Re: [RPRWG] Complete RPR/Java model for all three proposals
Unfortunately I have not come very far in simulating all three
proposals.
I have however put on my web site a complete Java simulation program.
It is a ring with 16 nodes, and have two classes of priority and
two sets of transit buffers.
The setup sends (very high load) random low prio. traffic between
all nodes, and one stream of (high prio.) TDM traffic between
two nodes.
The algorithm is mostly the one I presented at the September meeting
It uses N VOQs and N "choke"-points (where N+1 is the number of
stations on the ring). It uses some of the same low pass filters
that Gandalf is using (but with more agressive parameters).
http://www.ifi.uio.no/~steing/rpr/java-jan02
Stein