A system for data change propagation in heterogeneous information systems
Carmen Constantinescu, Uwe Heinkel, Ralf Rantzau, Bernhard Mitschang
Abstract
Carmen Constantinescu, Uwe Heinkel, Ralf Rantzau, Bernhard Mitschang
Abstract
Abstract: Today, it is common that enterprises manage several mostly heterogeneous information systems to supply their production and business processes with data. There is a need to exchange data between the information systems while preserving system autonomy. Hence, an integration approach that relies on a single global enterprise data schema is ruled out. This is also due to the widespread usage of legacy systems. We propose a system, called Propagation Manager, which manages dependencies between data objects stored in different information systems. A script specifying complex data transformations and other sophisticated activities, like the execution of external programs, is associated with each dependency. For example, an object update in a source system can trigger data transformations of the given source data for each destination system that depends on the object. Our system is implemented using current XML technologies. We present the architecture and processing model of our system and demonstrate the benefit of our approach by illustrating an extensive example scenario. 1.
OpenAlex reports 11 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract: Today, it is common that enterprises manage several mostly heterogeneous information systems to supply their production and business processes with data. There is a need to exchange data between the information systems while preserving system autonomy. Hence, an integration approach that relies on a single global enterprise data schema is ruled out. This is also due to the widespread usage of legacy systems. We propose a system, called Propagation Manager, which manages dependencies between data objects stored in different information systems. A script specifying complex data transformations and other sophisticated activities, like the execution of external programs, is associated with each dependency. For example, an object update in a source system can trigger data transformations of the given source data for each destination system that depends on the object. Our system is implemented using current XML technologies. We present the architecture and processing model of our system and demonstrate the benefit of our approach by illustrating an extensive example scenario. 1.
Key concepts: Computer science, Legacy system, Information system, XML, Data modeling, Distributed computing, Enterprise information integration, Schema (genetic algorithms)