ACHTUNG! ThruDB is no longer under development
Thrudb is a set of simple services built on top of the Facebook Apache Thrift framework that provides indexing and document storage services for building and scaling websites. Its purpose is to offer web developers flexible, fast and easy-to-use services that can enhance or replace traditional data storage and access layers.
- Client libraries for most languages
- Incremental backups and redo logging
- Multiple storage backends (BerkeleyDB, Disk, MySQL, S3 included)
- Memcache and Spread integration.
- Built for horizontal scalability
- Simple and powerful search service
- Thrudoc - Document storage service
- Thrudex - Indexing and search service
- Thruqueue - Persistent message queue service (read more)
- Throxy - Scaling service
A good place to start is our Technical Document
The wiki contains installation and configuration info. Please ask questions in our discussion group.
We haven't quite made it to a "release" yet so for now go ahead and grab the code with svn
We've begin an alternate implementation in java on github:http://github.com/tjake/thrudb/tree/master
See Thrudb in action on http://www.junkdepot.com, which is built with Thrudb as its data backend (no RDBMS) and is running on Amazon EC2 with virtually no chance of data loss (thanks to S3 storage backend).