Luis Ferreira Pires

A Framework for Dynamic Web Services Composition

Dynamic composition of web services is a promising approach and at the same time a challenging research area for the dissemination of service-oriented applications. It is widely recognised that service semantics is a key element for the dynamic composition of Web services, since it allows the unambiguous descriptions of a service’s capabilities and parameters. This paper introduces a framework for performing dynamic service composition by exploiting the semantic matchmaking between service parameters (i.e., outputs and inputs) to enable their interconnection and interaction. The basic assumption of the framework is that matchmaking enables finding semantic compatibilities among independently defined service descriptions. We also developed a composition algorithm that follows a semantic graph-based approach, in which a graph represents service compositions and the nodes of this graph represent semantic connections between services. Moreover, functional and non-functional properties of services are considered, to enable the computation of relevant and most suitable service compositions for some service request. The suggested end-to-end functional level service composition framework is illustrated with a realistic application scenario from the IST SPICE project.

On the notion of abstract platform in MDA development

Although platform-independence is a central property in MDA models, the study of platform-independence has been largely overlooked in MDA. As a consequence, there is a lack of guidelines to select abstraction criteria and modelling concepts for platform-independent design. In addition, there is little methodological support to distinguish between platform-independent and platform-specific concerns, which could be detrimental to the beneficial exploitation of the PIM-PSM separation-of-concerns adopted by MDA. This work is an attempt towards clarifying the notion of platform-independent modelling in MDA development. We argue that each level of platform-independence must be accompanied by the identification of an abstract platform. An abstract platform is determined by the platform characteristics that are relevant for applications at a certain level of platform-independence, and must be established by balancing various design goals. We present some methodological principles for abstract platform design, which forms a basis for defining requirements for design languages intended to support platform-independent design. Since our methodological framework is based on the notion of abstract platform, we pay particular attention to the definition of abstract platforms and the language requirements to specify abstract platforms. We discuss how the concept of abstract platform relates to UML.

A rigorous approach to relate enterprise and computational viewpoints

Multiviewpoint approaches allow stakeholders to design a system from stakeholder-specific viewpoints. By this, a separation of concerns is achieved, which makes designs more manageable. However, to construct a consistent multiviewpoint design, the relations between viewpoints must be defined precisely, so that the consistency of designs from these viewpoints can be verified. The goal of this paper is to make the consistency rules between (a slightly adapted version of) the RM-ODP enterprise and computational viewpoints more precise and to make checking the consistency between these viewpoints practically applicable. To achieve this goal, we apply a generic framework for relating viewpoints that includes reusable consistency rules. We implemented the consistency rules in a tool to show their applicability.

Towards runtime discovery, selection and composition of semantic services

Service-orientation is gaining momentum in distributed software applications, mainly because it facilitates interoperability and allows application designers to abstract from underlying implementation technologies. Service composition has been acknowledged as a promising approach to create composite services that are capable of supporting service user needs, possibly by personalising the service delivery through the use of context information or user preferences. In this paper we discuss the challenges of automatic service composition, and present DynamiCoS, which is a novel framework that aims at supporting service composition on-demand and at runtime for the benefit of service end-users. We define the DynamiCoS framework based on a service composition life-cycle. Framework mechanisms are introduced to tackle each of the phases and requirements of this life-cycle. Semantic services are used in our framework to enable reasoning on the service requests issued by end-users, making it possible to automate service discovery, selection and composition. We validate our framework with a prototype that we have built in order to experiment with the mechanisms we have designed. The prototype was evaluated in a testing environment using some use case scenarios. The results of our evaluation give evidences of the feasibility of our approach to support runtime service composition. We also show the benefits of semantic-based frameworks for service composition, particularly for end-users who will be able to have more control on the service composition process.

Situations in Conceptual Modeling of Context

In previous work, we have defined conceptual foundations that can be beneficially used in context modeling. These conceptual foundations include the separation of entity and context, and the characterization of context as either Intrinsic or Relational. This paper aims at extending this approach by introducing the ontological concept of Situation as means of composing the elements of our ontology (entities, intrinsic and relational contexts) to model particular states of affairs of interest. Our concepts have been inspired by and aligned with conceptual theories from the fields of philosophy and cognitive sciences.

On the role of basic design concepts in behaviour structuring

This paper presents some basic design concepts for the design of open distributed systems. These concepts should form the basis for the development of effective design methodologies. The paper discusses how design concepts, such as interaction, action and causality relation, can be used for modelling and structuring behaviours of functional entities in a distributed environment. The paper also addresses some consequences of the application of these design concepts such as the choice of language elements and operations to represent behaviour structure, the structuring of the design process, and the definition of design operations for behaviour refinement.

Performance in model transformations: experiments with ATL and QVT

Model transformations are increasingly being incorporated in software development processes. However, as systems being developed with transformations grow in size and complexity, the performance of the transformations tends to degrade. In this paper we investigate the factors that have an impact on the execution performance of model transformations. We analyze the performance of three model transformation language engines, namely ATL, QVT Operational Mappings and QVT Relations. We implemented solutions to two transformation problems in these languages and compared the performance of these transformations. We extracted metric values from the transformations to systematically analyze how their characteristics influence transformation execution performance. We also implemented a solution to a transformation problem in ATL in three functionally equivalent ways, but with different language constructs to evaluate the effect of language constructs on transformation performance. The results of this paper enable a transformation designer to estimate beforehand the performance of a transformation, and to choose among implementation alternatives to achieve the best performance. In addition, transformation engine developers may find some of our results useful in order to tune their tools for better performance.

An ontology-based approach for evaluating the domain appropriateness and comprehensibility appropriateness of modeling languages

In this paper we present a framework for the evaluation and (re)design of modeling languages. We focus here on the evaluation of the suitability of a language to model a set of real-world phenomena in a given domain. In our approach, this property can be systematically evaluated by comparing the level of homomorphism between a concrete representation of the worldview underlying the language (captured in a metamodel of the language), with an explicit and formal representation of a conceptualization of that domain (a reference ontology). The framework proposed comprises a number of properties that must be reinforced for an isomorphism to take place between these two entities. In order to illustrate the approach proposed, we evaluate and extend a fragment of the UML static metamodel for the purpose of conceptual modeling, by comparing it with an excerpt of a philosophically and cognitive well-founded reference ontology.

Designing a configurable services platform for mobile context‐aware applications

Context-aware services platforms aim at supporting the handling of contextual information in order to provide better user-tailored services. This paper proposes a novel services platform architecture to support mobile context-aware applications, giving emphasis to the configurability of the platform’s generic functionality. The paper introduces concepts and a language to cope with configurability aspects. The paper also reports on the implementation of a prototype, which implements a Web services-based context-aware services platform that runs on top of 3G networks

Enterprise interoperability with SOA: a survey of service composition approaches

Service-oriented architecture (SOA) claims to facilitate the construction of flexible and loosely coupled business applications, and therefore is seen as an enabling factor for enterprise interoperability. The concept of service, which is central to SOA, is very convenient to address the matching of needs and capabilities in enterprise collaborations. In order to satisfy more demanding needs or to rapidly adapt to changing needs it is possible to perform service composition in order to combine the capabilities provided through several available services. This paper presents a survey on recent approaches for service composition. To perform this study a conceptual framework for service composition is proposed. This framework allows studying how different approaches deal with the service composition life-cycle and provides basic guidelines for their analysis, evaluation and comparison. The proposed framework is used to analyse five representative service composition approaches.