Overview of Chunked Forest implementation
mainChunked Forest is an implementation of Forest that uses chunks—contiguous parts of the tree stored together in specific data formats. It is designed for a small memory footprint and fast loading of chunked node data.
Key characteristics include:
- Copy-on-write chunks: Optimized to be mutated in place when a chunk has only a single user (detected via reference counting).
- Efficient cloning: Allows for efficient cloning of forests without significant performance overhead in non-cloning scenarios.
- Multiple chunk formats: Uses different formats to optimize for various use cases.