Re: [quicwg/base-drafts] Persistent Congestion Time Threshold (#2365)

[Nick Banks <[e-mail]>]

nibanks commented on this pull request.



>  
      // Start a new congestion epoch if the last lost packet
      // is past the end of the previous recovery epoch.
-     CongestionEvent(largest_lost_packet.time_sent)
+     CongestionEvent(newest_lost_packet.time_sent)
+
+     // Collapse congestion window if persistent congestion
+     if (InPersistentCongestion(oldest_lost_packet.time_sent)):

Let me propose an example. Assume smooth_rtted = 1 sec, rttvar = 0, max_ack_delay = 0 for simplicity's sake.
```
t=0 | Send Pkt #1 (App Data)
t=1 | Send Pkt #2 (PTO 1)
t=3 | Send Pkt #3 (PTO 2)
t=7 | Send Pkt #4 (PTO 3)
t=8 | Recv ACK of Pkt #4
```
At t=8 you'd get a congestion event for packets 1 & 2 being lost. If you use Now() for the persistent congestion calculation, this would be declared as persistent congestion. If you used newest_lost_time it wouldn't. Which behavior do we want?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/pull/2365#discussion_r254472723