UDP Buffer Sizes
As of quic-go v0.19.x, you might see warnings about the receive and send buffer sizes.
Experiments have shown that QUIC transfers on high-bandwidth connections can be limited by the size of the UDP receive and send buffer. The receive buffer holds packets that have been received by the kernel, but not yet read by the application (quic-go in this case). The send buffer holds packets that have been sent by quic-go, but not sent out by the kernel. In both cases, once these buffers fill up, the kernel will drop any new incoming packet.
Therefore, quic-go tries to increase the buffer size. The way to do this is an OS-specific, and we currently have an implementation for linux, windows and darwin. However, an application is only allowed to do increase the buffer size up to a maximum value set in the kernel. Unfortunately, on Linux this value is rather small, too small for high-bandwidth QUIC transfers.
It is recommended to increase the maximum buffer size by running:
sysctl -w net.core.rmem_max=7500000
sysctl -w net.core.wmem_max=7500000
This command would increase the maximum send and the receive buffer size to roughly 7.5 MB. Note that these settings are not persisted across reboots.
Taken from: https://medium.com/@CameronSparr/increase-os-udp-buffers-to-improve-performance-51d167bb1360
On BSD/Darwin systems you need to add about a 15% padding to the kernel limit socket buffer. Meaning if you want a 25MB buffer (8388608 bytes) you need to set the kernel limit to 26214400*1.15 = 30146560. This is not documented anywhere but happens in the kernel here.
To update the value immediately to 7.5 MB, type the following commands as root:
sysctl -w kern.ipc.maxsockbuf=8441037Add the following lines to the /etc/sysctl.conf file to keep this setting across reboots:
kern.ipc.maxsockbuf=8441037