Environments & Operations

OptScale AI Environments & Operations covers MODEL TRAINING pages for shared runtime environments, cloud account links, automated instance power schedules, and external calendar or CI/CD integrations:

• Shared Environments for runtime reservation and environment management
• Cloud Connections for cloud account integration
• Power Schedules for automated infrastructure start and stop operations
• Integrations for calendar and CI/CD connectivity

For experiment tracking, datasets, models, and artifacts, see Experiment Tracking. For the complete sidebar structure, see Model Training overview.

Administrative actions in this section require the Organization manager role.

Shared Environments

The Shared Environments page provides controls for filtering environments, managing reservations, and monitoring availability. Use Table and Calendar views for reservation planning and operational visibility.

Use Table or Calendar to switch views, Only accessible by me and Any status to filter environments, + ADD to create a shared environment, and Search to filter by name or visible fields.

The main table displays the environment name, availability status, upcoming bookings, installed software, and available management actions.

Row actions

Table 1: Shared Environments row actions

Button	Action
	Book the environment for a selected time range.
	Release a reservation and make the environment available again.
	Temporarily disable bookings for the environment.
	Remove the environment from the organization catalog.

Create a shared environment

To register a shared environment:

1. Open MODEL TRAINING → Shared Environments.
2. Click + ADD.
3. Enter a Name and select a Resource type.
4. Configure optional properties such as SSH access, Description, IP, and Software.
5. Click CREATE.

New environments appear in the table together with status, booking, and software information.

Cloud Connections

The Cloud Connections page provides controls for registering and managing cloud provider integrations.

Use + ADD to start the connection wizard for a supported provider. Use Search to filter connections by name or provider type.

The main table displays the connection name and associated cloud provider type. Some connections may include expandable resource hierarchies for nested cloud scopes or accounts.

The footer displays the total number of configured cloud connections.

Power Schedules

Use Power Schedules to automate cloud instance start and stop operations across AWS, Azure, Google Cloud, and Alibaba Cloud resources. Each cloud resource can belong to only one schedule. A schedule may contain multiple daily power-on and power-off triggers.

The Power Schedules page provides controls for creating, filtering, and managing automated infrastructure schedules.

Use + ADD to create a power schedule. Use Search to filter schedules by name or visible fields.

The main table displays schedule information, including the schedule name, latest execution time, assigned cloud resources, configured triggers, time zone, validity period, and available management actions.

The Actions column provides controls for activating, deactivating, and deleting power schedules.

Use to activate a schedule.

Use to deactivate a schedule.

Use to delete a schedule.

The footer displays Total and Displayed counters for the current filter results.

Cloud permissions

Cloud resources require start and stop permissions before schedules can operate.

Table 2: Required cloud permissions

Provider	Permissions
AWS	ec2:StartInstances, ec2:StopInstances
Azure	Microsoft.Compute/virtualMachines/start/action, Microsoft.Compute/virtualMachines/deallocate/action
Alibaba Cloud	ecs:StartInstance, ecs:StopInstance
Google Cloud	compute.instances.start, compute.instances.stop

Schedule detail page

Select a schedule Name to open the detail view.

Summary cards display:

• last execution time
• assigned resource count
• configured time zone
• validity period

A visual 24-hour timeline displays powered-on and powered-off intervals throughout the day.

Use the Instances tab to manage assigned resources and the Triggers tab to review configured schedules.

Create a power schedule

To create a schedule:

1. Open MODEL TRAINING → Power Schedules.
2. Click + ADD.
3. Enter a schedule name and select a time zone.
4. Configure validity dates and triggers.
5. Click CREATE.

Manage schedule resources

To assign resources to a schedule:

1. Open the schedule detail page.
2. Go to the Instances tab.
3. Select ADD INSTANCES TO SCHEDULE.
4. Choose instances and click ADD.

To remove resources:

1. Select instances in the Instances tab.
2. Click REMOVE INSTANCES FROM SCHEDULE.

Edit or delete schedules

Use the schedule detail page to:

• update triggers
• modify validity dates
• activate or deactivate schedules
• delete schedules

Changes are applied after selecting EDIT or confirming the delete operation.

Integrations

Use Integrations to connect external scheduling and CI/CD systems to Shared Environment workflows.

Supported integrations may include:

• Google Calendar
• GitLab
• GitHub
• Jenkins

The page uses a card layout where each integration contains setup guidance and operational details.

Google Calendar

The Google Calendar integration synchronizes Shared Environment reservations with a shared calendar view.

Table 3: Google Calendar integration

Element	Purpose
Description	Explains how booking intervals appear in the calendar.
Status	Displays whether a calendar is connected.
CONNECT CALENDAR	Starts the calendar connection flow.

CI/CD integrations

GitLab, GitHub, and Jenkins integrations allow CI/CD pipelines to report environment status and deployment metadata to OptScale AI.

Table 4: CI/CD integrations

Integration	Typical use
GitLab	Deployment pipeline stage.
GitHub	GitHub Actions workflow step.
Jenkins	Post-build or deployment stage.

Pipeline examples use $ENV_COLLECTOR_URL to send deployment status and environment metadata updates.

Related documentation

Table 5: Related documentation

Topic	Where to read more
Tasks, models, datasets, artifacts	Experiment Tracking
Model Training navigation overview	Model Training
AI Gateway, policies, and FinOps	Core Services
Platform architecture	Architecture Overview