Apache® Cassandra™ is a popular NoSQL database that does certain things incredibly well. It can be always available, with multi-datacenter replication. It is also scalable and lets users keep their data anywhere. However, Cassandra is lacking in speed. Cassandra is not fast enough for some of today’s extreme OLTP workloads because it stores data on disk. Cassandra also shares certain limitations of other NoSQL databases such as limited querying capabilities.
Fortunately, there is a simple way to make Cassandra much faster and more flexible. You can add in-memory computing to Cassandra by combining it with Apache® Ignite™ or its production-ready version, the GridGain in-memory computing platform. Apache Ignite also provides ANSI SQL-99 compliance for in-memory SQL and ACID transaction guarantees.
This white paper discusses the lack of in-memory capabilities in Cassandra, why in-memory solutions make sense for today’s web and cloud applications, and how combining Cassandra with Apache Ignite creates an architecture that provides a substantial speed boost and a host of other benefits.
Cassandra Plus Apache Ignite: A Potent Combination
If you are using Cassandra – or considering using it – but are concerned that it will not meet the speed demands of extreme OLTP workloads, it may be time to consider in-memory technology. Apache Ignite and its production-ready version, the GridGain in-memory computing platform, can combine with Cassandra to provide a solution that accesses data in memory instead of on disk – an approach that is 1,000 times faster than disk-based approaches.
Adding Apache Ignite to Cassandra maintains Cassandra’s high availability and horizontal scalability while also providing several additional benefits. These benefits include more flexible query capabilities, horizontal and vertical scalability, and more robust consistency (ANSI SQL-99 compliance and ACID transaction guarantees). Plus, the migration path requires no data remodeling.
With these benefits and the substantial speed boost of in-memory computing, the combination of Cassandra and Apache Ignite is a solution well suited to heavy workloads and the demands of today’s web and cloud applications.
The key modules of the GridGain in-memory computing platform are that relevant to running extreme OLTP workloads with Apache Cassandra are:
- Data grid – Essentially an in-memory key value store that can be queried
- SQL grid - provides the ability to interact with data in-memory using ANSI SQL-99 via JDBC or ODBC APIs
- Compute grid - A stateless grid that provides high-performance computation in memory using clusters of computers and parallel processing
- Service grid - A service grid in which grid service instances are deployed across the distributed data and compute grids
- Streaming – The ability to consume an endless stream of information and process it in real-time
- Advanced clustering – The ability to automatically discover nodes, eliminating the need to restart the entire cluster when adding new nodes
Super Power Apache® Cassandra™ for Extreme OLTP Workloads with GridGain - White Paper
If you organization is developing or trying to improve an implementation of Cassandra for processing OLTP workloads, please download Super Power Apache® Cassandra™ for Extreme OLTP Workloads with GridGain, a new GridGain Systems white paper that takes a detailed look at OLTP platform requirements and how in-memory computing can increase the performance and scale beyond what Cassandra offers.