About the DEGAS Project

DEGAS addresses foundational aspects for the design of global applications by enhancing the state of the art in scientific as well as engineering principles.

The main concerns are the specification in UML and the qualitative and quantitative analysis of global applications. We plan to define the key features of global (wireless) applications that should be exposed at an abstract level of specification and analysis. We provide formal relations between the (possibly richer or incomplete) UML models and the process calculi specifications to connect the specification and the verification environment by hiding as much formal details from the designer as possible. The static and dynamic analysis with case-studies should bring to the definition of new linguistic constructs and new models to analyse and reason about performance and security of global systems.

To make effective the results of our research on the formal analysis of global applications we plan to define an interface that hides as much as possible the technical details. Furthermore, we rely on a standard Unified Modelling Language (UML) to ease formal methods into the software production process.
We first study the peculiar features of global applications and then we modify UML to make it suitable to their design. The challanges we approach in this task are the definition of techniques to extract specifications into process calculi from the possibly excessive or incomplete information in the UML description and to reflect the results of the semantic analyses in the UML style to the user.

The scientific objectives of DEGAS address foundational aspects for the design of global applications by enhancing the state of the art in scientific as well as engineering principles. The main topics concern:

(i) the specification in UML of global computing systems, and
(ii) the qualitative and quantitative analysis of such systems.

Under the above assumptions we plan to defie the distinguishing features of global applications that should be exposed at an abstract level of specification and analysis. We will also consider whether the interconnection structure of the application involves wireless links or not. We then define extensions to UML (according to the standard mechanisms) to model global computing systems.

We will provide formal relations between UML models and process calculi specifiations to connect the specification and verification part of our environment by singling out the mechanisms needed to hide formal details from the designer. This task together with the static and dynamic, qualitative and quantitative analysis techniques will help comparing and finding new language abstractions which are adequate to design global applications. The linguistic study joined with some case studies and with the definition of the interaction with UML models will bring forward new techniques to analyse and reason about global applications.

We are mainly concerned with quantitative aspects and security issues because they are key features for global (wireless) applications. Furthermore, security in wireless worlds cannot be studied independently of performance considerations (this is particularly significant for denial of service or cryptographic attacks). Therefore the framework of the project is an ideal arena to mix the two aspects yielding better analysis strategies than the ones available for classical nondeterministic models of systems.

Eventually, the overall scientific activity of the project will enhance the understanding of global computing systems as well as their design principles and analysis techniques.

The technological objectives of DEGAS are mainly concerned with proof-of-concept implementations with two main concerns:

(i) to shed light on the theoretical development and to validate ideas;
(ii) to prove that enhancement of engineering principles is feasible in the setting of global computing through the use of formal methods and standard modelling languages.

Under the above assumptions we plan to design and implement a prototype environment for assisting the designer of global applications from the early phases of software development onwards. The tool helps designers to specify in UML their applications taking into account the peculiar features of global applications determined in the development of the project. Furthermore, the modelling and analysis techniques for establishing security and quantitative properties are supported in a form which is hidden from the designer. In operation, the user interacts with the environment through the UML interface calling for some analysis whose results are reflected back using the UML notation.

More precisely, we plan to implement the following modules:

• an extractor of process calculi representations from UML specifications;
• an interpreter of process calculi descriptions into suitable representations (e.g., transition systems) for the dynamic analysis of performance and security properties;
• a security checker;
• a performance checker;
• a static checker;
• an interpreter of the results of the analysis into UML specifications.

All the modules will be integrated together with a uniform interface to form a single prototype design environment rather than a collection of tools. Eventually we will also produce a process plan for the application of the DEGAS environment to design general wireless global applications.

Since DEGAS is concerned with a design environment for global applications we select some examples of systems that clearly address the issues of globalism on which to concentrate our efforts. We choose wireless telecommunication applications as our focus domain because wireless technology is a booming area and the growing trend should continue for some years. Furthermore, mobile computing is a strategic area in which EU is willing to push its investments to maintain its leading position on the global market. In particular, to assess our development two possible case studies, selected according to the experience acquired in the field by partners MTCI (Motorola Technology Centre Italy) and OMNYS, can be the following.

Pilot service for mobile entertainment. This service is a multi-player game, highly interactive, to be accessed in ASP mode. From the technical point of view, it requires wireless multimedia messaging. For the business point of view, games and entertainment could be, after voice and SMS, the next killer application on the wireless Internet. The multimedia features of the service could be deployed in the realization of multi-player games, which will use the multimedia messaging service (MMS) to exchange images and sounds, which allows many users to act simultaneously on the same environment, as is possible on the Internet. The technology is expected to substantially evolve during the project, with the consolidation of GPRS and the first deployment of UMTS. Accordingly, the messaging features may be defined to follow the growth of the telecommunication networks, with the multimedia capability increasing as much as the technology will allow. The characteristics of the terminals (display size, computing power, MMI, user programmable) will drive the tailoring of the implementation of the game. For example the game will allow sending messages containing text and images in a format that can be easily sent on a GPRS connection (e.g. JPEG, emails, instant messages, WAP). A possible extension will allow enhancing the type of content that can be exchanged thanks to the potentialities of the UMTS technology.

Web based networked service system. This case study aims at designing a networked service system to provide support to micro web-based business that do not have by themselves the capability of developing proprietary solutions for e-Business. The participants within this networked system will be the service providers and the end users. The service providers will generate income through the business potentialities offered by the increased visibility and accessibility. For the end user, the system will offer easy and wide access to a variety of data information and services. The system will be accessible both through typical wired internet connections as well as through mobile devices via de facto standard protocols. The system will present the services offered by the various service providers according to a coherent layout, and will be the unique interface for service access. From the view point of DEGAS, this case study offers both performance and security issues that will help in the requirements elicitation phase as well as in the verification of the environment capabilities to support system design.

A preliminary step in the design of these new wireless applications is to isolate their distinguishing features with respect to standard ones that should be exposed at an abstract level of specification and analysis. In other words, should a development environment for distributed applications vary according to the characteristics of the communication infrastructure? A clean and precise answer to this question is a significative outcome of DEGAS.