Multi-criteria genetic algorithm applied to scheduling in multi-cluster environments

Figures & data

Figure 1 Chromosome design in Example 1.

Figure 2 Job schedule in Example 1.

Table 1 Number of generations for the GA to converge with the two initial population creation strategies

Workload	GA-Random	GA-Guided
1	600	390
2	220	50
3	300	10
4	190	15
5	300	210
6	50	10
7	350	50
8	1000	810
9	700	100
10	500	50

Figure 3 Crossover in Example 2.

Figure 4 GA profiling: Population size and number of iterations.

Figure 5 GA profiling: Mutation frequency.

Figure 6 α value evaluation: Makespan and Flowtime.

Table 2 Settings of GA-MF key parameters

Parameter	Value
Num. Iterations	120
Population Size	80
Mutation Frequency	50
α parameter	0.6

Figure 7 Makespan results for 30 different workloads composed of 20 jobs.

Figure 8 Flowtime results for 30 different workloads composed of 20 jobs.

Figure 9 Makespan results for 25 different workloads composed of 2000 jobs.

Figure 10 Flowtime results for 25 different workloads composed of 2000 jobs.