Ralf-Dieter Schimkat

A Service-Based Agent Framework for Distributed Symbolic Computation

We present OKEANOS, a distributed service-based agent framework implemented in Java, in which agents can act autonomously and make use of stationary services. Each agents behaviour can be controlled individually by a rule-based knowledge component, and cooperation between agents is supported through the exchange of messages at common meeting points (agent lounges). We suggest this general scheme as a new parallelization paradigm for Symbolic Computation, and demonstrate its applicability by an agent-based parallel implementation of a satisfiability (SAT) checker.

Retrieval for decision support resources by structured models

The number of available DSS within organizational Intranets will soon require efficient retrieval functionality. While Web retrieval technology performs excellent on documents, computational services need approaches that capture the semantics of resources. We present a retrieval approach that uses a variant of Structured Modeling to represent resources. It allows the use of similarity of models for retrieval. Exact similarity computation is shown to be NP-hard, and efficient heuristics for similarity computation and filter algorithms are introduced. We report an evaluation in a classroom experiment and give computational results on a benchmark library.

A lightweight, message-oriented application server for the WWW

In this paper, we present a lightweight system that loosely couples and integrates any kind of source systems with information retrieval capabilities. The system provides mechanisms for a rapid integration of source systems into the World Wide Web (WWW), allowing the generation of configurable collections of individual, heterogeneous source systems. The proposed system is based on a lightweight, messageoriented application server and an object-oriented client framework to provide an uniform, Java-based graphical user interface. The system offers application middleware functionalities that solve issues such as limited bandwidth and scalability of both sides ( W W W user clients and backend server systems) in a generic manner. Already integrated source systems include the system PRoGPd~SS (a method base system for mathematical algorithms with functionalities for server-side computing) and SPECTO (an XML-based, distributed monitoring system).

Scalability and interoperability in service-centric architectures for the Web

The Internet provides a computing infrastructure that is highly distributed by nature. In the future the number of actively and cooperatively interacting parties on the Internet, such as traditional computers, embedded devices or various service providers in the field of electronic commerce, will increase further. In such an environment, traditional client/server architectures reach the limit of their capability. Decentralized, distributed software architectures which actively support the object and code migration paradigm, are more suitable. We present a sophisticated approach to a scalable and interoperable service architecture for the Web. We propose a loosely coupled network of service nodes which supports scalability, integration and management of heterogenous services in an uniform way. Our proposed service infrastructure provides a platform to easily locate, connect, integrate and use services of any type. It addresses issues of bandwidth, security, distribution and integration, which are particularly important within the Web, at a general framework level.

Deploying Distributed State Information in Mobile Agent Systems

Agent-based distributed problem solving environments decompose a given problem into a set of sub problems which can be processed in parallel and independently by autonomous and mobile agents at each computation node. Such an autonomous agent primarily makes use of local information which is provided at the respective computation node. This kind of information is characterized by its potential incompleteness and inconsistency with regard to the overall distributed system state which is due to the lack of any centralized coordination facility. In this paper, we introduce the use of long-term knowledge repositories for autonomous agents without sacrifying the autonomy of the agents and without introducing any central management facility. We compare our approach to an agent-enabled distributed SAT prover which makes only use of local system state information. In that problem solving application a given search tree is distributed dynamically by autonomous mobile agents implemented in pure Java. To demonstrate the profit of using knowledge repositories in general, we integrated our agent system Okeanos into our XML-based monitoring system Specto and the lightweight, distributed event-based middleware Mitto. Our cooperative approach does not conflict with the decentralized parallelization algorithm of the distributed SAT prover. Empirical results show that our approach can contribute to the performance of distributed symbolic computation. In this example, a load balancing subsystem is implemented taking the now available global information about the system state appropriately into account.

A General, Web-Enabled Model Retrieval Approach

Models play a key role in the design of information systems. Apart from graphical tools supporting the creation of models, there is only limited support for a retrieval on models which goes beyond text-based search. The rapidly increasing number of business information systems demand the reuse of existing models which depends heavily on the effective retrieval of appropriate models. In this paper we introduce Firestorm, an object-oriented framework for general model retrieval based on so-called three-layer structured service models. Firestorm provides extensible algorithmic retrieval facilities in the sense that it can be applied to arbitrary model domains.

An Information System for Retrieving and Reasoning about Xml-based Mathematical Knowledge

Xml has become a key language for information interchange and integration over the World Wide Web. Representing and storing mathematical achievements and knowledge in a self-describing, extensible and open manner facilitates Web information systems which simplify co-operation among mathematicians world wide.

Living Hypertext — Web Retrieval Techniques

In this paper, we present a document metaphor called Living Documents for accessing and searching for digital documents in modern distributed information systems. Our approach is based upon a finegrained document concept which glues computational services, data and meta data together. Viewing documents as micro servers is particularly well suited in environments where the document’s content is changing continuously and frequently. Based on a case study of an existing state-of-the-art Web application, we show how to transform database-centric information systems into a hypertext of inter-linked Living Documents. We also discuss how to effectively use traditional as well as Web information retrieval techniques, namely topic distillation, in such hypertext environment. In particular, an extended version of Kleinberg’s [11] algorithm is presented.