As an alternative to MongoDB, Cassandra is highly available and uses a column-orientated approach. It is a database that was developed by Facebook, particularly with the goal of enabling inbox searches. It is utilised by both Facebook and Twitter, and it is also very popular wherever a degree of logging is required, including in financial services.
The Cassandra database makes use of a replication model that is in a Dynamo-style and as a result does not have a single point of failure. It uses a SQL-like language for queries called Cassandra Query Language or CQL. It can be used alongside Linux, Unix and Windows and many different languages.
Top features of Cassandra
- Cassandra scales linearly and offers very quick response times
- It supports the ACID principle
- Cassandra supports Hadoop and MapReduce
- It offers solid flexibility when it comes to distributing data
- Cassandra is very scalable and uses peer to peer architectural principles
Pros and Cons of Cassandra Database
Developers like Cassandra because it is very scalable and because there is no single point of failure in the database. Its multi-DC replication capabilities is also attractive, while developers also like the Cassandra integrates with other applications based on JVM.
However, Cassandra has limited support for aggregations and developers find that it does not offer predictable performs. It also lacks support for ad-hoc queries.