The MPLS-OPS Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] RE: Topology to LP converter
For post of
Date: Sat, 29 Nov 2003 20:57:26 -0500 (EST)
Errata:
For
Then we ask for values of n variables x to make
'objective function' f(x) as small as possible while
maintaining g(x) >= 0 (each of the m values of g(x) >=
0).
it would read better with
Then we ask for values of n variables x to make
'objective function' f(x) as small as possible while
maintaining g(x) >= 0 (each of the m values of g(x) is >=
0).
For
Experience showed that the simplex algorithm is
shockingly fast, in practice commonly requiring about 2
to 3 times m 'iterations' (where each iteration takes
about m**3 arithmetic operations) for a solution.
have m in one too many places. Could say that the algorithm
takes 2 to 3 times m**3 arithmetic operations. Or could say
Experience showed that the simplex algorithm is
shockingly fast, in practice commonly requiring about 2
to 3 times m 'iterations' (where each iteration takes
about m**2 arithmetic operations) for a solution.
For
Guess at the crucial angle; assign cost 1 per unit of
flow for all arcs with angle less than this critical
angle and 0 for the rest, and solve the min cost flow
problem for total flow amount n and with no more than 1
unit of flow on each arc.
change to
Guess at the crucial angle; assign cost 1 per unit of
flow for all arcs with angle greater than this critical
angle and 0 for the rest, and solve the min cost flow
problem for total flow amount n and with no more than 1
unit of flow on each arc.
Norman B. Waite, Ph.D.
Founder
Network Architectonics
9 Fox Run
Wappingers Falls, NY 12590
nbwaite@attglobal.net
845-227-7821
-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
|
|