diff --git a/src/runtime/virtcontainers/clh.go b/src/runtime/virtcontainers/clh.go
index 6984b73be5..801d53a9bd 100644
--- a/src/runtime/virtcontainers/clh.go
+++ b/src/runtime/virtcontainers/clh.go
@@ -480,12 +480,9 @@ func (clh *cloudHypervisor) CreateVM(ctx context.Context, id string, network Net
 	// Enable hugepages if needed
 	clh.vmconfig.Memory.Hugepages = func(b bool) *bool { return &b }(clh.config.HugePages)
 	if !clh.config.ConfidentialGuest {
-		hostMemKb, err := GetHostMemorySizeKb(procMemInfo)
-		if err != nil {
-			return nil
-		}
+		hotplugSize := clh.config.DefaultMaxMemorySize
 		// OpenAPI only supports int64 values
-		clh.vmconfig.Memory.HotplugSize = func(i int64) *int64 { return &i }(int64((utils.MemUnit(hostMemKb) * utils.KiB).ToBytes()))
+		clh.vmconfig.Memory.HotplugSize = func(i int64) *int64 { return &i }(int64((utils.MemUnit(hotplugSize) * utils.MiB).ToBytes()))
 	}
 	// Set initial amount of cpu's for the virtual machine
 	clh.vmconfig.Cpus = chclient.NewCpusConfig(int32(clh.config.NumVCPUs), int32(clh.config.DefaultMaxVCPUs))
@@ -882,6 +879,11 @@ func (clh *cloudHypervisor) ResizeMemory(ctx context.Context, reqMemMB uint32, m
 		return 0, MemoryDevice{}, err
 	}
 
+	maxHotplugSize := utils.MemUnit(*info.Config.Memory.HotplugSize) * utils.Byte
+	if reqMemMB > uint32(maxHotplugSize.ToMiB()) {
+		reqMemMB = uint32(maxHotplugSize.ToMiB())
+	}
+
 	currentMem := utils.MemUnit(info.Config.Memory.Size) * utils.Byte
 	newMem := utils.MemUnit(reqMemMB) * utils.MiB
 
diff --git a/src/runtime/virtcontainers/qemu.go b/src/runtime/virtcontainers/qemu.go
index c8e48953f6..f2c501ac3a 100644
--- a/src/runtime/virtcontainers/qemu.go
+++ b/src/runtime/virtcontainers/qemu.go
@@ -315,11 +315,7 @@ func (q *qemu) hostMemMB() (uint64, error) {
 }
 
 func (q *qemu) memoryTopology() (govmmQemu.Memory, error) {
-	hostMemMb, err := q.hostMemMB()
-	if err != nil {
-		return govmmQemu.Memory{}, err
-	}
-
+	hostMemMb := q.config.DefaultMaxMemorySize
 	memMb := uint64(q.config.MemorySize)
 
 	return q.arch.memoryTopology(memMb, hostMemMb, uint8(q.config.MemSlots)), nil
@@ -779,12 +775,8 @@ func (q *qemu) getMemArgs() (bool, string, string, error) {
 }
 
 func (q *qemu) setupVirtioMem(ctx context.Context) error {
-	maxMem, err := q.hostMemMB()
-	if err != nil {
-		return err
-	}
 	// backend memory size must be multiple of 4Mib
-	sizeMB := (int(maxMem) - int(q.config.MemorySize)) >> 2 << 2
+	sizeMB := (int(q.config.DefaultMaxMemorySize) - int(q.config.MemorySize)) >> 2 << 2
 
 	share, target, memoryBack, err := q.getMemArgs()
 	if err != nil {
@@ -1970,8 +1962,6 @@ func (q *qemu) hotplugMemory(memDev *MemoryDevice, op Operation) (int, error) {
 		return 0, err
 	}
 
-	currentMemory := int(q.config.MemorySize) + q.state.HotpluggedMemory
-
 	if memDev.SizeMB == 0 {
 		memLog.Debug("hotplug is not required")
 		return 0, nil
@@ -1985,17 +1975,7 @@ func (q *qemu) hotplugMemory(memDev *MemoryDevice, op Operation) (int, error) {
 		return 0, nil
 	case AddDevice:
 		memLog.WithField("operation", "add").Debugf("Requested to add memory: %d MB", memDev.SizeMB)
-		maxMem, err := q.hostMemMB()
-		if err != nil {
-			return 0, err
-		}
 
-		// Don't exceed the maximum amount of memory
-		if currentMemory+memDev.SizeMB > int(maxMem) {
-			// Fixme: return a typed error
-			return 0, fmt.Errorf("Unable to hotplug %d MiB memory, the SB has %d MiB and the maximum amount is %d MiB",
-				memDev.SizeMB, currentMemory, maxMem)
-		}
 		memoryAdded, err := q.hotplugAddMemory(memDev)
 		if err != nil {
 			return memoryAdded, err
@@ -2231,6 +2211,11 @@ func (q *qemu) ResizeMemory(ctx context.Context, reqMemMB uint32, memoryBlockSiz
 	case currentMemory < reqMemMB:
 		//hotplug
 		addMemMB := reqMemMB - currentMemory
+
+		if currentMemory+addMemMB > uint32(q.config.DefaultMaxMemorySize) {
+			addMemMB = uint32(q.config.DefaultMaxMemorySize) - currentMemory
+		}
+
 		memHotplugMB, err := calcHotplugMemMiBSize(addMemMB, memoryBlockSizeMB)
 		if err != nil {
 			return currentMemory, MemoryDevice{}, err
diff --git a/src/runtime/virtcontainers/qemu_test.go b/src/runtime/virtcontainers/qemu_test.go
index b50d73a917..b932189e8d 100644
--- a/src/runtime/virtcontainers/qemu_test.go
+++ b/src/runtime/virtcontainers/qemu_test.go
@@ -21,6 +21,7 @@ import (
 	"github.com/kata-containers/kata-containers/src/runtime/virtcontainers/persist"
 	"github.com/kata-containers/kata-containers/src/runtime/virtcontainers/types"
 	"github.com/kata-containers/kata-containers/src/runtime/virtcontainers/utils"
+	"github.com/pbnjay/memory"
 	"github.com/pkg/errors"
 	"github.com/stretchr/testify/assert"
 )
@@ -172,20 +173,20 @@ func TestQemuCPUTopology(t *testing.T) {
 
 func TestQemuMemoryTopology(t *testing.T) {
 	mem := uint32(1000)
+	maxMem := memory.TotalMemory() / 1024 / 1024 //MiB
 	slots := uint32(8)
 	assert := assert.New(t)
 
 	q := &qemu{
 		arch: &qemuArchBase{},
 		config: HypervisorConfig{
-			MemorySize: mem,
-			MemSlots:   slots,
+			MemorySize:           mem,
+			MemSlots:             slots,
+			DefaultMaxMemorySize: maxMem,
 		},
 	}
 
-	hostMemKb, err := GetHostMemorySizeKb(procMemInfo)
-	assert.NoError(err)
-	memMax := fmt.Sprintf("%dM", int(float64(hostMemKb)/1024))
+	memMax := fmt.Sprintf("%dM", int(maxMem))
 
 	expectedOut := govmmQemu.Memory{
 		Size:   fmt.Sprintf("%dM", mem),