPod)
 	// Delete pod from the pod spread topology maps.
-	meta.topologyPairsPodSpreadMap.removePod(deletedPod)
+	meta.podSpreadCache.removePod(deletedPod, meta.pod, node)
 	// All pods in the serviceAffinityMatchingPodList are in the same namespace.
 	// So, if the namespace of the first one is not the same as the namespace of the
 	// deletedPod, we don't need to check the list, as deletedPod isn't in the list.
@@ -556,9 +535,9 @@ func (meta *predicateMetadata) AddPod(addedPod *v1.Pod, nodeInfo *schedulernodei
 			}
 		}
 	}
-	// Update meta.topologyPairsPodSpreadMap if meta.pod has hard spread constraints
+	// Update meta.podSpreadCache if meta.pod has hard spread constraints
 	// and addedPod matches that
-	if err := meta.topologyPairsPodSpreadMap.addPod(addedPod, meta.pod, nodeInfo.Node()); err != nil {
+	if err := meta.podSpreadCache.addPod(addedPod, meta.pod, nodeInfo.Node()); err != nil {
 		return err
 	}
 
@@ -588,7 +567,7 @@ func (meta *predicateMetadata) ShallowCopy() PredicateMetadata {
 	newPredMeta.topologyPairsPotentialAffinityPods = meta.topologyPairsPotentialAffinityPods.clone()
 	newPredMeta.topologyPairsPotentialAntiAffinityPods = meta.topologyPairsPotentialAntiAffinityPods.clone()
 	newPredMeta.topologyPairsAntiAffinityPodsMap = meta.topologyPairsAntiAffinityPodsMap.clone()
-	newPredMeta.topologyPairsPodSpreadMap = meta.topologyPairsPodSpreadMap.clone()
+	newPredMeta.podSpreadCache = meta.podSpreadCache.clone()
 	newPredMeta.serviceAffinityMatchingPodServices = append([]*v1.Service(nil),
 		meta.serviceAffinityMatchingPodServices...)
 	newPredMeta.serviceAffinityMatchingPodList = append([]*v1.Pod(nil),

pkg/scheduler/algorithm/predicates/predicates.go

@@ -1796,15 +1796,15 @@ func EvenPodsSpreadPredicate(pod *v1.Pod, meta PredicateMetadata, nodeInfo *sche
 		return true, nil, nil
 	}
 
-	var topologyPairsPodSpreadMap *topologyPairsPodSpreadMap
+	var podSpreadCache *podSpreadCache
 	if predicateMeta, ok := meta.(*predicateMetadata); ok {
-		topologyPairsPodSpreadMap = predicateMeta.topologyPairsPodSpreadMap
+		podSpreadCache = predicateMeta.podSpreadCache
 	} else { // We don't have precomputed metadata. We have to follow a slow path to check spread constraints.
-		// TODO(Huang-Wei): get it implemented
+		// TODO(autoscaler): get it implemented
 		return false, nil, errors.New("metadata not pre-computed for EvenPodsSpreadPredicate")
 	}
 
-	if topologyPairsPodSpreadMap == nil || len(topologyPairsPodSpreadMap.topologyKeyToMinPodsMap) == 0 {
+	if podSpreadCache == nil || len(podSpreadCache.tpPairToMatchNum) == 0 {
 		return true, nil, nil
 	}
 
@@ -1821,25 +1821,24 @@ func EvenPodsSpreadPredicate(pod *v1.Pod, meta PredicateMetadata, nodeInfo *sche
 		if err != nil {
 			return false, nil, err
 		}
-		selfMatchNum := 0
+		selfMatchNum := int32(0)
 		if selfMatch {
 			selfMatchNum = 1
 		}
 
 		pair := topologyPair{key: tpKey, value: tpVal}
-		minMatchNum, ok := topologyPairsPodSpreadMap.topologyKeyToMinPodsMap[tpKey]
+		paths, ok := podSpreadCache.tpKeyToCriticalPaths[tpKey]
 		if !ok {
 			// error which should not happen
-			klog.Errorf("internal error: get minMatchNum from key %q of %#v", tpKey, topologyPairsPodSpreadMap.topologyKeyToMinPodsMap)
+			klog.Errorf("internal error: get paths from key %q of %#v", tpKey, podSpreadCache.tpKeyToCriticalPaths)
 			continue
 		}
 		// judging criteria:
 		// 'existing matching num' + 'if self-match (1 or 0)' - 'global min matching num' <= 'maxSkew'
-		matchNum := len(topologyPairsPodSpreadMap.topologyPairToPods[pair])
-
-		// cast to int to avoid potential overflow.
-		skew := matchNum + selfMatchNum - int(minMatchNum)
-		if skew > int(constraint.MaxSkew) {
+		minMatchNum := paths[0].matchNum
+		matchNum := podSpreadCache.tpPairToMatchNum[pair]
+		skew := matchNum + selfMatchNum - minMatchNum
+		if skew > constraint.MaxSkew {
 			klog.V(5).Infof("node '%s' failed spreadConstraint[%s]: matchNum(%d) + selfMatchNum(%d) - minMatchNum(%d) > maxSkew(%d)", node.Name, tpKey, matchNum, selfMatchNum, minMatchNum, constraint.MaxSkew)
 			return false, []PredicateFailureReason{ErrTopologySpreadConstraintsNotMatch}, nil
 		}

