The profits of HACASRPS and HACAS algorithms are shown in Figure 5.
From Figure 5, we can see that as cycle increases, the profits of both HACASRPS and HACAS algorithms increase.
Balancing the load of all the providers to make the system last longer is an important target of HACAS algorithm's provider selection scheme.
With the parameters in Table 1, theaverageloadsof HACAS and HACASRPS algorithms are 0.800 and 0.779, respectively.
Figure 6 tells us that the average loads of HACAS and HACASRPS algorithms are 0.8 and 0.78, respectively.
Moreover, the load variation of HACAS algorithm is much lower than that of HACASRPS algorithm.
From Figure 8, we can see that the load of the providers in HACASRPS algorithm fluctuates in a much wider range than that of HACAS algorithm.
Figure 9 reveals that the standard deviations of all the providers' loads of the ant's solution of HACAS algorithm are much lower than those of the HACASRPS algorithm.
Similar to Figure 7, Figure 10 further reveals the load variations of HACAS and HACASRPS algorithms in different simulation cycles when n = 30, from which we can derive similar observations with those drawn from Figure 7.
As shown in Section 5.2, the performance of HACAS algorithm is better than that of FCFS and HACASRPS algorithms.
In this paper, the parameters used in Table 1 can validate that HACAS algorithm is effective under heavy load environment (i.e., the applications' total resource requirements exceed the resources possessed by the system).