diff --git a/test/e2e/topology_updater.go b/test/e2e/topology_updater.go
index 043983a99..05e2b65f2 100644
--- a/test/e2e/topology_updater.go
+++ b/test/e2e/topology_updater.go
@@ -19,6 +19,7 @@ package e2e
 import (
 	"context"
 	"fmt"
+	"k8s.io/apimachinery/pkg/api/resource"
 	"time"
 
 	. "github.com/onsi/ginkgo/v2"
@@ -175,7 +176,12 @@ var _ = SIGDescribe("Node Feature Discovery topology updater", func() {
 			By("getting the initial topology information")
 			initialNodeTopo := testutils.GetNodeTopology(topologyClient, topologyUpdaterNode.Name)
 			By("creating a pod consuming resources from the shared, non-exclusive CPU pool (guaranteed QoS, nonintegral request)")
-			sleeperPod := testpod.GuaranteedSleeper("500m")
+			sleeperPod := testpod.GuaranteedSleeper(testpod.WithLimits(
+				corev1.ResourceList{
+					corev1.ResourceCPU: resource.MustParse("500m"),
+					// any random reasonable amount is fine
+					corev1.ResourceMemory: resource.MustParse("100Mi"),
+				}))
 
 			podMap := make(map[string]*corev1.Pod)
 			pod := f.PodClient().CreateSync(sleeperPod)
@@ -221,7 +227,12 @@ var _ = SIGDescribe("Node Feature Discovery topology updater", func() {
 			By("getting the initial topology information")
 			initialNodeTopo := testutils.GetNodeTopology(topologyClient, topologyUpdaterNode.Name)
 			By("creating a pod consuming exclusive CPUs")
-			sleeperPod := testpod.GuaranteedSleeper("1000m")
+			sleeperPod := testpod.GuaranteedSleeper(testpod.WithLimits(
+				corev1.ResourceList{
+					corev1.ResourceCPU: resource.MustParse("1000m"),
+					// any random reasonable amount is fine
+					corev1.ResourceMemory: resource.MustParse("100Mi"),
+				}))
 			// in case there is more than a single node in the cluster
 			// we need to set the node name, so we'll have certainty about
 			// which node we need to examine
diff --git a/test/e2e/utils/pod/pod.go b/test/e2e/utils/pod/pod.go
index 75d39aeaa..698d94b4d 100644
--- a/test/e2e/utils/pod/pod.go
+++ b/test/e2e/utils/pod/pod.go
@@ -25,7 +25,6 @@ import (
 	"github.com/onsi/ginkgo/v2"
 
 	corev1 "k8s.io/api/core/v1"
-	"k8s.io/apimachinery/pkg/api/resource"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/apimachinery/pkg/labels"
 	"k8s.io/apimachinery/pkg/util/uuid"
@@ -45,8 +44,8 @@ const (
 )
 
 // GuaranteedSleeper  makes a Guaranteed QoS class Pod object which long enough forever but requires `cpuLimit` exclusive CPUs.
-func GuaranteedSleeper(cpuLimit string) *corev1.Pod {
-	return &corev1.Pod{
+func GuaranteedSleeper(opts ...func(pod *corev1.Pod)) *corev1.Pod {
+	p := &corev1.Pod{
 		ObjectMeta: metav1.ObjectMeta{
 			Name: "sleeper-gu-pod",
 		},
@@ -54,20 +53,23 @@ func GuaranteedSleeper(cpuLimit string) *corev1.Pod {
 			RestartPolicy: corev1.RestartPolicyNever,
 			Containers: []corev1.Container{
 				{
-					Name:  "sleeper-gu-cnt",
-					Image: PauseImage,
-					Resources: corev1.ResourceRequirements{
-						Limits: corev1.ResourceList{
-							// we use 1 core because that's the minimal meaningful quantity
-							corev1.ResourceName(corev1.ResourceCPU): resource.MustParse(cpuLimit),
-							// any random reasonable amount is fine
-							corev1.ResourceName(corev1.ResourceMemory): resource.MustParse("100Mi"),
-						},
-					},
+					Name:      "sleeper-gu-cnt",
+					Image:     PauseImage,
+					Resources: corev1.ResourceRequirements{},
 				},
 			},
 		},
 	}
+	for _, o := range opts {
+		o(p)
+	}
+	return p
+}
+
+func WithLimits(list corev1.ResourceList) func(p *corev1.Pod) {
+	return func(p *corev1.Pod) {
+		p.Spec.Containers[0].Resources.Limits = list
+	}
 }
 
 // BestEffortSleeper makes a Best Effort QoS class Pod object which sleeps long enough