pkg/scheduler/algorithm/priorities/even_pods_spread.go

@@ -38,15 +38,15 @@ type topologyPair struct {
 
 type topologySpreadConstraintsMap struct {
 	// nodeNameToPodCounts is keyed with node name, and valued with the number of matching pods.
-	nodeNameToPodCounts map[string]int64
+	nodeNameToPodCounts map[string]int32
 	// topologyPairToPodCounts is keyed with topologyPair, and valued with the number of matching pods.
-	topologyPairToPodCounts map[topologyPair]*int64
+	topologyPairToPodCounts map[topologyPair]*int32
 }
 
 func newTopologySpreadConstraintsMap() *topologySpreadConstraintsMap {
 	return &topologySpreadConstraintsMap{
-		nodeNameToPodCounts:     make(map[string]int64),
-		topologyPairToPodCounts: make(map[topologyPair]*int64),
+		nodeNameToPodCounts:     make(map[string]int32),
+		topologyPairToPodCounts: make(map[topologyPair]*int32),
 	}
 }
 
@@ -54,7 +54,7 @@ func newTopologySpreadConstraintsMap() *topologySpreadConstraintsMap {
 // This function iterates <nodes> to filter out the nodes which don't have required topologyKey(s),
 // and initialize two maps:
 // 1) t.topologyPairToPodCounts: keyed with both eligible topology pair and node names.
-// 2) t.nodeNameToPodCounts: keyed with node name, and valued with a *int64 pointer for eligible node only.
+// 2) t.nodeNameToPodCounts: keyed with node name, and valued with a *int32 pointer for eligible node only.
 func (t *topologySpreadConstraintsMap) initialize(pod *v1.Pod, nodes []*v1.Node) {
 	constraints := getSoftTopologySpreadConstraints(pod)
 	for _, node := range nodes {
@@ -64,7 +64,7 @@ func (t *topologySpreadConstraintsMap) initialize(pod *v1.Pod, nodes []*v1.Node)
 		for _, constraint := range constraints {
 			pair := topologyPair{key: constraint.TopologyKey, value: node.Labels[constraint.TopologyKey]}
 			if t.topologyPairToPodCounts[pair] == nil {
-				t.topologyPairToPodCounts[pair] = new(int64)
+				t.topologyPairToPodCounts[pair] = new(int32)
 			}
 		}
 		t.nodeNameToPodCounts[node.Name] = 0
@@ -122,7 +122,7 @@ func CalculateEvenPodsSpreadPriority(pod *v1.Pod, nodeNameToInfo map[string]*sch
 			}
 
 			// <matchSum> indicates how many pods (on current node) match the <constraint>.
-			matchSum := int64(0)
+			matchSum := int32(0)
 			for _, existingPod := range nodeInfo.Pods() {
 				match, err := predicates.PodMatchesSpreadConstraint(existingPod.Labels, constraint)
 				if err != nil {
@@ -133,7 +133,7 @@ func CalculateEvenPodsSpreadPriority(pod *v1.Pod, nodeNameToInfo map[string]*sch
 					matchSum++
 				}
 			}
-			atomic.AddInt64(t.topologyPairToPodCounts[pair], matchSum)
+			atomic.AddInt32(t.topologyPairToPodCounts[pair], matchSum)
 		}
 	}
 	workqueue.ParallelizeUntil(ctx, 16, len(allNodeNames), processAllNode)
@@ -141,9 +141,9 @@ func CalculateEvenPodsSpreadPriority(pod *v1.Pod, nodeNameToInfo map[string]*sch
 		return nil, err
 	}
 
-	var minCount int64 = math.MaxInt64
+	var minCount int32 = math.MaxInt32
 	// <total> sums up the number of matching pods on each qualified topology pair
-	var total int64
+	var total int32
 	for _, node := range nodes {
 		if _, ok := t.nodeNameToPodCounts[node.Name]; !ok {
 			continue

pkg/scheduler/algorithm/priorities/even_pods_spread_test.go

@@ -17,7 +17,6 @@ limitations under the License.
 package priorities
 
 import (
-	"fmt"
 	"reflect"
 	"testing"
 
