A study and comparison of NoSQL databases

The advent of big data and the need for large scale data analysis has led to a rapid increase in the number of systems designed to serve data on the cloud. These systems address the need for OLTP, often adhering to BASE principles in order to maintain a high degree of availability and scalability. The wide range of systems, now available, make it difficult to determine which system will best meet a user's needs without testing each system individually. In 2010, Yahoo! filled the need for an apples-to-apples performance comparison by open sourcing their Yahoo! Cloud Serving Benchmark (YCSB) framework. The framework was designed with the goal of facilitating performance comparisons of the new generation of cloud data serving systems. YCSB provides a set of core workloads that reflect common real world use cases. This thesis discusses the architectures of several widely used NoSQL database systems as well as the results of applying YCSB workloads to the most readily available versions of those systems.