The catalog is a tool that allows you to manage all of your artifacts. This includes a number of things:

  • data logs
  • batch jobs (such as simulations)
  • ingest jobs (which run automatically as logs are offloaded)
  • software releases
  • private keys
  • container images

It provides capabilities for searching artifacts, understanding where your log capacity is going, organizing your artifacts into collections, and running jobs against those artifacts.

For example, you can create ingest jobs that kick off automatically when logs are offloaded. You can also run ad-hoc batch jobs on artifacts using the ark-batch-tool command.

You can visit the TBD catalog at https://catalog.tbdrobotics.com/. The catalog allows for granular permissions at the ‘organization’ level. This allows you to have your artifacts managed by the TBD catalog, but private to your organization.

Conceptually, the catalog is broken down into a few different services:

  • storage of the data itself, which is on an AWS S3 bucket
  • storage of ingested data, which is on a separate S3 bucket
  • state and data that can be queried, stored in a Postgres database
  • the api_gateway, which provides API access to the above
  • the scheduler, which runs and manages batch jobs
  • the frontend, which provides website (frontend) access to the above

The catalog is considered a one-stop-shop for all of your cloud access related to Ark.