BBR
also: Bottleneck Bandwidth and Round-trip time
BBR (Bottleneck Bandwidth and Round-trip time) is a modern TCP congestion control algorithm that optimizes network throughput by measuring available bandwidth and latency instead of relying on packet loss.
BBR is a congestion control algorithm developed by Google that fundamentally changes how TCP manages data flow over networks. Unlike older algorithms (like Reno or CUBIC) that interpret packet loss as a signal to slow down, BBR proactively measures the bottleneck bandwidth and round-trip time of a connection to determine the optimal sending rate.
BBR operates in two phases: it first estimates the bottleneck bandwidth by measuring how fast data can be sent, then maintains a small queue of data in flight proportional to bandwidth × latency. This approach delivers better performance on high-latency, high-bandwidth links—particularly useful for satellite connections, mobile networks, and intercontinental transfers.
To enable BBR on Linux, you modify the TCP congestion control algorithm: echo bbr | sudo tee /proc/sys/net/ipv4/tcp_congestion_control. BBR has been the default congestion control algorithm in the Linux kernel since version 5.18, though it can be explicitly selected on older kernels.