Heiko Pfeffer

Modeling and Controlling Dynamic Service Compositions

Service-oriented computing paradigms have considerably enhanced support for dynamic in modern computing environments by providing loosely coupled interactions between components. The key attainment is the ability to perform a late binding of service compositions, i.e., an allocation of actual services to service place holders during run-time. However, this attainment is limited to the binding of services and cannot operate on the service composition plans itself. Moreover, many service composition representations do not consider the realtime behavior of single services, failing to address a key need of service compositions within dynamic environments that are characterized by a high node mobility. Within this paper, we introduce a service composition model based on a bipartite graph representation that enables the rapid modification of already existing service composition plans. Here, the workflow within a service composition is modeled with a timed automaton while a labeled digraph ensures the correct flow of data between the single services.

A genetic algorithm for the adaptation of service compositions

The view on applications in large-scale open systems shifted to a service-oriented perspective, where each functional feature forming an application is regarded as service. The services which constitute an application can be physically spread over different network nodes and can be even provided by different administrative entities. According to the vision of the BIONETS project we are additionally facing a dynamically changing computing environment, which entails a dynamically changing set of available services. We investigate how service compositions, on which novel applications are based on, can flexibly be adapted to the changing conditions in the computing environment, while going beyond late-binding mechanisms. We apply methods of genetic programming to modify the structures describing service compositions to find compensation for types of services no longer available to applications. In this paper, we describe the current state of our efforts on complex algorithms for service composition transformation, based on the application of genetic operators to graph based service composition representations.

BIO-INSPIRED PROCESSING AND PROPAGATION OF SEMANTICS IN LOOSELY COUPLED COMPUTING ENVIRONMENTS

Human-centric computing has grown to be the major influence in todays computing research. Due to demand from industry and even lawmakers for easy-to-use computer systems, the user is now regarded as being the center of a ubiquitously available environment that supports the execution of task and anticipates user actions. This environment allows for the establishment of completely new ways for the delivery of legacy services and represents an opportunity for the introduction of a new type of services, addressing the user-focused service consumption. As a cause of this shift, the increasing saturation of everyday environments with computing devices can be identified. This saturation implies a numerical growth of computing systems and entails an increasing complexity, which negatively influences maintainability and manageability. Moreover, the shortcomings caused by the mobility of system elements, a common trait of human-centric environments, require consideration about the reliability of cooperative actions. In this paper, we present an approach that copes with complexity and dynamic while making service-oriented systems autonomous by the use of bio-inspired concepts. In particular, the aim is to make service architectures environment-aware. Thus, service architectures are supposed to adapt autonomously to their current environment like biological species do to survive. This approach requires services to obtain knowledge about characteristics and state of the environment through gathering semantically enhanced information about the context of the computing environment, which is intended to help in forming a virtual counterpart of the real world as reference for service adaptation. For this purpose, we illustrate the architecture for context provisioning in highly dynamic computing environments. As base for this architecture a middleware is introduced utilizing a loosely coupled interaction model. Moreover, a pheromone-based concept is outlined to optimize the dissemination of context data in the absence of adequate context sources.

Fuzzy Modeling of Resource Consumption for Service Composition Evaluation

The increasing numbers of different services that execute similar tasks offer the user a tough choice. Since most services perform nearly identical or their properties are not comparable easily, users are often not in the position to differ between various qualities of service. The situation becomes even worse if multiple single services are combined into complex service compositions that are elusive for the user. The distribution of services of a composition on a multiplicity of devices with heterogeneous performances requires for optimal resource allocation, since resources are limited and the user is interested in the best workload. In order to allow a user to formulate vague estimation of service behavior on resource consumption, this paper introduces a fuzzy model for the representation of such user preferences based on the model proposed by Pfeffer et al. [1]. The specified model can be combined easily with the existing model for fuzzy service properties. This allows for human-centric service composition evaluation on resource consumption. Both models allow a lightweight and fast evaluation process that is also introduced.

SmartWare: framework for autonomic application services

The combination of mobile and embedded computing devices holds potential for a novel landscape of application and services in the direct surrounding of the user. The realization of such a landscape requires solutions to cope with the volatile nature of the environment composition, the absence of central management infrastructure, and the heterogeneity of resources. To unburden developers of application services from searching these solutions themselves, we started developing a generic software framework called SmartWare. SmartWare is a collection of principles and tools that are intended to simplify and accelerate the development of application services at the edge of the Internet and beyond. In this paper we describe the features of SmartWare, a prototype implementation and a test application we realized with the framework prototype.

BIONETS: Self Evolving Services in Opportunistic Networking Environments

This paper presents the BIONETS opportunistic service evolution platform. The proposed platform allows pervasive services to evolve over time by exploiting opportunistic communications among mobile nodes on the one hand, and evolutionary computation techniques on the other. We present the main components of the platform, describing their functionalities and technical implementation. Finally, we present the hardware–in–the–loop approach we have followed to evaluate it, where a simulation platform, in charge or reproducing a large number of mobile nodes communicating wirelessly, is integrated with a real software prototype.

Decoupling animation development of realtime applications in distributed environments

The conditions arising for the realization of visualization-oriented applications have recently changed. Especially in pervasive environments, where predominantly small and limited devices are present, the need for a distributed realization of relevant visualization processes is present. Due to the impact of user interaction at runtime, issued by e.g. pervasive applications, animations needs to be describable by developers in a functional manner, relative to an animationpsilas current state. Therefore, we propose a framework for the development of animation-centric visualizations in distributed environments. Decoupling the development of animations in complexity and representation from their final visualization is thereby achieved by the modelling and realization of a distributed visualization architecture and an according functional animation related programming pattern. Finally, we describe stEVE, a case study for the suitability of our framework aiming at the demonstration of biologically inspired processes in next generation computing systems.

Context Data Dissemination in the Bio-inspired Service Life Cycle

Summary. Service provisioning in ad hoc and mobile environment introduced the need for new approaches to service adaptation and dissemination of appropriate context data for e.g. service management. Therefore, this paper introduces a novel bio-inspired service life cycle, leveraging enhanced service adaptation at the design time of services through guided service evolution, i.e. mutation and appropriate fitness selection of Service Individuals. Furthermore, we describe a context data dissemination algorithm to efficiently distribute the information needed in the bio-inspired service life cycle.