def greedy_LTD_start():
'''
Shortcut: called by the user, in order to retrieve several solutions and compare them each other
Parameters used to create the graphs are take as input from the user
'''
# Number of nodes
n = inc.input_int('Number of nodes', minValue = 1)
# Extreme values for the traffic matrix
TM_min = inc.input_int('Traffic matrix lower bound', minValue = 1)
TM_max = inc.input_int('Traffic matrix upper bound', minValue = TM_min)
# Delta values
delta_in = inc.input_int('Delta_in (max #rx per node)', minValue = 1)
delta_out = inc.input_int('Delta_out (max #tx per node)', minValue = 1)
# Traffic matrix
traffic_matrix = tm.random_TM(n, TM_min, TM_max)
# Results:
T1 = greedy_LTD_mesh(n, traffic_matrix, delta_in, delta_out)
T1_bis = LTD_random(n, len(T1.edges()), delta_in, delta_out, traffic_matrix)
T2 = greedy_LTD_ring(n, traffic_matrix, delta_in, delta_out)
T2_bis = LTD_random(n, len(T2.edges()), delta_in, delta_out, traffic_matrix)
for i in range(len(ns)):
# Number of nodes in the topology
n = ns[i]
# Retrieve number of nodes per row/column
nr = nrs[i]
nc = ncs[i]
# Estimated max flow, for every N
est_f_max_A = 0.0
est_f_max_B = 0.0
# Estimated computational time, for each N
est_time_A = 0.0
est_time_B = 0.0
# FOr every N-nodes topology, repeat th simulation several times
for s in range(n_simulations):
# Compute traffic matrix (same for A and B)
traffic_matrix = tm.random_TM(n, 0.5, 1.5)
# Experiment number
exp = 'Exp #%s, Sim #%s - ' % (str(t).zfill(2),
str(s).zfill(2))
# Information strings
exp_info = '%sN = %d, Nr = %d, Nc = %d' % (exp, n, nr, nc)
title = '%sManhattan LTD' % (exp)
print('\n\n%s' % (exp_info))
# SOLUTION
# Non-optimized
T_A = L2.LTD_manhattan(n, nr, nc, traffic_matrix, title, False,
False)
est_f_max_A += T_A['max_flow']
est_time_A += T_A['time']
# Optimized
title = '%sManhattan LTD Smart' % (exp)