[Tmrg] (limited) measurement of file size vs congestion level

[ricciato]

Hi Sally, all

[I am new to the list]

just an humble comment on the relationship between congestion and user 
behaviour.

1. there is a very preliminary study [1] investigating the issue

2. we have found and reported in [2] a case of severe congestion on 
"our" network, i.e. a mobile UMTS network that we extensively and 
constantly monitor in our project [3]. There we show that the presence 
of the bottleneck changes the statistics of the aggregate traffic, of 
course.
An analysis of the impact of user-behaviour (change in file-size, 
abandoning, re-clicks, etc.) based on the detailed packet-traces was 
always in our to-do list, but so far had not time to work that out  
(there are some complications in doing user-level analysis, e.g. the 
file-size does not correspond to the TCP connection size, as the 
relationship file:connections is NOT 1:1 in modern applications, we, p2p 
etc.)

3. However, based on our experience (we have seen many severe congestion 
events in this network), I can report the following qualitative 
observations

   A.  the user abandoning  process seems to be "with threshold" : if 
you consider the frequency of TCP RST as a gross indicator of user (or 
server) impatience, we saw that for mild congestion (right before the 
peak hour, on a congested link) the RST stay at physiological level 
(pretty low), while it sharply jumps to abnormally high values when the 
congestion becomes severe (during the peak hour)

  B. if you look at the distribution of the number of packets downloaded 
by each users in fixed timebins (e.g. 1 min), you see that after a 
capacity upgrade that removes a congestion points, such distribution 
changes, with more user downloading more packets (as expected).


4. My expectation is that the users regulate the  duration of the 
*session* and the total download rate (often across multiple parallel 
TCP connection) based on the experienced response time, there it is the 
*session* attributes (duration, rate), rather than the *file* ones, that 
are dependent on the congestion level. At the TCP level, this might 
means that it´s the connection arrival process that is mostly impacted, 
rtaher than the size (the latter is probably affected only in the tail 
of long files, which are probably truncated upon congestion).
Furthermore, after a certain threshold (severe congestion), users or 
servers suddenly get crazy and start to reclick/reset the downloads, and 
eventually give up the session.

ciao
fabio

[1] "User patience and the Web: a hands-on investigation", by Rossi, 
Casetti, Mellia, @ Globecom 2003.
[2] F. Ricciato, F. Vacirca, P. Svoboda, Diagnosis of Capacity 
Bottlenecks via Passive Monitoring in 3G Networks: an Empirical 
Analysis, Computer Networks, vol. 51, n.4, pp. 1205-1231, March 2007
[3]  http://userver.ftw.at/~ricciato/darwin/


Sally Floyd wrote:
> Lachlan -
>
> As an additional comment, my assumption would be that users' behavior
> is *heavily* affected by the level of congestion, but I don't know
> that much research has been done on this.  That is, I would assume
> that users would cut their Internet browsing short in times of heavy
> congestion (i.e., of slow response times and long download times),
> in terms of the number of TCP connections initiated.
>
> I wouldn't have had a good guess, however, what this would mean to
> the distribution of connection sizes.   It might mean different
> things on different type links.
>
> - Sally
> http://www.icir.org/floyd/
>
> _______________________________________________
> Tmrg-interest mailing list
> Tmrg-interest at ICSI.Berkeley.EDU
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/tmrg-interest
>   



Sally Floyd wrote:
> Lachlan -
>
> As an additional comment, my assumption would be that users' behavior
> is *heavily* affected by the level of congestion, but I don't know
> that much research has been done on this.  That is, I would assume
> that users would cut their Internet browsing short in times of heavy
> congestion (i.e., of slow response times and long download times),
> in terms of the number of TCP connections initiated.
>
> I wouldn't have had a good guess, however, what this would mean to
> the distribution of connection sizes.   It might mean different
> things on different type links.
>
> - Sally
> http://www.icir.org/floyd/
>
> _______________________________________________
> Tmrg-interest mailing list
> Tmrg-interest at ICSI.Berkeley.EDU
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/tmrg-interest
>