Skip to content

Optimize dataset loading I/O #75

Closed

Description

@dentiny

Loading a dataset currently performs four sequential operations: an existence check, schema loading, column loading, and row loading.

Combine schema and column discovery into one metadata request, run metadata and row loading concurrently, and let FastAPI execute the synchronous LanceDB calls in its thread pool. This reduces request round trips and avoids blocking the async event loop with synchronous I/O.

Implemented by #73.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions