Class GraphLayout

Declaration

Documentation

This assigns (x, y) locations to all tasks quickly and dynamically as new
tasks are added.  This scales to a few thousand nodes.

It is commonly used with distributed/bokeh/scheduler.py::TaskGraph, which
is rendered at /graph on the diagnostic dashboard.

Methods

• ▷ def __init__(self, scheduler)

  source link
• ▶ def reset_index(self)

  Reset the index and refill new and new_edges

  source link

  From time to time TaskGraph wants to remove invisible nodes and reset
  all of its indices.  This helps.

• ▶ def transition(self, key, start, finish, *args, **kwargs) override

  inherited doc Run whenever a task changes state

  source link

  This method overrides distributed.diagnostics.plugin.SchedulerPlugin.transition.

  Parameters

  • key : string
  • start : string
    

    Start state of the transition.
    One of released, waiting, processing, memory, error.

  • finish : string
    

    Final state of the transition.

  • *args, **kwargs : More options passed when transitioning
    

    This may include worker ID, compute time, etc.

• ▶ def update_graph(self, scheduler, dependencies=None, priority=None, tasks=None, **kwargs) override

  inherited doc Run when a new graph / tasks enter the scheduler

  source link

  This method overrides distributed.diagnostics.plugin.SchedulerPlugin.update_graph.

Inherited methods

• Methods inherited from distributed.diagnostics.plugin.SchedulerPlugin:
  add_client, add_worker, remove_client, remove_worker, restart

Reexports

• Imported in distributed.diagnostics.tests.test_graph_layout.
• Imported in distributed.diagnostics.
• Imported in distributed.dashboard.components.scheduler.

Generated by nedoc v0.9 at 2020-12-29 14:05