diff --git a/README.md b/README.md
index c82bcab..22d7c5b 100644
--- a/README.md
+++ b/README.md
@@ -64,6 +64,7 @@ The default telemd runs the following instruments:
 * `kubernetes_cgrp_cpu` the cpu usage time of individual Kubernetes Pod containers
 * `kubernetes_cgrp_blkio` the total block io usage in bytes for individual Kubernetes Pod containers
 * `kubernetes_cgrp_memory` the total memory (RAM) usage in bytes for individual Kubernetes Pod containers
+* `kubernetes_cgrp_net` the total network io usage in bytes for individual Kubernetes Pod containers
 
 ### Info keys
 
diff --git a/internal/telemd/cfg.go b/internal/telemd/cfg.go
index d192ea6..f98251a 100644
--- a/internal/telemd/cfg.go
+++ b/internal/telemd/cfg.go
@@ -77,6 +77,7 @@ func NewDefaultConfig() *Config {
 		"kubernetes_cgrp_cpu":    1 * time.Second,
 		"kubernetes_cgrp_blkio":  1 * time.Second,
 		"kubernetes_cgrp_memory": 1 * time.Second,
+		"kubernetes_cgrp_net":    1 * time.Second,
 	}
 
 	return cfg
diff --git a/internal/telemd/daemon.go b/internal/telemd/daemon.go
index 96631dd..b16f468 100644
--- a/internal/telemd/daemon.go
+++ b/internal/telemd/daemon.go
@@ -53,6 +53,7 @@ func (daemon *Daemon) initInstruments(factory InstrumentFactory) {
 		"kubernetes_cgrp_cpu":    factory.NewKubernetesCgroupCpuInstrument(),
 		"kubernetes_cgrp_blkio":  factory.NewKubernetesCgroupBlkioInstrument(),
 		"kubernetes_cgrp_memory": factory.NewKubernetesCgroupMemoryInstrument(),
+		"kubernetes_cgrp_net":    factory.NewKubernetesCgroupNetInstrument(),
 	}
 
 	activeNetDevice, err := findActiveNetDevice()
diff --git a/internal/telemd/instrument.go b/internal/telemd/instrument.go
index d1ab047..852ccbb 100644
--- a/internal/telemd/instrument.go
+++ b/internal/telemd/instrument.go
@@ -33,6 +33,7 @@ type InstrumentFactory interface {
 	NewDockerCgroupMemoryInstrument() Instrument
 	NewKubernetesCgroupBlkioInstrument() Instrument
 	NewKubernetesCgroupMemoryInstrument() Instrument
+	NewKubernetesCgroupNetInstrument() Instrument
 	NewPsiCpuInstrument() Instrument
 	NewPsiMemoryInstrument() Instrument
 	NewPsiIoInstrument() Instrument
@@ -75,12 +76,16 @@ type DockerCgroupv1NetworkInstrument struct {
 type DockerCgroupv2NetworkInstrument struct {
 	pids map[string]string
 }
+
 type DockerCgroupv1MemoryInstrument struct{}
 type DockerCgroupv2MemoryInstrument struct{}
 
 type KubernetesCgroupCpuInstrument struct{}
 type KubernetesCgroupBlkioInstrument struct{}
 type KuberenetesCgroupMemoryInstrument struct{}
+type KubernetesCgroupv1NetworkInstrument struct {
+	pids map[string]string
+}
 
 func (CpuUtilInstrument) MeasureAndReport(channel telem.TelemetryChannel) {
 	then := readCpuUtil()
@@ -531,6 +536,53 @@ func (c *DockerCgroupv1NetworkInstrument) MeasureAndReport(channel telem.Telemet
 	}
 }
 
+func (c KubernetesCgroupv1NetworkInstrument) MeasureAndReport(channel telem.TelemetryChannel) {
+	var kubepodRootDir = "/sys/fs/cgroup/cpuacct/kubepods"
+	var bestEffortDir = kubepodRootDir + "/" + "besteffort"
+	var burstableDir = kubepodRootDir + "/" + "burstable"
+	var guaranteedDir = kubepodRootDir + "/" + "guaranteed"
+
+	for _, kubepodDir := range [3]string{bestEffortDir, burstableDir, guaranteedDir} {
+
+		go func(kubepodDir string) {
+			for _, containerDir := range fetchKubernetesContainerDirs(kubepodDir) {
+				containerId := filepath.Base(containerDir)
+
+				pid, ok := c.pids[containerId]
+
+				if !ok {
+					// refresh pids
+					pids, err := containerProcessIds()
+					if err != nil {
+						log.Println("unable to get container process ids", err)
+						continue
+					}
+					c.pids = pids
+
+					pid, ok = c.pids[containerId]
+					if !ok {
+						log.Println("could not get pid of container after refresh", containerId)
+						continue
+					}
+				}
+
+				rx, tx, err := readTotalProcessNetworkStats(pid)
+				if err != nil {
+					if os.IsNotExist(err) {
+						delete(c.pids, containerId) // delete now and wait for next iteration to refresh
+					} else {
+						log.Println("error parsing network stats of pid", pid, err, err)
+					}
+					continue
+				}
+
+				channel.Put(telem.NewTelemetry("kubernetes_cgrp_net/"+containerId[:12], float64(rx+tx)))
+			}
+		}(kubepodDir)
+	}
+
+}
+
 func (c *DockerCgroupv2NetworkInstrument) MeasureAndReport(channel telem.TelemetryChannel) {
 	prefix := "docker-"
 
@@ -932,6 +984,9 @@ func (d defaultInstrumentFactory) NewKubernetesCgroupBlkioInstrument() Instrumen
 func (d defaultInstrumentFactory) NewKubernetesCgroupMemoryInstrument() Instrument {
 	return KuberenetesCgroupMemoryInstrument{}
 }
+func (d defaultInstrumentFactory) NewKubernetesCgroupNetInstrument() Instrument {
+	return KubernetesCgroupv1NetworkInstrument{}
+}
 
 func (d defaultInstrumentFactory) NewDockerCgroupMemoryInstrument() Instrument {
 	cgroup := checkCgroup()
diff --git a/internal/telemd/sysparse.go b/internal/telemd/sysparse.go
index 7cec428..92c2125 100644
--- a/internal/telemd/sysparse.go
+++ b/internal/telemd/sysparse.go
@@ -266,6 +266,10 @@ func getContainerId(pid string) (string, error) {
 				// cgroup v1
 				return split, nil
 			}
+		} else if strings.Contains(line, "kubepods") {
+			// kubernetes
+			// freezer:/kubepods/besteffort/podae778fdf-394c-4356-9625-ea50666783b1/2cc54a6877a50da0b6a2a5340dd1e8c5707a1d7d4b363e03b7cde76d2569f0c0
+			return strings.Split(line, "/")[4], nil
 		}
 	}
 	return "", errors.New("Did not find container for PID " + pid)