@@ -32,8 +31,8 @@ func Test_topologySpreadConstraintsMap_initialize(t *testing.T) {
 		name                string
 		pod                 *v1.Pod
 		nodes               []*v1.Node
-		wantNodeNameMap     map[string]int64
-		wantTopologyPairMap map[topologyPair]*int64
+		wantNodeNameMap     map[string]int32
+		wantTopologyPairMap map[topologyPair]*int32
 	}{
 		{
 			name: "normal case",
@@ -46,17 +45,17 @@ func Test_topologySpreadConstraintsMap_initialize(t *testing.T) {
 				st.MakeNode().Name("node-b").Label("zone", "zone1").Label("node", "node-b").Obj(),
 				st.MakeNode().Name("node-x").Label("zone", "zone2").Label("node", "node-x").Obj(),
 			},
-			wantNodeNameMap: map[string]int64{
+			wantNodeNameMap: map[string]int32{
 				"node-a": 0,
 				"node-b": 0,
 				"node-x": 0,
 			},
-			wantTopologyPairMap: map[topologyPair]*int64{
-				{key: "zone", value: "zone1"}:  new(int64),
-				{key: "zone", value: "zone2"}:  new(int64),
-				{key: "node", value: "node-a"}: new(int64),
-				{key: "node", value: "node-b"}: new(int64),
-				{key: "node", value: "node-x"}: new(int64),
+			wantTopologyPairMap: map[topologyPair]*int32{
+				{key: "zone", value: "zone1"}:  new(int32),
+				{key: "zone", value: "zone2"}:  new(int32),
+				{key: "node", value: "node-a"}: new(int32),
+				{key: "node", value: "node-b"}: new(int32),
+				{key: "node", value: "node-x"}: new(int32),
 			},
 		},
 		{
@@ -70,14 +69,14 @@ func Test_topologySpreadConstraintsMap_initialize(t *testing.T) {
 				st.MakeNode().Name("node-b").Label("zone", "zone1").Label("node", "node-b").Obj(),
 				st.MakeNode().Name("node-x").Label("node", "node-x").Obj(),
 			},
-			wantNodeNameMap: map[string]int64{
+			wantNodeNameMap: map[string]int32{
 				"node-a": 0,
 				"node-b": 0,
 			},
-			wantTopologyPairMap: map[topologyPair]*int64{
-				{key: "zone", value: "zone1"}:  new(int64),
-				{key: "node", value: "node-a"}: new(int64),
-				{key: "node", value: "node-b"}: new(int64),
+			wantTopologyPairMap: map[topologyPair]*int32{
+				{key: "zone", value: "zone1"}:  new(int32),
+				{key: "node", value: "node-a"}: new(int32),
+				{key: "node", value: "node-b"}: new(int32),
 			},
 		},
 	}
@@ -445,43 +444,6 @@ func TestCalculateEvenPodsSpreadPriority(t *testing.T) {
 	}
 }
 
-func makeNodesAndPods(pod *v1.Pod, existingPodsNum, allNodesNum, filteredNodesNum int) (existingPods []*v1.Pod, allNodes []*v1.Node, filteredNodes []*v1.Node) {
-	var topologyKeys []string
-	var labels []string
-	// regions := 3
-	zones := 10
-	for _, c := range pod.Spec.TopologySpreadConstraints {
-		topologyKeys = append(topologyKeys, c.TopologyKey)
-		labels = append(labels, c.LabelSelector.MatchExpressions[0].Key)
-	}
-	// build nodes
-	for i := 0; i < allNodesNum; i++ {
-		nodeWrapper := st.MakeNode().Name(fmt.Sprintf("node%d", i))
-		for _, tpKey := range topologyKeys {
-			if tpKey == "zone" {
-				nodeWrapper = nodeWrapper.Label("zone", fmt.Sprintf("zone%d", i%zones))
-			} else if tpKey == "node" {
-				nodeWrapper = nodeWrapper.Label("node", fmt.Sprintf("node%d", i))
-			}
-		}
-		node := nodeWrapper.Obj()
-		allNodes = append(allNodes, node)
-		if len(filteredNodes) < filteredNodesNum {
-			filteredNodes = append(filteredNodes, node)
-		}
-	}
-	// build pods
-	for i := 0; i < existingPodsNum; i++ {
-		podWrapper := st.MakePod().Name(fmt.Sprintf("pod%d", i)).Node(fmt.Sprintf("node%d", i%allNodesNum))
-		// apply labels[0], labels[0,1], ..., labels[all] to each pod in turn
-		for _, label := range labels[:i%len(labels)+1] {
-			podWrapper = podWrapper.Label(label, "")
-		}
-		existingPods = append(existingPods, podWrapper.Obj())
-	}
-	return
-}
-
 func BenchmarkTestCalculateEvenPodsSpreadPriority(b *testing.B) {
 	tests := []struct {
 		name             string
@@ -521,7 +483,7 @@ func BenchmarkTestCalculateEvenPodsSpreadPriority(b *testing.B) {
 	}
 	for _, tt := range tests {
 		b.Run(tt.name, func(b *testing.B) {
-			existingPods, allNodes, filteredNodes := makeNodesAndPods(tt.pod, tt.existingPodsNum, tt.allNodesNum, tt.filteredNodesNum)
+			existingPods, allNodes, filteredNodes := st.MakeNodesAndPods(tt.pod, tt.existingPodsNum, tt.allNodesNum, tt.filteredNodesNum)
 			nodeNameToInfo := schedulernodeinfo.CreateNodeNameToInfoMap(existingPods, allNodes)
 			b.ResetTimer()
 			for i := 0; i < b.N; i++ {

