Define new JDL files to run the jobs you wrote last class:

• define a JDL assuming your jobs are parametric
• define a JDL assuming your jobs are a collection
• submit your new JDL files, wait for the results and compare the makespan of your new experiments with the one performed in the last class
• Use some graphical tool to plot curves of makespan for all experiments according to the number of jobs submitted
• Write your own scheduler such that each one of your jobs will choose to run in the "best" resource (define your choice of "best")
• Compare your new experiments with the previous ones.