Link	Redis	Bangdb	BerkleyDB	LevelDB
Best used	For rapidly changing data with a foreseeable database size (should fit mostly in memory)	For random reads and writes, rapidly changing data, data not needed to fit in memory as it works with overflow decently	For rapidly changing data, key value store, object store, static data, data should fit mostly in memory. Works with larger memory but write degrades badly	For rapidly changing data, key value store, object store, static data. Works with overflow of data out of memory but read degrades
Usage example	Stock prices. Analytics. Real-time data collection. Real-time communication.	key value store, real time data, static data, session data, object store	key value store, real time data, static data, session data, object store	key value store, real time data, static data, session data, object store
Main focus	Speed	Speed, Robustness, Crash Recovery, Scale from embedded to network to whole data grid. Available in many flavors. Elastic	Speed, Robustness, Data recovery, Available as Embedded db	Speed, Robustness, Available as Embedded db, very high sequential write
License	BSD	BSD	Oracle	BSD
Projects using it	craigslist.org github.com guardian.co.uk Disqus stackoverflow.com flickr.net tweetdeck.com blizzard.com
Technical details
Latest version	6.0.4	0.5	18.1	1.22
Release date	May 28th 2020	May 25th 2012	Jun. 14th 2018	May 3rd 2019
Initial release	2009	Apr. 25th 2012	1994	Jul. 30th 2011
Transaction support	Yes	Yes	Yes	No
Replication	Master/slave	P2P	Master/slave
Protocol	Telnet-like	Custom API, HTTP	Custom API	Custom API
Development language	C/C++	C/C++	C	C++
Platforms	Cross-platform	Cross-Platform	Cross Platform	Cross Platform
Additional
Website	redis.io	bangdb.com	oracle.com/...	google.com/...
Wikipedia	wikipedia.org/...	wikipedia.org/...	wikipedia.org/...	wikipedia.org/...
Flavor	Network	Embedded, Network, Elastic Cache	Embedded	Embedded
Access Model	Client Server	Embedded as part of process (Released) Client Server Distributed data grid - P2P	Embedded	Embedded