Current techniques used to design and implement database systems do not provide support for efficient implementation of crosscutting concerns in the database software, e.g., failure detection, database policies such as concurrency control and scheduling, and synchronization. Aspect-oriented software development (AOSD) is a new technique that provides an efficient way of modularizing crosscutting concerns in software systems. In This work we evaluate the effectiveness of applying AOSD to database systems thereby paving way for successful application of aspect languages to the database domain. Our focus is on embedded database systems, as a representative for a class of database systems. We show, by analyzing and re-engineering one commercial well-known embedded database system (Berkeley database), that aspect-orientation has promise, especially in enabling development of tailorable, maintainable, and evolvable database systems.