mirror of
https://github.com/LibreQoE/LibreQoS.git
synced 2025-01-08 21:43:01 -06:00
Jitter plot
This commit is contained in:
parent
b1b6605b37
commit
88d191604d
@ -84,6 +84,11 @@ the AQM component, doesn't work as well as it could.
|
||||
|
||||
[[./img/middlebox_compared.png]]
|
||||
|
||||
Putting cake on the CPE also results in a metric ton less jitter.
|
||||
|
||||
[[./img/tonlessjitter.png]]
|
||||
|
||||
|
||||
Similarly, many devices are themselves the bottleneck, still, so they
|
||||
accumulate a ton of RTT themselves, and monitoring the RTT and doing
|
||||
something about it would possibly help.
|
||||
@ -105,7 +110,6 @@ worse over time. The only thing we have for quic is the: [[https://www.ietfjourn
|
||||
which is probably [[https://http3-explained.haxx.se/en/quic/quic-spinbit][not widely implemented]]. The best insight we actually
|
||||
have is queue accomulation and packet drop/mark behaviors.
|
||||
|
||||
|
||||
** HTB is bursty
|
||||
|
||||
** DSCP
|
||||
@ -114,4 +118,6 @@ What DSCPs are in common use today?
|
||||
|
||||
** DROP_MONITOR
|
||||
|
||||
We have 2600 (not kidding, 2600) places where packet
|
||||
We have 2600 (not kidding, 2600) places where packets can be dropped
|
||||
in the kernel, not to mention other parts of the network. A nice
|
||||
feature would be to be able to track retransmits relative to drops.
|
||||
|
BIN
sim/img/tonlessjitter.png
Normal file
BIN
sim/img/tonlessjitter.png
Normal file
Binary file not shown.
After Width: | Height: | Size: 47 KiB |
Loading…
Reference in New Issue
Block a user