Cassandra is an open source distributed database management system. It is simple key value store, designed to handle large data stored across multiple commodity hardware. Data is automatically replicated to multiple nodes for high availability. Failed nodes can be replaced with no down times. Every node in the cluster is identical, so there is no single point of failure problem.
Cassandra is suitable for applications that can’t afford data lose even if entire system goes down. Its data model is designed for distributed data on a very large scale. It trades ACID compliant data practices for important advantages in performance, availability and operational manageability. Read and write throughput both increase linearly as new machines are added, with no interruption to application.