Data Classification
Check your docs version
This version of the Anyscale docs is deprecated. Go to the latest version for up to date information.
| Name | Description | Examples (not exhaustive) | Location |
|---|---|---|---|
| Application logs and customer data | Unstructured application data or semi-structured application data generated by customer code. Customer data refers to the data that customers use. | Ray core, Ray Tune logs, session command jobs, job logs, driver and worker logs | Customer Data Plane |
| Ingested application logs (used to enable log viewer feature) | Structured application logs generated by customer code and by Ray. The log ingestion is opted-in by default for Anyscale Hosted clouds and clouds created through anyscale cloud setup CLI. For clouds created using anyscale cloud register CLI, you must opt-in to the log viewer feature using anyscale cloud config update. | Ray jobs and services logs in log viewer. | Anyscale Control Plane |
| System Logs | Unstructured or semi-structured system data generated by executing Ray or Anyscale components or processes | Autoscaler, GCS, other system logs | Anyscale Control Plane |
| Observability metrics (Grafana or Ray dashboard metrics) | Structured data that contains selected KPI’s of the system | CPU or Disk utilization, Memory | Anyscale Control Plane |
| Metadata | Structured data that describes other data in the system | Start time, stop time, Cloud name, Entry point, YAML file | Anyscale Control Plane |
| Runtime environment metadata | List of dependencies that the application needs to run. Anyscale stores the definitions to understand what to install dynamically. | Filenames, packages, module names, working directory, environment variables | Anyscale Control Plane |
| User actions | Actions performed by users against Anyscale | Login or logout, creating a organization or project, clusters, archiving a job or service, etc. | Anyscale Control Plane |