All things related to multicast

[Prashant Rajvaidya <>]

> Am I correct in thinking that the sudden peaks showing up in the CAIDA
> Mantra monitoring graphs are a result of the recent storms? I didn't keep
> track of the exact times that were reported, but they look about right.

yep! peaks in the plot for number of groups (in "Session
Monitoring" section of mantra-www) are because of the storm. we generate
these results on the basis of information from mroute-tables as well as
msdp-cache. most of the times, especially in stormY situations,
strong influence of statistics from msdp-cache is well evident in the
plots.

> The main page is http://www.caida.org/tools/measurement/mantra/, and the
> most interesting graphs are the session monitors, at
> http://www.caida.org/tools/measurement/Mantra/session-mon/session-mon.html

i think that the plots available in "MSDP Monitor" section of mantra-www
are pretty interesting as well. statistics depicted in the plots
available here are specific to msdp. following is the url
http://www.caida.org/tools/measurement/Mantra/msdp-mon/msdp-mon.html

results available through the main mantra pages represent an aggregated
view; i.e. we aggregate snapshots obtained from the point of view of
different routers. 

it is interesting to see how spikes differ from router to router (tells a
lot about propagation of SA messages, as well). i am soon going to put up
a page with all the results relevant to "msdp-storm" on mantra mirrors.

*note: in addition to mantra-www at www.caida.org, a mirror is also
available at http://imj.ucsb.edu/mantra/

--prashant


 ,_,
(O,O)
(   ) "there is no spoon!"
-"-"-------------------------------------------------------------
Prashant Rajvaidya    |  Department of Computer Science
Ph.D. Grad -- "slave" |  University of California, Santa Barbara
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
tel(O): 805-893-7520 (Networking and Multimedia Systems Lab)
-----------------------------------------------------------------