Workspace APIs

This topic describes the Python SDK APIs for the workspace operations.

Create a Workspace

The create workspace API provisions a workspace that spans multiple clusters and namespaces already onboarded with the EGS controller. This workspace acts as an isolated environment where end users can request GPU resources and deploy workloads.

Syntax

egs.create_workspace(workspace_name, clusters, namespaces, username, email, authenticated_session=None)

Parameters

Input Parameter	Type	Description	Required
workspace_name	String	The name of the workspace to be created.	Mandatory
clusters	List of strings	The set of clusters that should be part of the workspace. The cluster must be registered with the EGS controller before it can be onboarded onto a workspace.	Mandatory
namespaces	List of strings	The set of namespaces that should be part of the workspace. These namespaces will be created on the workspace cluster if they don’t exist.	Mandatory
username	String	The name of the user requesting the workspace.	Mandatory
email	String	The email of the user requesting the workspace.	Mandatory
authenticated_session	[AuthenticatedSession]	The authenticated session with the EGS controller. The default value is None. If no authenticated session has been provided, SDK will try to use the SDK default. If no SDK default has been found, an exception will be raised.	Optional

Response Object Returned

Returns	Description
String	The name of workspace.

Exceptions Raised

Raises	Description
exceptions.Unauthorized	This exception is raised when the workspace creation is not allowed to the API Key.
exceptions.WorkspaceAlreadyExists	This exception is raised when a workspace with the given name already exists.

Example

import egs
from egs.exceptions import ApiKeyInvalid, ApiKeyNotFound, WorkspaceAlreadyExists
    try:
        # Authenticate the EGS
        auth = egs.authenticate("https://egs-core-apis.example.com", "5067bd55-1aef-4c84-8987-3e966e917f07")
        workspace_name = egs.create_workspace("llm-gpt-4o", ["worker-1", "worker-2"], ["llm-gpt-40-dev"], "John Doe", "john.doe@avesha.io", auth)

    except ApiKeyInvalid as e:
        # Do something when API key is not found
        print("API Key is Invalid")
        print(e)
    except ApiKeyNotFound as e:
        # Do something when API key is not found
        print("API Key not found")
        print(e)
    except WorkspaceAlreadyExists as e:
        # Do something when API key is not found
        print("WorkspaceAlreadyExists found")
        print(e)
    except Exception as e:
        print(e)
        print(f"Exception: {e}")

Delete a Workspace

The delete workspace API deletes the provisioned workspace.

Syntax

egs.delete_workspace(workspace_name, authenticated_session=None) 

Parameters

Input	Parameter Type	Description	Required
workspace_name	String	The name of the workspace to be deleted.	Mandatory
authenticated_session	[AuthenticatedSession]	The authenticated session with the EGS controller. The default value is None, If no authenticated session has been provided, SDK will try to use the SDK default. If no SDK default has been found, an exception will be raised.	Optional

Response Object Returned

Returns	Description
Boolean	The value is true, if the workspace deletion is successful. The value is false, if the workspace deletion is unsuccessful.

Exceptions Raised

Raises	Description
exceptions.Unauthorized	This exception is raised when the workspace deletion is not allowed to the API Key.

Example

import egs
from egs.exceptions import ApiKeyInvalid, ApiKeyNotFound, WorkspaceAlreadyExists
    try:
        # Authenticate the EGS
        auth = egs.authenticate("https://egs-core-apis.example.com", "5067bd55-1aef-4c84-8987-3e966e917f07")
        workspace_name = egs.create_workspace("llm-gpt-4o", ["worker-1", "worker-2"], ["llm-gpt-40-dev"], "John Doe", "john.doe@avesha.io", auth)
        delete_status = egs.delete_workspace("llm-gpt-4o", authenticated_session=auth) 
        print(delete_status)
    except ApiKeyInvalid as e:
        # Do something when API key is not found
        print("API Key is Invalid")
        print(e)
    except ApiKeyNotFound as e:
        # Do something when API key is not found
        print("API Key not found")
        print(e)
    except WorkspaceAlreadyExists as e:
        # Do something when API key is not found
        print("WorkspaceAlreadyExists found")
        print(e)
    except Exception as e:
        print(e)
        print(f"Exception: {e}")  

List Workspaces

The list workspaces API returns the list of workspaces that the EGS controller controls.

Syntax

egs.list_workspaces(authenticated_session=None) 

Parameters

Input	Parameter Type	Description	Required
authenticated_session	[AuthenticatedSession]	The authenticated session with EGS controller. The default value is None. If no authenticated session has been provided, SDK will use the SDK default. If no SDK default value is found, an exception will be raised.	Optional

Response Object Returned

Returns	Description
List[class Workspace]	The list of EGS workspaces.

Exceptions Raised

Raises	Description
exceptions.Unauthorized	This exception is raised when the operation is not allowed to the API Key.

Type: class Workspace

Parameter	Type	Description	Required
name	String	The name of the workspace.	Mandatory
clusters	List of strings	The list of clusters that are attached to the workspace.	Mandatory
namespaces	List of strings	The list of namespaces that are attached to the workspace.	Mandatory

Example

import egs 
auth = egs.authenticate("https://egs-core-apis.example.com", "5067bd55-1aef-4c84-8987-3e966e917f07") 
workspaces = egs.list_workspaces(auth) 

Get Workspace Kubeconfig

The workspace kubeconfig API returns the Kubernetes kubeconfig YAML in the text format whicch is used directly with Kubernetes SDK and/or with the kubectl CLI for the deployment of workloads on the cluster.

Syntax

egs.get_workspace_kubeconfig(workspace_name, cluster_name, authenticated_session=None)

Parameters

Input	Parameter Type	Description	Required
workspace_name	String	The name of the workspace for which the kubeconfig should be generated.	Mandatory
cluster_name	String	The name of the cluster.	Mandatory
authenticated_session	[AuthenticatedSession]	The authenticated session with the EGS controller. The default value is None. If no authenticated session has been provided, SDK will try to use the SDK default. If no SDK default is found, an exception will be raised.	Optional

Response Object Returned

Returns	Description
String	The kubeconfig YAML file in the text format.

Exceptions Raised

Raises	Description
exceptions.Unauthorized	This exception is raised when requesting the kubeconfig YAML is not allowed to the API Key.

Example

import egs
from egs.exceptions import ApiKeyInvalid, ApiKeyNotFound, WorkspaceAlreadyExists
    try:
        # Authenticate the EGS
        auth = egs.authenticate("https://egs-core-apis.example.com", "5067bd55-1aef-4c84-8987-3e966e917f07")
        workspace_name = egs.create_workspace("llm-gpt-4o", ["worker-1", "worker-2"], ["llm-gpt-40-dev"], "John Doe", "john.doe@avesha.io", auth)

        kube_config = egs.get_workspace_kubeconfig("llm-gpt-4o",
                                                    "worker-1",
                                                    authenticated_session=auth)
        print(kube_config)
    except ApiKeyInvalid as e:
        # Do something when API key is not found
        print("API Key is Invalid")
        print(e)
    except ApiKeyNotFound as e:
        # Do something when API key is not found
        print("API Key not found")
        print(e)
    except WorkspaceAlreadyExists as e:
        # Do something when API key is not found
        print("WorkspaceAlreadyExists found")
        print(e)
    except Exception as e:
        print(e)
        print(f"Exception: {e}")