We propose an object-oriented approach to the development of distributed applications emphasizing separation of concerns. Our approach combines the needs of transparency, encapsulation of distribution issues, and support for non-traditional models, where cooperation and sharing are desired. The development process is constructive, thus allowing partial verification of results. We recognize seven concerns: fragmentation, replication, naming, concurrency, failure, configuration, and communication. Each concern as perceived in three levels of abstraction: model, policy and mechanism. We also propose an integration process centered on development stages. Both concerns and stage perspectives are part of an integrated and flexible development process.

This paper describes the major issues in the design and implementation of orthogonal persistence in IK. A single and uniform programming paradigm is used to manipulate objects in a persistent and distributed environment. Object references can be freely passed during remote invocations or stored persistently. IK supports orthogonal persistence with type inheritance. Objects are stored persistently when reachable from an Eternal Root, regardless of their type. For programmers, objects are created and manipulated uniformly, independently of the time they persist. Persistent objects are dynamically grouped at run-time into clusters to encapsulate fine-grain language level objects into coarser-grain entities. We present a novel approach to integrate object clustering, naming and garbage collection in persistent systems, and present experimental results.

The paper describes the user interface (UI) environment developed for the SOMIW (Secure Open Multimedia Integrated Workstation) project. The basic goals of the project were the definition of a comprehensive architecture handling all relevant UI concepts and the implementation of an interface generator integrated in the programming environment. Others goals were related to the specific requirements of a multimedia workstation (image, voice synthesis and recognition), distribution and integration with the operating system. The architecture of the user interface management system is based on a conceptual virtual environment defined by an object-oriented computational model, and visualisation and input models. This environment is described in the initial sections, followed by an outline of the specification and generation tools. We conclude by describing the internal details of IMAGES, the implementation techniques in development and the target machines.

The authors discuss the definition and design of a generic reliable communication architecture on a widely used host-independent platform, such as a local area network (LAN). Two relevant aspects are the use of nonreplicated LANs and self-checking components. The protocol is innovative, in the sense that, although clockless and running on a nonreplicated network, it displays bounded execution times. Thus the architecture is capable of reliably addressing realtime. Support of high-performance real-time applications with this architecture is being seriously considered in the present phase of project Delta-4, a CED Esprit II consortium designing an open, dependable distributed architecture. The authors' considerations regarding synchronism properties of clockless protocols are being applied in this context.>

Comandos is a project within the European Strategic Programme for Research on Information Technology - ESPRIT and it stems from the identified need of providing simpler and more integrated environments for application development in large distributed systems.

Abstract This article analyses the structure of the distributed operating system of the SMD project— DIOS. The description of the basic structure is followed by a global view of the services and the supporting kernel of the system. Each service is analysed according to its functionality and implementation. DIOS was implemented on bare machines and on top of commercial operating systems. The port of the system to hetereogeneous environments is discussed and some implementation details are described. In parallel with the research on DIOS a multiprocessor station has been developed. These two projects had clear influences on each other; to illustrate some of the major interdependencies, the performance of DIOS in their particular architecture is analysed. The paper ends with information on the current status of the project and on some planned directions of work.