Jürgen Wäsch

Deriving service models in cross-organizational workflows

Competitive markets force companies to form virtual enterprises by outsourcing activities to external service providers. The workflow concept has been very successful in streamlining business processes by automating the coordination of activities, but has so far been limited to use within single organizations. To address the problems of cross-organizational workflows, we use a service-oriented workflow model. Within this approach, we present a technique to derive a model of the external services, based on continuous-time Markov chains, by analysing their externally observable behaviour. This allows one to assess the quality of the external services without compromising the autonomy of the service providers.

Tamino - An Internet Database System

Software AGs Tamino is a novel database server designed to fit the needs of electronic business and worldwide information exchange via the Internet. It is not just an on-top solution based on a database system originally designed for use in other application areas. Rather, it is entirely designed for the specific scenario of HTTP-based access to data represented in XML. These data can stem from various sources, and can be combined on the fly when a corresponding request is encountered. This paper sketches the architecture and the functional features of Tamino, and justifies its various design decisions.

Adaptive Outsourcing in Cross-Organizational Workflows

The workflow concept has been very successful in streamlining business processes by automating the coordination of activities, but has so far been limited to the use within single organizations. Any attempt to distribute workflows among different organizations has to face the problems posed by the complex relationship among autonomous organizations and their services. To address these problems we propose a service-oriented model for cross-organizational workflows. Modeling the workflow execution as a cooperation of services allows different organizations to interact via well-defined interfaces. We further show how the execution can be optimized by selecting services depending on their contribution to quality criteria of the workflow.

History merging as a mechanism for concurrency control in cooperative environments

Cooperative applications need proper transactional support for coordinating joint activities, sharing of data, and semantically correct exchange of information of collaborating users. Conventional transaction models based on the ACID properties do not meet the typical requirements of cooperative applications. The CoACT model (M. Rusinkiewicz et al., 1995) is designed for supporting cooperative work in multi user environments. CoACT includes a novel algorithm implementing the semantically correct exchange of information among concurrent activities of cooperating users by means of merging histories of user activities. This technique of merging activity histories is applicable in various fields beyond CSCW systems, e.g., in the area of mobile computing to manage disconnected operation.

HyperStorM: an extensible object-oriented hypermedia engine

ABSTRACT It is a well-known problem that developers of hypermediaapplications need assistance for modeling and maintainingapplication-specific hypermedia structures. In the past, vari-ous hypermedia engines have been proposed to support thesetasks. Until now, hypermedia engines either provided a fixedhypermedia data model and left extensions to the hyper-media application or they left the modeling of the hyper-media data completely to the application developer and onlyprovided storage functionality which had to be plugged intothe data model by the application developer. As an alterna-tive, we propose an extensible object-oriented hypermediaengine which supports the specification of application se-mantics as application classes within the hypermedia en-gine, thereby supporting complex operations maintainingapplication-specific as well as application-independentconstraints. In the HyperStorM hypermedia engine, the storage layer andthe application layer of a hypermedia system are implement-ed within the object-oriented database management systemVODAK. Only the presentation layer is kept outside theOODBMS. This approach facilitates both the reuse of data-base functionality as well as the flexibility necessary to sup-port the efficient development of different kinds ofhypermedia applications. First evaluations show that our ap-proach presents a much more powerful hypermedia enginethan previous approaches, thus giving a new quality to hy-permedia application development.

Flexible Design and Efficient Implementation of a Hypermedia Document Database System by Tailoring Semantic Relationships

In this paper we present the design concepts and data modeling approach that was used to define a general application framework for storing hypermedia documents in the object-oriented database management system VODAK. We exploit the capabilities of the VODAK data model to introduce new hypermedia modeling primitives at the meta level. We show that representing the hypermedia semantics within the DBMS in this way is clearly advantageous for efficiency of the design and implementation of hypermedia document storage engines. As an example, we give a concrete realization of a DBMS-based hypermedia engine for the SEPIA hypermedia authoring system developed at GMD-IPSI. With this example we show that the data model extensions provided are flexible enough to represent also complex semantic hypermedia concepts, that the development cycle of a hypermedia engine can become extremely fast and that the resulting implementation has adequate performance for interactive applications.

Transactional support for cooperative applications

Cooperative work on shared, persistent data requires computing system support to coordinate the work of multiple users and to ensure data consistency. Conventional database transaction models do not meet the requirements of cooperative applications. Isolation of transactions, as guaranteed by the ACID properties, contradicts the need of cooperation between users. In this paper, we investigate different advanced transaction models that target at improved support for cooperative applications. To that extent we analyze typical cooperative application scenarios and derive from that basic requirements for consistent access to shared databases. In particular we discuss in detail the cooperative transaction model that has been developed within the TRANSCOOP project. This model supports alternating periods of individual and joint work, and allows to exchange and share information consistently. It provides transactional execution guarantees both for the work of single users and for the overall cooperative effort. The model is applicable to a wide spectrum of cooperative applications ranging from creative design applications to structured workflow-like applications.

The Transcoop Transaction Model

This chapter describes the TRANSCOOP cooperative transaction model CoACT, and its underlying paradigm of cooperative work. As motivated in Chapter 4 new application areas of the transaction concept impose new requirements on transaction models. In all of the discussed application areas there is a need to support the cooperation of human actors engaged in common tasks and working towards a common goal. a transaction model supporting cooperation should make the sharing and exchange of information among co-workers as natural as possible while still preserving a notion of consistency.

Global nested transaction management for ODMG-compliant multi-database systems

Object technology has received considerable attention in the re cent years for the integration of various local data sources in a multi-database system (MDBS). An object-oriented data model like ODMG-93 can serve as the canonical data model for the global layer in the MDBS. The ODMG-93 standard defines a closed nested ACID transaction model without intra-transaction parallelism. Although the nested transaction model provides special support for distributed computing enviromnents, the implications of its usage as a global transaction model for MDBS have so far not been examined. In this paper, we present a simple and practical method to implement global nested transaction management for ODMGcompliant multi-database systems. The main contribution of our work is the adaption of the ticket method to achieve global serializability for global nested transactions that conform to ODMG. Our approach is also applicable to OMG’s common object request broker architecture. We have successfully implemented our global nested transaction model within the IRO-DB multi-database system.

Cooperative Data Management and its Application to Mobile Computing

Mobile computing and workgroup computing are emerging technologies which have so far been treated independently. Current approaches to support cooperative work neglect the special characteristics of mobile environments like limited bandwidth or temporary disconnection. On the other hand, approaches to support disconnected operation rely on the assumption that the degree of data-sharing is low which is obviously not appropriate for cooperative work. In this paper, we utilize the COACT cooperative transaction model to provide support for parallel activities in mobile environments. We present a system architecture that is able to cope with the special characteristics of mobile environments and a formal framework for the consistent information exchange between mobile users. The paper shows how the COACT history merge algorithm reduces conflicts by exploiting operation semantics and offering consistent sequences of operations. We believe that our new approach is a viable solution to the growing demand for cooperation in mobile environments.