rtt measurement in loss recovery state
when we use rtt for pacing. the rtt measurement is usually too conservative for loss recovery state.
posted by David at 1:14 AM
walk in the networks
something related to my graduate study
Friday, September 02, 2005
Previous Posts
- "strange case 2" in FACK.
- paced tcp suffers from more timeout?
- default queue-limit for each queue is 50pkt
- inconsistency between delsnd timer and burstsnd ti...
- double check some results
- Yao's techniqe
- plan for the rest of this week
- Add overhead on the processing time
- a summary of the pace vs non-paced issue
- graph summary
0 Comments:
Post a Comment
<< Home