Bernhard Volz

Towards an open meta modeling environment

Conventional modeling environments support either only a two layered meta hierarchy or do not provide (full) support for advanced modeling paradigms that go beyond the capabilities of the Meta Object Facility (MOF). Within this article we introduce the foundation of a meta modeling environment that supports Powertypes, Clabjects, Deep Instantiation and Materialization.

A Meta Modeling Framework for Domain Specific Process Management

Process Management has become an acknowledged technology for application integration. However, different applications leverage from different process modeling capabilities. Thus, domain specific process management becomes more and more relevant. In this paper we present our solution for an abstract process modeling method and language based on an extensible meta modeling framework which has two main advantages compared to standard MDA tools. First, we can easily implement modeling patterns (here: powertypes and type/usage concept). Second, we can use more than two meta layers which results in a more clear structure and in higher flexibility (here a separation between general process modeling principles and domain specific languages that can better express domain specific semantics).

Towards a Generic Cloud-Based Virtual Research Environment

Virtual collaboration is an important aspect for the success of scientific projects, especially if participating researchers are distributed over the whole globe. In the recent past some systems -- so called virtual research environments -- were presented to support collaborative work restricted to certain research domains. Within this article a concept of a generic framework for building personal, cloud-based virtual research environments easily is proposed. Such an environment could be defined by composing arbitrary services, appropriate to the requirements of a particular scientist. Due to low funds in some scientific areas, we also provide a flexible billing strategy using the cloud specific pay-per-use model. Thus, each service has just to be paid as long as it is utilized.

Agent Assignment for Process Management: Agent Performance Evaluation Framework

Convergence of data mining and process management is ideal – but still limited. An example of such a convergence is presented in the form of APE Framework that addresses the problem of static-agent-assignment-strategies in Workflow Management Systems (WfMS) – one cause of poor business process performance since all eligible agents may be assigned to a task instead of only assigning those which are expected to be “successful”. To solve this problem, the APE Framework first identifies successful agents by investigating the history of workflow executions with methods from data mining and taking the process model into account as a source for domain knowledge. It then updates its findings into an organizational database – a data source wherefrom WfMS make assignments. Thus, through this convergence, the APE Framework enables WfMS to allocate only successful agents – instead of merely static list of all eligible agents.

A conceptual modeling and execution framework for process based scientific applications

In recent years, scientists are dealing more and more with data intensive and complex applications. Many scientific workflow systems emerged which adapt technology and methods stemming from the workflow management area and that should support scientists in understanding and working with their complex scenarios. However as these systems often descend from problem solving environments, many of them are missing a well structured conceptual method for process modeling and execution as a foundation. In this publication we present a comprehensive and well structured method for developing and analyzing process based scientific applications. This method is constituted by a process modeling framework, a data integration framework and a model driven approach to build up infrastructures for process modeling and execution.

Agent Assignment for Process Management: Goal Modeling for Continuous Resource Management

Workflow Management Systems (WfMS) support modeling and execution of business processes, but they lack to define a criteria that can be used to determine how successfully certain processes are being performed by authorized agents. As a consequence, agents go on and on with their work even they have a poor success history and thus cause a process to become inefficient. Therefore, this paper introduces means for including goal modeling into workflow modeling, enabling a WfMS not only to support performance evaluation mechanisms but also to select those agents for a certain task who will most likely be performing best.

DaltOn: An Infrastructure for Scientific Data Management

It is a common characteristic of scientific applications to require the integration of information coming from multiple sources. This aspect usually confronts end-users with data management issues which involve the transportation of data from one system to another as well as the syntactic and semantic integration of data, i.e. data come in different formats and have different meanings. In order to deal with these issues in a systematic and well structured way, we propose a sophisticated framework based on process modeling. In this paper, we present the three major conceptual architectural abstractions of the system and detail its execution.

Deriving biological applications from domain specific process models

Abstract In this paper we present how the process modeling and execution tools iPM and iPE can be used to model and execute biological processes. The main focus of this paper is on the flexibility of iPM and iPE with respect to the customization to the biological application domain. We will demonstrate the flexibility of our modeling methodology by giving two examples: Modeling the invocation semantics of web services used in the biological application domain and the processing of streamed data.

Semantic data integration in the DaltOn system

Due to the large volume and high complexity of data, end-users are often confronted with data management issues such as syntactic and semantic integration of data (data comes in different formats and has different meanings) as well as the pure movement of data in between information systems. In order to cope with these issues in a systematic and well structured manner, we propose an elaborate framework based on process modeling, data provision, data integration and a repository which tracks all data management issues are the central components of our approach. It is out of the scope of this paper to reflect all the components in detail, instead the main focus is to present how our framework deals with semantic data integration using ontologies in theory and in practice by giving a real-world example. In this example we describe a self medication application from the pharmaceutical domain.

Data Integration with the DaltOn Framework --- A Case Study

Data integration has gained a great attention in current scientific applications due to the increasingly high volume of heterogeneous data and proliferation of diverse data generating devices such as sensors. Recently evolved workflow systems contributed a lot towards scientific data integration by exploiting ontologies. Even though they offer good means for modeling computational workflows, they were proved not to be sufficiently strong in addressing data related issues in a transparent and structured manner. The DaltOn system improves the productivity of scientists by providing a framework which copes with these issues in a transparent and well structured manner. In this paper we will elaborate its application in a real world scenario taken from meteorological research where data are retrieved from a sensor network and are integrated into a central scientific database.