topsim.core.scheduler.Scheduler

class topsim.core.scheduler.Scheduler(env, buffer, cluster, planner, algorithm)

Methods

__init__(env, buffer, cluster, planner, ...)
allocate_ingest(observation, pipelines, planner)	Ingest is 'streaming' data to the buffer during the observation How we calculate how long it takes remains to be seen
allocate_tasks(observation)	For the current observation, we need to allocate tasks to machines based on:
check_ingest_capacity(observation, ...)	Check the cluster and buffer to ensure that we have enough capacity to run the INGEST pipeline for the provided observation
is_idle()	Determine if the scheduler has completed its work
print_state()
run()	Starts the 'per-TIMESTEP' process loop for the Scheduler actor.
scheduler_status()	The status of the scheduled observation(s) and whether or not the scheduler has been delayed yet.
shutdown()
start()	Set the SchedulerStatus to RUNNING.
to_df()	Convert scheduling timestep data into dataframe for the Monitor actor.
to_summary()	Generate summary information when a key event occurs during the current timestep