In this report, we give an overview of the SFB 501 approach on developing abstractions and concepts needed to achieve the long-term objective of realizing a genuine universal media server. Such a media server is supposed to supplement repositories of any kind (e. g., digital libraries) by providing a high-level interface that allows applications to access and process media data over a network (including the world-wide web). We focus primarily on the problem of optimally processing raw media data at the server. The traditional data independence abstraction, which is to be realized by a universal media server, is known to be impeding the processing of media data with optimal performance and/or quality. However, instead of abandoning data independence (which is most usual for media servers today), we propose an enhancement that is particularly tailored to the characteristics of networked media data processing: transformation independence. Conceptually, transformation independ...

Query optimizers generate plans to retrieve data specified by queries. Query optimization for object databases (i.e., object-oriented and object-relational databases) is an immature field, and stands to benefit from adaptation of techniques that have proved useful for relations. One technique uses query-to-query transformations to rewrite queries into queries that are potentially more amenable to plan generation. For transformations to be useful, they must preserve the semantics of the queries they rewrite (correctness) and usually result in queries that generate better plans (effectiveness). Object databases complicate the expression of correct and effective transformations. Transformation correctness is problematic even for relational queries. Especially error-prone are transformations that rewrite complex nested queries (queries containing other queries) or queries that return duplicates. Objects make correctness more difficult because object queries can be far more compl...

Query optimizers are complex subsystems of database management systems. Modifying query optimizers to admit new algorithms or storage structures is quite difficult, but partly alleviated by extensible approaches to optimizer construction. Rule-based optimizers are a step in that direction, but from our experience, the rule sets of such optimizers are rather monolithic and brittle. Conceptually minor changes often require wholesale modifications to a rule set. Consequently, much can be done to improve the extensibility of rule-based optimizers. As a remedy, we present a tool called Prairie that is based on an algebra of layered optimizers. This algebra naturally leads to a building-blocksapproach to rule-set construction. Defining customized rule sets and evolving previously defined rule sets is accomplished by composing building-blocks. We explain an implementation of Prairie and present experimental results that show how classical relational optimizers can be synthesized from building-blocks, and that the efficiency of query optimization is not sacrificed. 1