JÄMFÖRELSE MELLAN ORACLE RDBMS, ORACLE NOSQL OCH MONGODB

Databases are present everywhere in our modern society and the amount of data that have to be stored is constantly increasing, which means that it’s now more important than ever to be able to handle massive data sets e\u82ffectively. NoSQL databases2 were developed to solve this problem by e\u81fficiently storing large amounts of data and enable fast access to that data. Since NoSQL databases only became popular within the last ten years, they haven’t been as well researched as relational databases. An in-depth evaluation is carried out on six distinct features, where one part is comparative performance tests. Th\u91e other features are: scalability, consistency, availability, durability and reliability. MongoDB and Oracle NoSQL are the NoSQL databases used and together with Oracle RDBMS as relational database make up the basis for a comparative study of the above mentioned features.Th\u91e results showed that there are big diff\u82erences between how data is handled in NoSQL compared to relational databases that will aff\u82ect the choice of database, e.g. that NoSQL tends to prioritize that clients can reach the database over non-contradictory data and lowering the demands on transaction management to increase performance and storage capacity. Furthermore, the performance tests showed that both NoSQL databases performed be\u8ber than the relational database regardless of the data set size. MongoDB was clearly the fastest on reading operations, while Oracle NoSQL performed write operations the fastest most of the time. Both NoSQL databases are impacted less by a growing data set than the relational database for both read and write operations