# filesystem-catalog-original

**Repository Path**: magic-flute_1/filesystem-catalog-original

## Basic Information

- **Project Name**: filesystem-catalog-original
- **Description**: A prototype for implementing datalake catalog management based on arbitrary file systems.

一个用于研究基于纯文件系统进行可靠的数据湖catalog管理的原型项目.
- **Primary Language**: Java
- **License**: BSD-3-Clause
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-06-10
- **Last Updated**: 2025-06-22

## Categories & Tags

**Categories**: Uncategorized

**Tags**: DataLake, catalog, hadoop, s3, oss

## README

# filesystem-catalog-original
A prototype for implementing catalog management based on arbitrary file systems.

We want to minimize the complexity of the infrastructure required to build a datalake.

Our goals are:
- The submission strategy should be pluggable.
- Try to use the same set of submission logic to complete the submission actions in different file systems.

Things yet to be done:
- Multi-table transactions.
- Serializable submission.

see demo.fscatalog.process.FileTrackerCommitStrategyV2

If you have any ideas, feel free to communicate with me. plashspeed@foxmail.com.