LigthDNF is a DNF algorithm framework based on reversed index. It is designed for complex conditioned targeting indexing in online advertising, a few advantages will be included as below:
- High efficiency for query, response within limited microseconds(TODO)
- Lower memory usage, as low as possible(TODO)
- High toleration for error or exception(TODO)
- Capable of backup/serialization for data in specific time(TODO)
08/11/2017: basic functions for insert/select within single thread are supported.
To get started, please follow the Installation Guide and Quick Start.
- Wiki
- Installation Guide
- Quick Start (TODO)
- Examples (TODO)
This project is triggered by a paper written by Qian Wang, more details you can refer here.