Documentation

# kortical.api.worker_group

WARNING

Creating worker groups will incur further compute costs.

This object represents a worker group on Kortical. This is a set of computers of a given specification (e.g number of cores, RAM, GPU). You can configure various tasks, projects and environments to run on a specific worker group. For example, you might want to dedicate one worker group for training models and another for a Production environment, so performance by one is not affected by the other.

The WorkerGroup class includes the following attributes:

  • id - The ID assigned to the worker group upon creation.
  • name - The name of the worker group.
  • state - current status (e.g creating, resizing, running, terminating).
  • worker_type - the specification of the computers that belong to the worker group.
  • current_size/required_size - the actual/desired number of computers in the worker group.

In the Kortical CLI, you can view this by running:

>>> kortical worker list

Worker Groups:
+----+------------+---------+----------------+--------------+---------------+
| id |    name    |  state  |  worker_type   | current_size | required_size |
+====+============+=========+================+==============+===============+
| 1  | system     | Running | 4-CPU-26GB-RAM | 3            | 3             |
+----+------------+---------+----------------+--------------+---------------+
| 2  | default    | Running | 4-CPU-26GB-RAM | 2            | 2             |
+----+------------+---------+----------------+--------------+---------------+
| 3  | Production | Running | 2-CPU-13GB-RAM | 2            | 2             |
+----+------------+---------+----------------+--------------+---------------+

#

# list

This function returns a full list of worker groups used in Kortical.

No Inputs

Returns

  • [WorkerGroup, ...] - A list of WorkerGroup objects.
from kortical.api.worker_group import WorkerGroup

worker_groups = WorkerGroup.list()

#

# get_worker_group

This function returns the specified worker group from Kortical.

Inputs

  • worker_group_or_id - Name or ID of the worker group you want.

Returns

  • WorkerGroup - A WorkerGroup object.
from kortical.api.worker_group import WorkerGroup

default_worker_group = WorkerGroup.get_worker_group('default')

#

# list_worker_types

This function returns a list of worker group types.

No Inputs

Returns

  • [str, ...] - A list of worker group types.
from kortical.api.worker_group import WorkerGroup

worker_types = WorkerGroup.list_worker_types()

#

# get_default_worker_type

This function gets the default worker type.

No Inputs

Returns

  • str - The default type.
from kortical.api.worker_group import WorkerGroup

default_worker_type = WorkerGroup.get_default_worker_type()

#

# create_worker_group

This function creates a worker group in Kortical.

Inputs

  • worker_group_name - Name of the new worker group.
  • worker_type - Type of the new worker group; if unsure, run worker_group.list_worker_types() to see available options.
  • size - Number of computers in the new worker group.

Returns

  • WorkerGroup - A WorkerGroup object.
from kortical.api.worker_group import WorkerGroup

new_worker_group = WorkerGroup.create_worker_group('production_environment', TYPE_HERE, 2)

#

# resize

This function resizes a worker group in Kortical.

Inputs

  • size - Number of computers in the new worker group.

Returns

  • WorkerGroup - A WorkerGroup object.
from kortical.api.worker_group import WorkerGroup

# Get the worker group
uat_environment_worker_group = WorkerGroup.get_worker_group('UAT')

# Update
uat_environment_worker_group.rersize(size=3)

#

# delete

This function deletes a worker group from Kortical.

No Inputs

Returns

  • None
from kortical.api.worker_group import WorkerGroup

# Get the worker group
uat_environment_worker_group = WorkerGroup.get_worker_group('UAT')

# Delete
uat_environment_worker_group.delete()

← kortical.api.experiment_framework →