Background
To have a better idea of simulation life-cycle and efficiency; after collecting all the metadata into a centralized locations, we should make a variation of plots.
- accumulative core hour overtime. (x-axis as date. y-axis as the core hours used over time)
- Jobs running per day. ( x-axis as date. y-axis as jobs finished/running per day)
2a. different plots for each step (e.g. emod3d, post-emod, hf, bb, imcalc) - Comparison plot of estimated and actual runtime of each step. ( accumulative plot, where x-axis is date.)
Tasks
- decide/implement what library to read the json file with (currently preferring PANDAS) - 2H
- script to plot total jobs running per day over a period of time
2a. the script should take 'step' name as argument. (example. 'all' , 'emod3d' or 'bb') - 3H - script to plot total core hour usage over a period of time
3a. the script should take 'step' name as argument. (example. 'all' , 'emod3d' or 'bb') - 3H
3b. the script should have an argument that will plot the estimated core hours as a comparison -4h