Abstract. More than ever before schema transformation is a prevalent problem that needs to be addressed to accomplish for example the migration of legacy systems to the newer OODB systems, the generation of structured web pages from data in database systems, or the integration of systems with different native data models. Such schema transformations are typically composed of a sequence of schema evolution operations. The execution of such sequences can be very time-intensive, possibly requiring many hours or even days and thus effectively making the database unavailable for unacceptable time spans. While researchers have looked at the deferred execution approach for schema evolution in an effort to improve availability of the system, to the best of our knowledge ours is the first effort to provide a direct optimization strategy for a sequence of changes. In this paper, we propose heuristics for the iterative elimination and cancellation of schema evolution primitives as well as for the merging of database modifications of primitives such that they can be performed in one efficient transformation pass over the database. In addition we show the correctness of our optimization approach, thus guaranteeing that the initial input and the optimized output schema evolution sequence produce the same final schema and data state. We provide proof of the algorithm's optimality by establishing the confluence property of our problem search space, i.e., we show that the iterative application of our heuristics always terminates and converges to a unique minimal sequence. Moreover, we have conducted experimental studies that demonstrate the performance gains achieved by our proposed optimization technique over previous solutions.

The age of information management and with it the advent of increasingly sophisticated technologies have kindled a need in the database community and others to re-structure existing systems and move forward to make use of these new technologies. Legacy application systems are being transformed to newer state-of-the-art systems, information sources are being mapped from one data model to another, a diversity of data sources are being transformed to load, cleanse and consolidate data into modern data-warehouses [CR99]. Re-structuring is thus a critical task for a variety of applications. For this reason, most object-oriented database systems (OODB) today support some form of re-structuring support [Tec94, Obj93, BKKK87]. This existing support of current OODBs [BKKK87, Tec94, Obj93] is limited to a pre-defined taxonomy of simple fixed-semantic schema evolution operations. However, such simple changes, typically to individual types only, are not sufficient for many advanced applications [Bré96]. More radical changes, such as combining two types of redefining the relationship between two types, are either very difficult or even impossible to achieve with current commercial database technology [Tec94, Obj93]. In fact, most OODBs would typically require the user to write ad-hoc programs to accomplish such transformations. Research that has begun to look into the issue of complex changes [Bré96, Ler96] is still limited by providing a fixed set of some selected (even if now more complex) operations. To address these limitations of the current restructuring technology, we have proposed the SERF framework which aims at providing a rich environment for doing complex user-defined transformations flexibly , easily and correctly [CJR98b]. The goal of our work is to increase the usability and utility of the SERF framework and its applicability to re-structuring problems beyond OODB evolution. Towards that end, we provide re-usable transformations via the notion of SERF Templates that can be packaged into libraries, thereby increasing the portability of these transformations. We also now have a first cut at providing an assurance of consistency for the users of this system, a semantic optimizer that provides some performance improvements via enhanced query optimization techniques with emphasis on the re-structuring primitives [CNR99]. In this demo we give an overview of the SERF framework, its current status and the enhancements that are planned for the future. We also present an example of the application of SERF to a domain other than schema evolution, i.e., the web restructuring.

No abstract