Bandwidth Manager (beta)

This guide explains how to configure Cilium’s bandwidth manager to optimize TCP and UDP workloads and efficiently rate limit individual Pods if needed through the help of EDT (Earliest Departure Time) and eBPF.

The bandwidth manager does not rely on CNI chaining and is natively integrated into Cilium instead. Hence, it does not make use of the bandwidth CNI plugin. Due to scalability concerns in particular for multi-queue network interfaces, it is not recommended to use the bandwidth CNI plugin which is based on TBF (Token Bucket Filter) instead of EDT.

Cilium’s bandwidth manager supports the Pod annotation which is enforced on egress at the native host networking devices. The bandwidth enforcement is supported for direct routing as well as tunneling mode in Cilium.

The annotation is not supported and also not recommended to use. Limiting bandwidth happens natively at the egress point of networking devices in order to reduce or pace bandwidth usage on the wire. Enforcing at ingress would add yet another layer of buffer queueing right in the critical fast-path of a node via ifb device where ingress traffic first needs to be redirected to the ifb’s egress point in order to perform shaping before traffic can go up the stack. At this point traffic has already occupied the bandwidth usage on the wire, and the node has already spent resources on processing the packet. annotation is ignored by Cilium’s bandwidth manager.


Bandwidth Manager requires a v5.1.x or more recent Linux kernel.


First, make sure you have Helm 3 installed.

If you have (or planning to have) Helm 2 charts (and Tiller) in the same cluster, there should be no issue as both version are mutually compatible in order to support gradual migration. Cilium chart is targeting Helm 3 (v3.0.3 and above).

Download the Cilium release tarball and change to the kubernetes install directory:

curl -LO
tar xzf master.tar.gz
cd cilium-master/install/kubernetes

As a beta feature, the Cilium bandwidth manager is disabled by default for new deployments via Helm. The option is controllable through bandwidthManager with a possible setting of true and false (default).

helm install cilium ./cilium \
  --namespace kube-system \
  --set bandwidthManager=true

The native host networking devices are auto detected as native devices which have the default route on the host or have Kubernetes InternalIP or ExternalIP assigned. InternalIP is preferred over ExternalIP if both exist. To change and manually specify the devices, set their names in the devices helm option, e.g. devices='{eth0,eth1,eth2}'. Each listed device has to be named the same on all Cilium managed nodes.

Verify that the Cilium Pod has come up correctly:

kubectl -n kube-system get pods -l k8s-app=cilium
NAME                READY     STATUS    RESTARTS   AGE
cilium-crf7f        1/1       Running   0          10m

In order to verify whether the bandwidth manager feature has been enabled in Cilium, the cilium status CLI command provides visibility through the BandwidthManager info line. It also dumps a list of devices on which the egress bandwidth limitation is enforced:

kubectl exec -it -n kube-system cilium-xxxxx -- cilium status | grep BandwidthManager
BandwidthManager:       EDT with BPF   [eth0]

Assuming we have a multi-node cluster, in the next step, we deploy a netperf Pod on the node named foobar. The following example deployment yaml limits the egress bandwidth of the netperf Pod on the node’s physical device:

apiVersion: apps/v1
kind: Deployment
  name: netperf
      run: netperf
  replicas: 1
        run: netperf
      annotations: "10M"
      nodeName: foobar
      - name: netperf
        image: cilium/netperf
        - containerPort: 12865

Once up and running, netperf client can be invoked from a different node in the cluster using the Pod’s IP (in this case directly. The test streaming direction is from the netperf deployment towards the client, hence TCP_MAERTS:

netperf -t TCP_MAERTS -H
MIGRATED TCP MAERTS TEST from ( port 0 AF_INET to () port 0 AF_INET
Recv   Send    Send
Socket Socket  Message  Elapsed
Size   Size    Size     Time     Throughput
bytes  bytes   bytes    secs.    10^6bits/sec

 87380  16384  16384    10.00       9.56

As can be seen, egress traffic of the netperf Pod has been limited to 10Mbit per second.

In order to introspect current endpoint bandwidth settings from BPF side, the following command can be run:

kubectl exec -it -n kube-system cilium-xxxxx -- cilium bpf bandwidth list
491        10M

Each Pod is represented in Cilium as an Endpoint which has an identity. The above identity can then be correlated with the cilium endpoint list command.


  • Bandwidth enforcement currently does not work in combination with L7 Cilium Network Policies. In case they select the Pod at egress, then the bandwidth enforcement will be disabled for those Pods.