Skip to content

Testing network bandwidth between local web browser and Rancher management cluster

Article Number: 000021608

Environment

Rancher management cluster

Situation

The Tuning and Best Practices for SUSE® Rancher Prime at Scale documentation has a Browser Requirements section which has a recommendation for network bandwidth to the Rancher management cluster. How can I collect this data in my environment to compare to the recommendation?

Resolution

Collection

  1. install the  iperf3 tool (on macOS, use brew install iperf3)
  2. run the following commands from a machine with kubectl access to the upstream Rancher cluster: 

kubectl run iperf3-server --image=networkstatic/iperf3 --port=5201 --command -- iperf3 -s
kubectl port-forward pod/iperf3-server 5201:5201 &
iperf3 -c localhost -p 5201
3. The output will look like this: 

Connecting to host localhost, port 5201
[  7] local ::1 port 57800 connected to ::1 port 5201
[ ID] Interval           Transfer     Bitrate
[  7]   0.00-1.00   sec   121 MBytes  1.01 Gbits/sec
[  7]   1.00-2.00   sec   111 MBytes   933 Mbits/sec
[  7]   2.00-3.00   sec   110 MBytes   921 Mbits/sec
[  7]   3.00-4.00   sec   110 MBytes   925 Mbits/sec
[  7]   4.00-5.00   sec   110 MBytes   924 Mbits/sec
[  7]   5.00-6.01   sec   110 MBytes   921 Mbits/sec
[  7]   6.01-7.00   sec   110 MBytes   924 Mbits/sec
[  7]   7.00-8.00   sec   110 MBytes   925 Mbits/sec
[  7]   8.00-9.01   sec   110 MBytes   927 Mbits/sec
[  7]   9.01-10.01  sec   103 MBytes   867 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate
[  7]   0.00-10.01  sec  1.08 GBytes   928 Mbits/sec                  sender
[  7]   0.00-10.08  sec  1.07 GBytes   915 Mbits/sec                  receiver

iperf Done

Interpretation

The last two table rows, under the Bitrate column, represent actual raw bandwidth.

For optimal Rancher performance at scale, SUSE recommends at least 72 Mbits/s "receiver" bandwidth (equivalent to a single 802.11n Wi-Fi 4 link stream).