diff --git a/test/e2e/autoscaling/autoscaling_timer.go b/test/e2e/autoscaling/autoscaling_timer.go
index 2bb9a879823..cad8aa1a021 100644
--- a/test/e2e/autoscaling/autoscaling_timer.go
+++ b/test/e2e/autoscaling/autoscaling_timer.go
@@ -34,7 +34,7 @@ import (
 var _ = SIGDescribe("[Feature:ClusterSizeAutoscalingScaleUp] [Slow] Autoscaling", func() {
 	f := framework.NewDefaultFramework("autoscaling")
 
-	SIGDescribe("Autoscaling a service", func() {
+	ginkgo.Describe("Autoscaling a service", func() {
 		ginkgo.BeforeEach(func() {
 			// Check if Cloud Autoscaler is enabled by trying to get its ConfigMap.
 			_, err := f.ClientSet.CoreV1().ConfigMaps("kube-system").Get(context.TODO(), "cluster-autoscaler-status", metav1.GetOptions{})
diff --git a/test/e2e/autoscaling/horizontal_pod_autoscaling.go b/test/e2e/autoscaling/horizontal_pod_autoscaling.go
index bc9d51f790b..9e277b7e6cc 100644
--- a/test/e2e/autoscaling/horizontal_pod_autoscaling.go
+++ b/test/e2e/autoscaling/horizontal_pod_autoscaling.go
@@ -34,7 +34,7 @@ var _ = SIGDescribe("[Feature:HPA] Horizontal pod autoscaling (scale resource: C
 	titleUp := "Should scale from 1 pod to 3 pods and from 3 to 5"
 	titleDown := "Should scale from 5 pods to 3 pods and from 3 to 1"
 
-	SIGDescribe("[Serial] [Slow] Deployment", func() {
+	ginkgo.Describe("[Serial] [Slow] Deployment", func() {
 		// CPU tests via deployments
 		ginkgo.It(titleUp, func() {
 			scaleUp("test-deployment", e2eautoscaling.KindDeployment, false, f)
@@ -44,7 +44,7 @@ var _ = SIGDescribe("[Feature:HPA] Horizontal pod autoscaling (scale resource: C
 		})
 	})
 
-	SIGDescribe("[Serial] [Slow] ReplicaSet", func() {
+	ginkgo.Describe("[Serial] [Slow] ReplicaSet", func() {
 		// CPU tests via ReplicaSets
 		ginkgo.It(titleUp, func() {
 			scaleUp("rs", e2eautoscaling.KindReplicaSet, false, f)
@@ -55,7 +55,7 @@ var _ = SIGDescribe("[Feature:HPA] Horizontal pod autoscaling (scale resource: C
 	})
 
 	// These tests take ~20 minutes each.
-	SIGDescribe("[Serial] [Slow] ReplicationController", func() {
+	ginkgo.Describe("[Serial] [Slow] ReplicationController", func() {
 		// CPU tests via replication controllers
 		ginkgo.It(titleUp+" and verify decision stability", func() {
 			scaleUp("rc", e2eautoscaling.KindRC, true, f)
@@ -65,7 +65,7 @@ var _ = SIGDescribe("[Feature:HPA] Horizontal pod autoscaling (scale resource: C
 		})
 	})
 
-	SIGDescribe("ReplicationController light", func() {
+	ginkgo.Describe("ReplicationController light", func() {
 		ginkgo.It("Should scale from 1 pod to 2 pods", func() {
 			scaleTest := &HPAScaleTest{
 				initPods:                    1,
diff --git a/test/e2e/node/kubelet.go b/test/e2e/node/kubelet.go
index 35064865453..ed71ac50c0c 100644
--- a/test/e2e/node/kubelet.go
+++ b/test/e2e/node/kubelet.go
@@ -276,7 +276,7 @@ var _ = SIGDescribe("kubelet", func() {
 		ns = f.Namespace.Name
 	})
 
