| Abstract |
Database Management Systems (DBMSs) have long been developed with the focus on
transactional applications. Such applications are usually characterized by a large number of
updates and very little data retrieval in comparison. However, in the past decade demand has
shifted from these transactional applications towards more analytical applications. After many
years of storing data, analyzing that data has become very important in many sectors. This
analysis is very query-intensive, yet many popular relational DBMS s still have storage systems
optimized for transactional use, as well as restrictive query languages. Monet is a non-traditional
database system that provides high performance for query-intensive applications. Monet can be
accessed through a SQL layer or directly through a low level algebraic interface. The application
developed in this thesis project will implement a clean relational algebra layer on top of Monet,
while attempting to maintain its high performance. |
