Apache Cassandra is a 2008 distributed database created by Facebook engineers. A skyrocketing number of users is proof that traditional, relational database engines and existing solutions are unable to provide sufficient solution efficiency. Upon successful launch, the Cassandra development team decided to share the project with a vast user community through Apache Incubator. To date, Apache Cassandra is the most efficient NoSQL “wide-row” class database. At the same time it is fully scalable on any equipment.
Apache Cassandra has been designed so as to meet all non-functional requirements. The main features of the solution include:
Cassandra creators chose horizontal scalability, which is why adding a new nod increases both read and save throughput without the application being unavailable.
Replication and data security
Cassandra cluster replication strategy has been designed so as to make it possible for a system designer to have a configuration that is ideally suitable, taking into account multiple data centres and server placement on racks. Appropriately configured Cassandra replication ensures data safety.
Lack of SPOF (Single Point of Failure)
each cluster element plays an identical role, therefore one nod malfunction does not result in system availability issues.
Data transaction and cohesion
saving and reading may be configured to control record locking levels – from solving, at which point saving is always successful, to blocking all replicas till the end. The most popular solution is to operating at quorum cohesion level (e.g. 50 per cent plus 1), which enables most business cases to operate within parameters. At the same time, the system designer may control the cohesion level even for individual queries, which makes it possible for Cassandra to be used for many standard and non-standard uses.
One of the most significant facilities for application creators was introducing the 0.8 version of the CQL (Cassandra Query Language). With the introduction of an additional abstraction layer between internal data structures and client applications, Cassandra has gained many supporters, both amongst open source software enthusiasts and corporate decision-makers at organizations which had previously only used relational databases.
Clear and complete documentation together with wide community support causes test run deployment launch costs to be relatively low. Simultaneously, the platform’s popularity resulted in there being no problems on the market with finding both specialists in this technology and companies offering comprehensive support of the solution.
Apache Cassandra has a vast selection of drivers for a variety of technologies rolled out either as official versions by the Datastax team, developing the platform’s basic functionalities, or as alternative versions issued by the community. Because of a wide array of available options, Cassandra enables user-friendly integration of all popular technologies and platforms.
Solid design foundations and high stability have caused Apache Cassandra to be the solution of choice for many big players from a variety of business backgrounds. Cassandra has been successfully implemented by many organizations, e.g. CERN, Apple, Netflix, eBay, GitHub, Reddit. At the same time, due to its maturity, the solution has been approved and implemented by more than 1500 financial sector companies, e.g. ING, UBS, Credit Suisse.
See other technologies, which we use in this area
NoSQL / BigData