It's available in Ubuntu 17.04. Add these lines to /etc/sysctl.conf:

    net.core.default_qdisc=fq
    net.ipv4.tcp_congestion_control=bbr

I got 400Mbps (on a 1Gbps link) from Seattle to New York with a single TCP-BBR stream, vs ~50Mbps before :-).

You could probably saturate that link with a single TCP stream -- even without BBR -- by simply choosing the correct send and receive buffer sizes in your test program. (I assume you're using iperf or a similar program to perform your tests.)

BBR and other congestion-control algorithms come into play when the underlying IP network is congested. To test them, you have to have control of the congestion rate of the underlying circuit, which you might not have if you're testing this on an ordinary public IP network.

Thank you! This is helpful also about fq vs fq_codel - https://groups.google.com/forum/m/#!topic/bbr-dev/4jL4ropdOV...

How are you measuring this?

To get this on Ubuntu 16.04 LTS, get and install the following:

  wget kernel.ubuntu.com/~kernel-ppa/mainline/v4.9.26/linux-headers-4.9.26-040926_4.9.26-040926.201705031231_all.deb kernel.ubuntu.com/~kernel-ppa/mainline/v4.9.26/linux-image-4.9.26-040926-generic_4.9.26-040926.201705031231_i386.deb kernel.ubuntu.com/~kernel-ppa/mainline/v4.9.26/linux-image-4.9.26-040926-generic_4.9.26-040926.201705031231_amd64.deb
  sudo dpkg -i linux-headers-4.9*.deb linux-image-4.9*.deb

It's in kernel 4.9

The code is already in net-next. This means you currently have to build your own kernel if you want to use it. This should be enough to get you going: https://github.com/google/bbr/blob/master/Documentation/bbr-...