CHOReOS will implement a framework for scalable choreography development. The goal is to enable domain experts to develop decentralized ultra-large scale (ULS) solutions composed of heterogeneous services that are adaptable and QoS (Quality-of-Service) aware. Prior to this, these solutions were only possible with the support of dedicated IT professionals to provide the skills needed for architectural design and software engineering.
CHOReOS will deliver formally grounded abstractions and models, dynamic choreography-centric development processes, governance and service-oriented middleware manipulated via an Integrated Development Runtime Environment (IDRE) aimed at overcoming the ULS impact on software system development.
Here is what to expect from the CHOReOS project:
- Abstractions and models
- A choreography-based development environment
- Service-oriented middleware
- Governance mechanisms
- The Future Internet Ultra-Large Scale (ULS) on any imaginable dimension: This greatly challenges the scalability of the engineering of services, regarding run-time aspects (e.g., scaling to the load, the wide distribution, the heterogeneity and/or the dynamics of the system) as much as design-time aspects (e.g., scaling to the size of the service base, the continuous evolution of the networking environment and/or the number of services to coordinate). In more detail, within CHOReOS, we concentrate on the following dimensions of the Ultra-Large Scale Internet: very to ultra large number of services to be coordinated, very to ultra large service base, wide distribution, high heterogeneity of the networked systems that range from tiny scale sensors/actuators to infrastructure servers, high dynamics of the networked environment, and very high load.
- Service-oriented software development evolves from a mostly static process to a completely dynamic user-centric one: Traditionally, Internet service development has been seen as a technologycentric and centrally controlled process. By contrast, a significant feature of the ULS Future Internet is that the frontier between service-oriented software development and service usage is blurring. Indeed, the endless openness and dynamics of the Future Internet makes design-time assumptions continuously evolving, especially regarding the context of use and the underlying technologies. Moreover, the development of services can no longer be assumed to be handled by IT-experts only. Endusers should be able to play an active role in the overall development of services, so that changes in requirements as much as in context of use can be adequately and agilely handled.