[tor-bugs] #4486 [Analysis]: Research: should N23 actually help in practice?

[Tor Bug Tracker & Wiki]

#4486: Research: should N23 actually help in practice?
-------------------------------------+--------------------------------------
 Reporter:  arma                     |          Owner:  robgjansen
     Type:  task                     |         Status:  new       
 Priority:  normal                   |      Milestone:            
Component:  Analysis                 |        Version:            
 Keywords:  performance flowcontrol  |         Parent:  #4506     
   Points:                           |   Actualpoints:            
-------------------------------------+--------------------------------------
Changes (by robgjansen):

  * status:  needs_information => new


Comment:

 Replying to [comment:39 arma]:

 > Does n23 look better when you load down the network a lot more,

 The
 [https://trac.torproject.org/projects/tor/attachment/ticket/4486/20121113-apollo-n23
 -heavy-combined.pdf 'heavy load' graphs are here]. There were 1.5 times
 the number of clients as in the experiments that produced
 [https://trac.torproject.org/projects/tor/attachment/ticket/4486/20121025-apollo-n23-combined.pdf
 the previous 'normal load' set of graphs], though I noticed that the
 overall load was nowhere near 1.5 times. It again seems like N3Initial50
 is better for first byte, as well as last byte of 50K downloads, but
 download times again worsen for 1M and 5M files.

 I'm skeptical of the results given our low confidence that the N23 code
 actually does what we think it should do.

 > or when you make some client first links crappy (#4487)?

 Running that now (despite my comment about code confidence), and will
 report results in #4487.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/4486#comment:41>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online