ProjectB-RedundancyInQueueingNetworks/parSim.py at master · yoniyes/ProjectB-RedundancyInQueueingNetworks · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
import time
import datetime
import os
import numpy as np
import multiprocessing
import matplotlib.pyplot as plt
import QueueNetworkSimulation as qns


def poolInit(q):
    singleRun.q = q


def singleRun(args):
    (simObj, arrivalRate, effectiveServiceRate, result_num) = args
    try:
        return [result_num, arrivalRate, simObj[0].sims[result_num].singleRun(arrivalRate, effectiveServiceRate)]
    except Exception:
        print('arrivalRate, result_num = %f, %d' % (arrivalRate, result_num))


class parSim(qns.QueueNetworkSimulation):

    def __init__(self, size, dispatchPolicyStrategy, convergenceConditionStrategy, plotStrategy=None, services=[],
                 workloads=[], historyWindowSize=10000, numOfRounds=100, verbose=False, T_min=0, T_max=10000000,
                 guess=False):
        qns.QueueNetworkSimulation.__init__(self, size, dispatchPolicyStrategy, convergenceConditionStrategy,
                                            plotStrategy, services, workloads, historyWindowSize, numOfRounds, verbose,
                                            T_min, T_max, guess)
        self.size = size
        self.sims = [qns.QueueNetworkSimulation(size, dispatchPolicyStrategy, convergenceConditionStrategy,
                                     plotStrategy=plotStrategy, services=services, workloads=workloads,
                                     historyWindowSize=historyWindowSize, numOfRounds=numOfRounds, verbose=verbose,
                                     T_min=T_min, T_max=T_max, guess=guess)
                     for k in range(numOfRounds)]
        self.results = [[0.0] * self.numOfRounds for i in range(2)]

    def reset(self):
        for simu in self.sims:
            simu.reset()
        self.results = [[0.0] * self.numOfRounds for i in range(2)]

    def parRun(self):
        if __name__ == '__main__':
            result_queue = multiprocessing.Queue()
            starttime = time.time()
            start_time = datetime.datetime.now()

            numOfRounds = len(self.sims)
            effectiveServiceRate = self.sims[0].getEffectiveServiceRate()
            arrivalRate = [0.0] * numOfRounds
            if effectiveServiceRate != 0:
                arrivalRate = np.arange(0, effectiveServiceRate, float(effectiveServiceRate) / float(numOfRounds))
            # self.results[0] = arrivalRate

            args = [([self], arrivalRate[i], effectiveServiceRate, i) for i in range(numOfRounds-1, -1, -1)]

            pool = multiprocessing.Pool(processes=8, initializer=poolInit, initargs=[result_queue])
            res = pool.imap(singleRun, args)
            pool.close()
            pool.join()
            # print res.get()
            r = res.next()
            while r:
                # print r
                self.results[0][r[0]] = r[1]
                self.results[1][r[0]] = r[2]
                try:
                    r = res.next()
                except:
                    break

            # print self.results

            # processes = []
            # for i in range(numOfRounds-1, -1, -1):
            #     sim = self.sims[i]
            #     p = multiprocessing.Process(target=sim.singleRun, args=(arrivalRate[i],
            #                                                             effectiveServiceRate, result_queue, i))
            #     processes.append(p)
            #     p.start()
            #
            # for process in processes:
            #     process.join()

            print "DONE"

            # while not result_queue.empty():
            #     result = result_queue.get()
            #     self.results[1][result[0]] = result[1]

            end_time = datetime.datetime.now()
            if self.verbose:
                print "INFO:    Simulation ended at:"
                print "INFO:        time                            :   " + str(end_time)

            # Save results to file.
            if self.verbose:
                print "INFO:    Saving results to file [ " + datetime.datetime.now().strftime(
                    "%Y%m%d-%H%M%S") + "_queue_net_sim.dump ]"
            fd = open(datetime.datetime.now().strftime("%Y%m%d-%H%M%S") + '_queue_net_sim.dump', 'w')
            for val in self.results[0]:
                fd.write(str(val) + ",")
            fd.write("\n")
            for val in self.results[1]:
                fd.write(str(val) + ",")
            fd.write("\n")
            fd.write("\n")
            fd.write("INFO:        time started                    :   " + str(start_time) + "\n")
            fd.write("INFO:        time ended                      :   " + str(end_time) + "\n")
            fd.write("INFO:        number of servers               :   " + str(self.size) + "\n")
            fd.write("INFO:        dispatch policy                 :   " +
                     self.dispatchPolicyStrategy.getName() + "\n")
            fd.write("INFO:        redundancy                      :   " +
                     str(self.dispatchPolicyStrategy.getRedundancy()) + "\n")
            fd.write("INFO:        params                          :   " +
                     self.dispatchPolicyStrategy.getParamStr() + "\n")
            fd.write("INFO:        convergence condition           :   " +
                     self.convergenceConditionStrategy.getName() + "\n")
            fd.write("INFO:        convergence precision           :   " +
                     str(self.convergenceConditionStrategy.getPrecision()) + "\n")
            fd.close()
            print('overall it took {} seconds'.format(time.time() - starttime))

    def plot(self, x=None, y=None, plotStrategy=None):
        self.plotStrategy.plot(self.results[0], self.results[1])


if __name__ == '__main__':
    rounds = 30
    # sim = parSim(3, qns.DispatchPolicyStrategy.RandomDStrategy(alpha=10, beta=1000, p=0.95, d=2, bias=2.0),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim = parSim(4, qns.DispatchPolicyStrategy.FixedSubsetsStrategy(alpha=10, beta=1000, p=0.95, redundancy=2),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim = parSim(4, qns.DispatchPolicyStrategy.FixedSubsetsStrategy(alpha=10, beta=2000, p=0.43, redundancy=1),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim.parRun()
    # # sim.plot()
    # sim = parSim(4, qns.DispatchPolicyStrategy.FixedSubsetsStrategy(alpha=10, beta=2000, p=0.43, redundancy=2),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim.parRun()
    # # sim.plot()
    # sim = parSim(4, qns.DispatchPolicyStrategy.FixedSubsetsStrategy(alpha=10, beta=2000, p=0.43, redundancy=4),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim.parRun()
    # # sim.plot()

    #sim = parSim(3, qns.DispatchPolicyStrategy.RandomDStrategy(alpha=10, beta=2000, p=0.8, d=2, bias=0.225*2),
    #             qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.0001), verbose=True,
    #             numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim1 = parSim(3, qns.DispatchPolicyStrategy.RandomQueueStrategy(alpha=10, beta=2000, p=0.8),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.0001), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim1.parRun()
    # sim = parSim(3, qns.DispatchPolicyStrategy.RouteToIdleQueuesStrategy(alpha=10, beta=2000, p=0.8),
    #              qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.0001), verbose=True,
    #              numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    # sim.parRun()
    # sim1.plot()
    # sim.plot()
    sim = parSim(3, qns.DispatchPolicyStrategy.RoundRobinRedundancyDStrategy(alpha=10, beta=2000, p=0.8, d=2, n=3, bias=0.225*2),
                 qns.ConvergenceConditionStrategy.VarianceConvergenceStrategy(epsilon=0.00005), verbose=True,
                 numOfRounds=rounds, historyWindowSize=20000, T_min=500000, T_max=1000000000, plotStrategy=qns.PLOT())
    sim.parRun()
    sim.plot()