| ScyllaDB | |
|---|---|
 Scylla monster, mascot of the Scylla database | |
| Developer | ScyllaDB Inc. |
| Initial release | September 22, 2015; 10 years ago (2015-09-22) |
| Stable release | |
| Written in | C++ |
| Operating system | Linux |
| Type | distributeddata store |
| License | ScyllaDB Software License Agreement, previouslyGNU AGPL |
| Website | www |
| Repository | |
ScyllaDB is asource-availabledistributedNoSQLwide-columndata store. It was designed to be compatible withApache Cassandra while achieving significantly higher throughputs and lower latencies. It supports the same protocols as Cassandra (CQL) and the same file formats (SSTable), but is a completely rewritten implementation, using theC++20 language replacing Cassandra's Java, and theSeastar[1] asynchronous programming library replacing classic Linux programming techniques such as threads, shared memory and mapped files. In addition to implementing Cassandra's protocols, ScyllaDB also implements theAmazon DynamoDB API.[2]
ScyllaDB uses asharded design on each node, meaning that eachCPU core handles a different subset of data. Cores do not share data, but rather communicate explicitly when they need to. The ScyllaDB authors claim that this design allows ScyllaDB to achieve much better performance on modernNUMASMP machines, and to scale very well with the number of cores. They have measured as much as 2 million requests per second on a single machine,[3] and also claim that a ScyllaDB cluster can serve as many requests as a Cassandra cluster 10 times its size – and do so with lower latencies.[4] Independent testing has not always been able to confirm such 10-fold throughput improvements, and sometimes measured smaller speedups, such as 2x.[5] A 2017 benchmark fromSamsung observed the 10x speedup on high-end machines – the Samsung benchmark reported that ScyllaDB outperformed Cassandra on a cluster of 24-core machines by a margin of 10–37x depending on theYCSB workload.[6]
ScyllaDB is available on-premises, on major public cloud providers, or as a DBaaS (ScyllaDB Cloud).[7]
ScyllaDB was started in December 2014 by thestartupCloudius Systems (later renamed ScyllaDB Inc.), previously known for having createdOSv. Co-founders wereAvi Kivity and Dor Laor. ScyllaDB was released as open source in September 2015,[8] under theAGPL license. In December 2024, ScyllaDB moved to asource available license.[9] Its source code and development process is still open to the public on publicGitHub repositories, but usage on a cluster beyond a certain size requires purchasing a license.