-	SIGDescribe("Clean up pods on node", func() {
+	ginkgo.Describe("Clean up pods on node", func() {
 		var (
 			numNodes        int
 			nodeNames       sets.String
@@ -384,7 +384,7 @@ var _ = SIGDescribe("kubelet", func() {
 	})
 
 	// Test host cleanup when disrupting the volume environment.
-	SIGDescribe("host cleanup with volume mounts [sig-storage][HostCleanup][Flaky]", func() {
+	ginkgo.Describe("host cleanup with volume mounts [sig-storage][HostCleanup][Flaky]", func() {
 
 		type hostCleanupTest struct {
 			itDescr string
diff --git a/test/e2e/node/kubelet_perf.go b/test/e2e/node/kubelet_perf.go
index a2668264c81..5d8406bb0f5 100644
--- a/test/e2e/node/kubelet_perf.go
+++ b/test/e2e/node/kubelet_perf.go
@@ -216,7 +216,7 @@ var _ = SIGDescribe("Kubelet [Serial] [Slow]", func() {
 		result := om.GetLatestRuntimeOperationErrorRate()
 		framework.Logf("runtime operation error metrics:\n%s", e2ekubelet.FormatRuntimeOperationErrorRate(result))
 	})
-	SIGDescribe("regular resource usage tracking [Feature:RegularResourceUsageTracking]", func() {
+	ginkgo.Describe("regular resource usage tracking [Feature:RegularResourceUsageTracking]", func() {
 		// We assume that the scheduler will make reasonable scheduling choices
 		// and assign ~N pods on the node.
 		// Although we want to track N pods per node, there are N + add-on pods
@@ -268,7 +268,7 @@ var _ = SIGDescribe("Kubelet [Serial] [Slow]", func() {
 			})
 		}
 	})
-	SIGDescribe("experimental resource usage tracking [Feature:ExperimentalResourceUsageTracking]", func() {
+	ginkgo.Describe("experimental resource usage tracking [Feature:ExperimentalResourceUsageTracking]", func() {
 		density := []int{100}
 		for i := range density {
 			podsPerNode := density[i]
diff --git a/test/e2e/storage/vsphere/pv_reclaimpolicy.go b/test/e2e/storage/vsphere/pv_reclaimpolicy.go
index fce1e8355b7..b8f1d0f4b93 100644
--- a/test/e2e/storage/vsphere/pv_reclaimpolicy.go
+++ b/test/e2e/storage/vsphere/pv_reclaimpolicy.go
@@ -50,7 +50,7 @@ var _ = utils.SIGDescribe("PersistentVolumes [Feature:vsphere][Feature:ReclaimPo
 		framework.ExpectNoError(framework.WaitForAllNodesSchedulable(c, framework.TestContext.NodeSchedulableTimeout))
 	})
 
-	utils.SIGDescribe("persistentvolumereclaim:vsphere [Feature:vsphere]", func() {
+	ginkgo.Describe("persistentvolumereclaim:vsphere [Feature:vsphere]", func() {
 		ginkgo.BeforeEach(func() {
 			e2eskipper.SkipUnlessProviderIs("vsphere")
 			Bootstrap(f)
diff --git a/test/e2e/storage/vsphere/pvc_label_selector.go b/test/e2e/storage/vsphere/pvc_label_selector.go
index 906e94787b1..3245f18569b 100644
--- a/test/e2e/storage/vsphere/pvc_label_selector.go
+++ b/test/e2e/storage/vsphere/pvc_label_selector.go
@@ -75,7 +75,7 @@ var _ = utils.SIGDescribe("PersistentVolumes [Feature:vsphere][Feature:LabelSele
 
 	})
 
-	utils.SIGDescribe("Selector-Label Volume Binding:vsphere [Feature:vsphere]", func() {
+	ginkgo.Describe("Selector-Label Volume Binding:vsphere [Feature:vsphere]", func() {
 		ginkgo.AfterEach(func() {
 			ginkgo.By("Running clean up actions")
 			if framework.ProviderIs("vsphere") {