Overview
An overview of the concept and purpose of LOC data processes.
What is a Data Process?
A data process in LOC is a workflow or a data pipeline, which includes a series of logic to be executed.
A data process has its own revision system. If one of the included logic is modified, you'll need to edit the data process as well, which will create a new revision.
Every user asset created in LOC - for example, a data process - will have an unique identity string called PID (permanent ID).
Workspace, Projects and Scenarios
A data process belongs to a scenario under a project; both can be created by users.
Projects belong to a workspace or unit. Currently, the default workspace or unit is not editable.
Agent Configurations and Environment Variables
See: Agent Configuration
A logic, once added into a data process, can have its own environment variables and references to agent configurations (for connecting whitelisted external data sources).
Data Process Execution
Invoke an Execution
See: Trigger
A trigger can invoke an execution with one or more data processes as tasks.
Execution Result and Error Handling
See: Execution and Task
The execution and task results - including errors - will be stored as execution results.