Peer Hasselmeyer

Managing Dynamic Service Dependencies

We anticipate that software development will be service-centric in the near future. Applications will be created from existing services that are distributed throughout a network. Sure enough, management of those components will be mandatory. While service management is usually service-specific, a few areas can be identified that can be addressed in a generic way. One of these areas is the management of dependencies between services. Despite its genericity, this problem is currently mostly addressed in a service-specific way. This paper details the need for a generic dependency management approach, identifies the key properties of dependencies as well as the requirements for dependency management schemes, describes the model derived from the requirements, and presents an implementation using the Jini connection technology.

A Framework for the Integration of Legacy Devices into a Jini Management Federation

The administration of heterogeneous networks with many devices is a tedious and time-consuming task. Todays approaches only provide static configuration files and make the addition and removal of devices a manual chore. In this paper we present a framework for the integration of legacy devices based on Jini, Sun Microsystems new technology for federating network devices and services. We introduce extended proxy objects called nannies that take care of non-Jini-enabled devices and handle the relevant management events, guide a device through bootstrapping, register it with the lookup service, and provide the implementation of the administrative interfaces of the Jini API. Through this approach both Jini-enabled and legacy devices can be handled homogeneously in a Jini Management Federation.

Towards SLA-Supported resource management

Achievements and experiences in projects with focus on resource management have shown that the goals and needs of High Performance Computing service providers have not or only inadequately been taken into account in Grid research and development. Mapping real-life business behaviour and workflows within the service provider domain to the electronic level implies focusing on the business rules of the provider as well as on the complexity of the jobs and the current state of the HPC system. This paper describes an architectural approach towards a business-oriented and Service Level Agreement-supported resource management, valuable for High Performance Computing providers to offer and sell their services. With the introduction of a Conversion Factory the authors present a component that is able to combine the Service Level Agreement, the system status, and all business objectives of the provider in order to address the business needs of service providers in the Grid.

Trade-offs in a Secure Jini Service Architecture

Jini is an infrastructure built on top of the mobile code facilities of the Java programming language enabling clients and services to spontaneously engage in arbitrary usage scenarios. For a small home or office environment the currently available infrastructure might be adequate, but for mission-critical applications it lacks essential security properties. In the sequel we identify weak points in the Jini architecture and its protocols and propose an extension to the architecture that provides a solution to the identified security problems. We describe the design choices underlying our implementation which aims at maximum compatibility with the existing Jini specifications.

An infrastructure for the management of dynamic service networks

Infrastructures for creating and deploying dynamically cooperating distributed services have become popular. An area that these infrastructures address only insufficiently is the administration of services and their relationships. This article takes an in-depth look at some management topics that are particularly important in a dynamic service environment: service life cycle and service dependencies. It states requirements for these areas that a management system has to fulfill. The article also describes MADYSON (Management Architecture for Dynamic Service-Oriented Networks), a management infrastructure that was developed based on these requirements.

Profiling and internet connectivity in automotive environments

This demo combines active DB technology in open, heterogeneous environments with the Web presence requirements of nomadic users. It illustrates these through profiling of users and Internet-enabled vehicles. A scenario is developed in which useful functionality is provided, such as instrument adjustments, maintenance and diagnostic information handling with the corresponding workflows, and convenience features, such as position-dependent language translation support and traffic information. The customization mechanism relies on an active functionality service.

A Loosely Coupled Federation of Distributed Management Services

This paper describes an architecture for management services. The architecture consists of a number of small components, each performing a highly specialized task. Together, they form a dynamic, highly automated, yet integrated management system. The components are plug-and-play services that use Jini to dynamically establish communication links among themselves and form transient management federations. Our architecture is built around a configuration service which provides for consistent configuration of managed resources. We identify a number of common management scenarios and demonstrate how they can be supported by our system.

Monitoring component interaction in Jini federations

Jini is an infrastructure for spontaneous ad hoc service networks. It allows clients to find services without prior knowledge of their network surroundings. For service interaction proxy objects are used which are supplied by service providers. These proxy objects interact directly with the service provider. Compared to architectures that use a virtually central communications broker (like a CORBA ORB or an e-speak Core), this method offers a large amount of flexibility in the selection of an appropriate communication protocol. On the downside, debugging a distributed application using this approach is rather hard, as the interactions between clients and servers are not visible. This paper describes an approach using Javas dynamic proxies that allows component interaction in a Jini federation to be traced. By putting the functionality into the Jini lookup service, the approach is generic and transparent for both services and clients.