. Future middleware platforms will need to be more configurable in order to meet the demands of a wide variety of application domains. Furthermore, we believe that such platforms will also need to be re-configurable, for example to enable systems to adapt to changes in the underlying systems infrastructure. A number of technologies are emerging to support this level of configurability and re-configurability, most notably middleware platforms based on the concepts of open implementation and reflection. One problem with this general approach is that widespread changes can often be made to the middleware platform, potentially jeopardizing the integrity of the overall system. This paper discusses the role of software architecture in maintaining the overall integrity of the system in such an environment. More specifically, the paper discusses extensions to the Aster framework to support the re-configuration of a reflective (component-based) middleware platform in a constrained m...

Middleware technologies such as CORBA and DCOM have been developed as a means of tackling heterogeneity and complexity problems inherent in distributed systems. However, more work still need to be done to develop methodologies for the construction of distributed objects. In addition, little attention has been paid to the development of methodologies for the configuration of computational resources among distributed objects. This paper introduces a resource configuration description language (RCDL) for the specification of the resource management of distributed systems. This language is based on both a resource model and a task model. The former offers various levels of abstraction for resources, resource factories and resource mangers. The latter then provides a fine- and a coarse-grained approach to allocate resources to both application services and middleware services by breaking such services into task hierarchies. Finally, we use reflection as a principled means to obtain a clear separation of concerns between the functional and non-functional behaviour (e.g. resource management) of distributed systems.

Abstract. The characteristics of advanced applications vary a lot and the transaction concept must adapt to fulfil the varying transactional requirements. However, current transaction services do not support the required flexibility in a principled way. We believe that the need for transaction service flexibility can be supported through reflection, where the system is capable of inspecting and changing its own behaviour. The first part of this paper describes a framework for such a reflective transactional system, in which different transaction services can be deployed and modified, and from where a suitable transaction service can be chosen for the execution of a transaction. As concurrent transactions may request different transaction services, the framework allows concurrently active transaction services within the same environment. To ensure correctness when heterogeneous transaction services are concurrently active, a transaction service manager within this framework, provides, among other tasks, transaction service synchronization. The second part of the paper concentrates on service synchronization, and presents algorithms that demonstrate how correctness can be guaranteed when incompatible services are requested. 1

A method is presented for the automatic construction of all possible valid compositions of different middleware software architectures. This allows reusing the latter in order to create systems providing a set of different non-functional properties. These compositions are constructed by using only the structural information of the architectures i.e., their configurations. Yet, they provide a valuable insight on the different properties of the class of systems that can be constructed when a particular set of non-functional properties is required.

Abstract. The concept of software architecture has created a new scenario for incorporating non-functional and transactional requirements into the software design. Transactional and non-functional requirements can be included in an architecture-based software development through formal approaches in which rst-order and temporal logic are utilised to deal with them. In this paper, we present an approach in which transactional and non-functional requirements are formally incorporated into a special class of software architectures, known as dynamic software architectures. In order to demonstrate how this proposal can be utilised in a real application, an appointment system is presented. 1

The specification of a software architecture using different ADLs allows system designers to carry out a number of complementary analyses. In this position paper, we go one step further in this direction by advocating the need for specifying distinct views of a software architecture, each characterizing a specific type of properties (i.e. functional, interaction, and quality properties). Multi-view description of a software architecture raises the issue of combining a set of architectural views so as to derive the resulting overall architecture. We propose some hints on how this can be handled. 1 Introduction It is now recognized that the construction of complex software systems can greatly benefit from the software architecture paradigm [9]. The software architecture of a system describes the system's gross organization using an ADL, which provides notations for the abstract specification of its architectural elements. Practically, the construction of a system from its software archi...

The urgent need to deal with problems that are frequently met in many different families of application led to the evolution and standardization of a software layer that lies between the application and the underlying operating system. This layer is widely known as middleware, and provides reusable solutions to problems like heterogeneity, interoperability, security, transactions, fault tolerance etc. Middleware is typically build based on services provided by an infrastructure. Well known examples of such infrastructures are those complying to the Corba standard, Dcom, Ejb etc. Implementing software based on the reusable solutions provided by middleware infrastructures certainly reduces the cost of the overall software development process. Developers are released from implementing communication protocols, security policies, transaction models. Given a middleware infrastructure, all they have to do is combine available middleware services in a way that satisfies the requirements of a p...

This paper presents a novel approach that simplifies real-time multimedia communication applications development and service provision. Such applications are no longer required to code the creation and management of their real-time communication needs, but are only required to declare them to our middleware. These applications typically require media streams for audio and video, and data channels for control purposes. Moreover they should be able to adapt to available network and host resources. This is particularly important for multi-party applications operating in heterogeneous environments where each party may have different resources available to them. In addition the nature of the heterogeneity may vary over the lifetime of the session for example as network congestion varies or as the terminal resources are shared with other applications or users. A further problem, still, is that the application developer and service provider typically need to address security and charging requ...

In this paper, we present an aspect-oriented specification technique that supports the specification of component-based distributed systems. Importantly, this technique also supports the synthesis of quality of service management components from particular aspects of the specification. We describe how, by using a tool to support our aspect-oriented environment, we can first specify and verify QoS management subsystems and then synthesise components that can be placed into a running system. We focus on dynamic QoS management functions, particularly monitoring and adaptation. We illustrate our approach by giving a simple example of the dynamic QoS management of an audio stream. 1. INTRODUCTION Component technologies are now established as a central technique for the construction of complex applications. According to Szyperski [Szyperski98], a component can be viewed as "as a unit of composition with contractually specified interfaces and explicit context dependencies only". He then ext...

: The indissoluble bonds of computers and failures have produced a plurality of fault tolerant techniques to satisfy, potentially, any dependability requirement. As a consequence, the development of dependable systems is not based on inventing the mechanism that provides the desired dependability guarantees. Rather, it is based on selecting from the existing techniques the one that best meets the system's dependability requirements. Then, some codiøcation of the selected technique can be used as the search-key for retrieving from a repository of fault tolerant mechanisms the one that implements the selected technique. Hence, the development of dependable systems becomes a process that transforms a set of dependability constraints into a fault tolerant mechanism that meets them. The focus of our work is to ensure the rigorous development of dependable systems by creating a formal basis for the aforementioned selection process. More precisely, this formal basis consists of: a system mode...