Roadmap¶

The project introduction explains the overview and goals of DataFusion, and our development efforts largely align to that vision.

Planning `EPIC`s¶

DataFusion uses GitHub issues to track planned work. We collect related tickets using tracking issues labeled with [EPIC] which contain discussion and links to more detailed items.

Epics offer a high level roadmap of what the DataFusion community is thinking about. The epics are not meant to restrict possibilities, but rather help the community see where development is headed, align our work, and inspire additional contributions.

As this project is entirely driven by volunteers, we welcome contributions for items not currently covered by epics. However, before submitting a large PR, we strongly suggest and request you start a conversation using a github issue or the dev@arrow.apache.org mailing list to make review efficient and avoid surprises.

The current list of EPICs can be found here.

Quarterly Roadmap¶

A quarterly roadmap will be published to give the DataFusion community visibility into the priorities of the projects contributors. This roadmap is not binding and we would welcome any/all contributions to help keep this list up to date.

2023 Q4¶

Improve data output (COPY, INSERT and DataFrame) output capability #6569
Implementation of ARRAY types and related functions #6980
Write an industrial paper about DataFusion for SIGMOD #6782

2022 Q2¶

DataFusion Core¶

IO Improvements
- Reading, registering, and writing more file formats from both DataFrame API and SQL
- Additional options for IO including partitioning and metadata support
Work Scheduling
- Improve predictability, observability and performance of IO and CPU-bound work
- Develop a more explicit story for managing parallelism during plan execution
Memory Management
- Add more operators for memory limited execution
Performance
- Incorporate row-format into operators such as aggregate
- Add row-format benchmarks
- Explore JIT-compiling complex expressions
- Explore LLVM for JIT, with inline Rust functions as the primary goal
- Improve performance of Sort and Merge using Row Format / JIT expressions
Documentation
- General improvements to DataFusion website
- Publish design documents
Streaming
- Create StreamProvider trait

Ballista¶

Make production ready
- Shuffle file cleanup
- Fill functional gaps between DataFusion and Ballista
- Improve task scheduling and data exchange efficiency
- Better error handling
  - Task failure
  - Executor lost
  - Schedule restart
- Improve monitoring and logging
- Auto scaling support
Support for multi-scheduler deployments. Initially for resiliency and fault tolerance but ultimately to support sharding for scalability and more efficient caching.
Executor deployment grouping based on resource allocation