pkg/scheduler/core/generic_scheduler.go

@@ -1107,7 +1107,7 @@ func selectVictimsOnNode(
 	removePod := func(rp *v1.Pod) {
 		nodeInfoCopy.RemovePod(rp)
 		if meta != nil {
-			meta.RemovePod(rp)
+			meta.RemovePod(rp, nodeInfoCopy.Node())
 		}
 	}
 	addPod := func(ap *v1.Pod) {

pkg/scheduler/testing/BUILD

@@ -6,6 +6,7 @@ go_library(
     name = "go_default_library",
     srcs = [
         "fake_lister.go",
+        "workload_prep.go",
         "wrappers.go",
     ],
     importpath = "k8s.io/kubernetes/pkg/scheduler/testing",

pkg/scheduler/testing/workload_prep.go

@@ -0,0 +1,67 @@
+/*
+Copyright 2019 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 testing
+
+import (
+	"fmt"
+
+	"k8s.io/api/core/v1"
+)
+
+// MakeNodesAndPods serves as a testing helper for EvenPodsSpread feature.
+// It builds a fake cluster containing running Pods and Nodes.
+// The size of Pods and Nodes are determined by input arguments.
+// The specs of Pods and Nodes are generated with the following rules:
+// - If `pod` has "node" as a topologyKey, each generated node is applied with a unique label: "node: node<i>".
+// - If `pod` has "zone" as a topologyKey, each generated node is applied with a rotating label: "zone: zone[0-9]".
+// - Depending on "lableSelector.MatchExpressions[0].Key" the `pod` has in each topologySpreadConstraint,
+//   each generated pod will be applied with label "key1", "key1,key2", ..., "key1,key2,...,keyN" in a rotating manner.
+func MakeNodesAndPods(pod *v1.Pod, existingPodsNum, allNodesNum, filteredNodesNum int) (existingPods []*v1.Pod, allNodes []*v1.Node, filteredNodes []*v1.Node) {
+	var topologyKeys []string
+	var labels []string
+	zones := 10
+	for _, c := range pod.Spec.TopologySpreadConstraints {
+		topologyKeys = append(topologyKeys, c.TopologyKey)
+		labels = append(labels, c.LabelSelector.MatchExpressions[0].Key)
+	}
+	// build nodes
+	for i := 0; i < allNodesNum; i++ {
+		nodeWrapper := MakeNode().Name(fmt.Sprintf("node%d", i))
+		for _, tpKey := range topologyKeys {
+			if tpKey == "zone" {
+				nodeWrapper = nodeWrapper.Label("zone", fmt.Sprintf("zone%d", i%zones))
+			} else if tpKey == "node" {
+				nodeWrapper = nodeWrapper.Label("node", fmt.Sprintf("node%d", i))
+			}
+		}
+		node := nodeWrapper.Obj()
+		allNodes = append(allNodes, node)
+		if len(filteredNodes) < filteredNodesNum {
+			filteredNodes = append(filteredNodes, node)
+		}
+	}
+	// build pods
+	for i := 0; i < existingPodsNum; i++ {
+		podWrapper := MakePod().Name(fmt.Sprintf("pod%d", i)).Node(fmt.Sprintf("node%d", i%allNodesNum))
+		// apply labels[0], labels[0,1], ..., labels[all] to each pod in turn
+		for _, label := range labels[:i%len(labels)+1] {
+			podWrapper = podWrapper.Label(label, "")
+		}
+		existingPods = append(existingPods, podWrapper.Obj())
+	}
+	return
+}