ASM Foundations of Database Management Bernhard Thalheim 1 and Klaus-Dieter Schewe 2 1 Christian Albrechts University Kiel, Department of Computer Science Olshausenstr. 40, D-24098 Kiel, Germany 2 Massey University, Information Science Research Centre Private Bag 11 222, Palmerston North, New Zealand thalheim@is.informatik.uni-kiel.de k.d.schewe@massey.ac.nz Abstract. Database structuring is well understood since decades. The operating of databases has been based in the past on temporal logics and did not yet get an easy to understand formal underpinning. There- fore, conceptions like transaction and recovery are mainly discussed at the logical or operational level. This paper shows that database struc- turing and functionality can be defined within a uniform language. We base database semantical on the operational semantics of abstract state machines (ASM). This uniform mechanism allows to define the structur- ing, the functionality, the distribution and the interactivity of a database system in a way that supports abstract consideration at various layers of abstraction, that supports refinement of specifications to more detailed ones and that support proof of properties. 1 Adequacy and Deficiencies of Database Technology 1.1 Strength and Weaknesses of Database Technology Database systems are currently broadly used for support of data-intensive ser- vices. These broad usage is based on advantages such as the following: Consistent storage of data: Data are uniquely stored in the most actual version. Each user gets the same data. Inconsistency can be avoided. Furthermore, redundancy can be reduced and standards can be enforced. Multi-user support: Data can be consistently shared among different users. Also, conflicting requirements can be balanced. Security is enforced by restricting and managing access to data. Data can be consistently distributed within a network. Integration into component-ware: Currently, database systems are turned into middle-ware components in information-intensive applications. Database op- erating is based on the transaction paradigm. A transaction is a logical unit of work. Database systems are designed to support transactions. Nevertheless, database engines do not completely support complex applications such as internet services, real-time applications, stream information systems and web information systems. LNBIP 5, Springer, 2008, 318-331