Question about dlas implementation

[BingyangWu]

Tiresias/simulator/run_sim.py

Line 1226 in 959f9b0

jump_time = int(math.ceil((JOBS.queue_limit[qid] - rjob['executed_time'])/rjob['num_gpu']) + event_time)

Hi! I am using dlas scheduler in your simulator. However, I am confused about the calculation for jump_time. According to my knowledge, it represents the next time when a job is demoted to another queue. In this case, it should be calculated as jump_time = int(math.ceil((JOBS.queue_limit[qid])/rjob['num_gpu'] - rjob['executed_time']) + event_time) . Could you please explain this calculation? Looking forward to your apply!

[kzhang28]

When using GPU * TIME as the time unit, both queue_limit[qid] and rjob['executed_time] would be using unit GPU*TIME. So I think the original implementation makes sense.