Catálogo de publicaciones - libros

The performance requirement for the transformation of messages within electronic business processes is our motivation to investigate in automatic capacity planning methods. Performance typically means the throughput and response time of a system. Finding a configuration of a distributed system satisfying performance goals is a complex search problem that involves many design parameters, like hardware selection, job distribution and process configuration. Performance models are a powerful tool to analyse potential system configurations, however, their evaluation is expensive, such that only a limited number of possible configurations can be evaluated. In this paper we give an overview of our automatic system design method and discuss the arising problems to achieve the performance during the runtime of the systems. Furthermore we make a discussion on the impact of our strategy on the current trends in distributed systems.

Service oriented architectures facilitate loosely coupled collaborations, which are established in a decentralized way. One challenge for such collaborations is to guarantee consistency, that is, fulfillment of all constraints of individual services and deadlock-freeness. This paper presents an overview on a decentralized approach to consistency checking, which utilizes only bilateral views of the collaboration.

This work presents a schema evolution methodology able to propagate structural and semantic modifications occurring in the local schemas of a federated database to the integrated schema. Our approach is to regard this problem from a schema integration point of view. Our theoretical framework is based on a declarative schema integration methodology, which reduces schema integration to the resolution of a set of equivalence correspondences between arbitrarily complex local subschemas.

Content-Addressable Networks (CAN) are able to manage huge sets of (key,value)-pairs and cope with very high workloads. They follow the peer-to-peer (P2P) paradigm in order to build scalable, distributed data structures on top of the Internet. CAN are designed to drive Internet-scale applications like distributed search engines, multimedia retrieval systems and more. In these scenarios, the nearest-neighbor (NN) query model is very natural: the user specifies a query key, and the engine responds with the set of query results closest to the key. Implementing NN queries in CAN is challenging. As with any P2P system, global knowledge about the peers responsible for parts of the query result is not available, and the communication overhead is the most critical factor. In this paper, we present our approach to realize efficient NN queries in CAN. We evaluate our NN query processing scheme by experiments with a CAN implementation in a setting derived from web applications. The results of our experiments with 10.000 peers are positive: even large result sets with a precision of 75% can be obtained by invoking less than 1.6 peers on average. In addition, our NN protocol is suitable for prefetching in settings with sequences of consecutive queries for similar keys.

The Semantic Web aims to provide access to information for humans and machines. Practical implementations of Semantic Web technologies need to consider aspects such as scalability and reliability to be attractive for industrial practitioners. Our approach shows a portal infrastructure which benefits on the one hand from the access facilities provided by Semantic Web technologies and on the other hand from the applicability of current Web technologies for industrial strength applications. The main achievement of the approach is an up-and-running network of content-interchanging portals. Accompanying tools and methods are provided for setting up such networks from scratch or to put them on top of existing information sources. In the paper, special attention is given to the ontology-supported query refinement service that enables a user to find the relevant results even if his query does not match his information need ideally.

This paper discusses the requirements and tasks in annotating existing Web pages with additional structural and semantic information. It suggests annotating existing Web pages with the concepts of a domain model, and annotating the links between the Web pages with the relations between these concepts. It also suggests explicitly representing high-level hypermedia structures, so-called hypertext composites and contexts, and annotating the corresponding composite and context pages with such structures. RDF and RDF Schema are adopted to represent the domain models and the resulting annotations. The architecture of a prototype annotation tool is outlined and corresponding requirements for automatic annotation support are discussed.

The Dublin Core Metadata Initiative (DCMI) maintains a vocabulary of several dozen metadata terms, notably the fifteen-element “Dublin Core.” These terms (and their historical versions) are identified with URI references, described in Web documents and machine-processable schemas, indexed in registries, cited in application profiles, and of course used in metadata records. Within DCMI, however, the emphasis has been less on growing this small vocabulary than on clarifying how the DCMI vocabulary can “play well” with other, complementary vocabularies in a Semantic Web that recombines semantics from multiple sources for specialized purposes. This priority has led DCMI to examine models for referencing other vocabularies and to clarify the modeling bases for interoperability among heterogeneous systems.

The distribution of large content files, like videos, over a large number of users is a demanding and costly operation if done using a traditional client/server architecture. Peer-to-peer based file-sharing systems can be used as an alternative for content distribution.

The eDonkey file-sharing network is one of the most successful peer-to-peer file-sharing networks, especially in Germany. eDonkey forms a hybrid network that capitalizes both on the client/server and peer-to-peer paradigms in the design of its architecture.

In this paper, we describe the eDonkey protocol, the constructed overlay network, the critical operations and their characteristics, as well as the results of measurements of the network and transport layer and of the user behavior. The measurements were made with the client software and with an open-source eDonkey server we extended explicitly for these measurements. Our study shows that eDonkey is particularly well suited for content distribution and not surprisingly also intensively used for the distribution of large files, mainly videos.

Today more and more media data is produced completely in the digital domain without the need of analogue input. This brings an increase of flexibility and efficiency in media handling, as distributed access, duplication and modification are possible without the need to move or touch physical data carriers. But this also reduces the security of the process: Without physical originals to refer to, changes in the material can remain unnoticed, at the end making the manipulated data the new original. Theft and illegal copies in the digital domain can happen without notice and loss of quality. We therefore see the need of setting up secure media production environments, where access control, integrity and copyright protection as well as traceability of individual copies are enabled. Addressing this need, we design a framework for media production environments, where mechanisms like encryption, digital signatures and digital watermarking help to enable a flexible yet secure handling and processing of the content.

In Germany, the public safety system is largely organized by the German Federal States, which operate, among other equipment, a fleet of Nuclear, Biological and Chemical Reconnaissance Vehicles (NBC RVs) to take measurements in contaminated areas. Currently, NBC RV staff verbally report measured data to a Control Center Unit (CCU) over the assigned Public Safety Organization (PSO) analog voice radio channel. This procedure has several disadvantages. The channel is not secure and its capacity is wasted, which places a limit on the achievable throughput and thus on the number of NBC RVs that can be operational simultaneously. Also, while data is being reported, other PSO members are blocked from sending, and operating personnel is distracted from other work. To overcome these problems, we propose a heterogeneous and flexible communication platform that complies with reliability and coverage requirements for PSO. More specifically, our proposed system is designed to replace current ways of communicating between NBC RVs and the CCU. A drastically higher amount of data can then be transmitted to the CCU, and it can be processed in a much more effective manner in the CCU as well as in cooperating PSO units. Ultimately, this will improve NBC RV missions and consequently shorten PSO response time when dealing with NBC disasters.