nanog mailing list archives
Re: Trident3 vs Jericho2
From: Vincent Bernat <bernat () luffy cx>
Date: Fri, 09 Apr 2021 16:38:24 +0200
❦ 9 avril 2021 17:20 +03, Saku Ytti:
If we'd change TCP sender to bandwidth estimation, and newly created window space would be serialised at estimated receiver rate then we would need dramatically less buffers. However this less aggressive TCP algorithm would be outcompeted by new reno reducing bandwidth estimation to approach zero. Luckily almost all traffic is handled by few players, if they agree to change to well behaved TCP (or QUIC) algorithm, it doesn't matter much if the long tail is badly behaving TCP.
I think many of them are now using BBR or BBR v2. It would be interesting to know how it impacted switch buffering. -- As flies to wanton boys are we to the gods; they kill us for their sport. -- Shakespeare, "King Lear"
Current thread:
- Trident3 vs Jericho2 Dmitry Sherman (Apr 09)
- Re: Trident3 vs Jericho2 Mike Hammett (Apr 09)
- Re: Trident3 vs Jericho2 Tom Beecher (Apr 09)
- Re: Trident3 vs Jericho2 Mike Hammett (Apr 09)
- Re: Trident3 vs Jericho2 Saku Ytti (Apr 09)
- Re: Trident3 vs Jericho2 Vincent Bernat (Apr 09)
- Re: Trident3 vs Jericho2 Tom Beecher (Apr 09)
- Re: Trident3 vs Jericho2 Mike Hammett (Apr 09)
- Re: Trident3 vs Jericho2 William Herrin (Apr 09)
- Re: Trident3 vs Jericho2 lobna gouda (Apr 09)