[e2e] How can I improve the converge time of route simulation procedure

[Jing Shen]

Hi,

I'm trying to simulate some routing algorithms with a event driven simulator. MCI backbone topology
and some of topology to Waxman's result is used. But, we find the simulation time is so long that
it may last for 2-3 days! The amount of connection request may be over 550,0000! If we always take this
way, the experiment may last for month! 

Would you please give some indication on how to decrease the converge time of simulation. (we use 
a confidence level 0.99 and confidence interval 0.05 )

Each word will be  highly appreciated.

Regards

Jing Shen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.postel.org/pipermail/end2end-interest/attachments/20020411/aa541701/attachment.html