scaling.readme

R. Lo Cigno, P. Larcheri 802.11e closed-loop scheduling Description: ns-2 package for 802.11e clo

# Created by Polly Huang, USC/ISI
# http://www-scf.usc.edu/~bhuang
# 7/16/98

What's the story about session sim's improvement on memory consumption?

The bottleneck has been determined to be the routing table, which is
O(n^2) for flat routing. 

In addition, n (number of routing entries) is always 2^k. That is why 
we observe roughly the same routing table size for 513-node and 
1024-node simulations (k=10).

The following table lists _approximately_ the memory consumption of
session sim with various memory conservation solutions.

# nodes in topo		Mem consumption (in MB)
		session		with delay_bind		with hierarchial
		connectivity ~1.8			connectivity ~3-4
512-1023	   16		  10			  
1024-2047	   46		  40			  16 (1040 node)
2048-4095	  180		 160			  40 (2080 node)
4096-8191	  720		 640			 169 (5120 node)
8192-16384	 2886		2560			1049 (10075 node)

The hierarchial and flat routing simulations use different topologies 
with similar number of nodes. We're trying to unify/simplify the API 
for specifying flat or hierarchial routing and to provide benchmark 
topologies for future evaluation.


What's the largest topology ever simulated?

By 7/16/98 11:15 EDT, the largest topology ever simulated, using 
SessionSim, has 10,000 nodes.