This thesis investigates the deployment modeling in the scope of Progress,a research vision that aims to tackle the increasing complexity of embedded softwaresystems by adopting a software-component approach. The first phase of the Progress deployment process, which is in the focus of this thesis, defines virtualnodes architecture as an abstraction of target platform devices where componentsare allocated. Based on the Progress development process analysis, the thesis identifiesconcerns that need to be addressed by the ProCom component model to supportthe concepts of virtual nodes and allocation, proposes the extension of the ProCommeta-model and the design of allocation in general. The thesis also provides an implementation of a tool support incorporated into the Eclipse application that forms the basis of the Progress IDE. The implementation,whose main goals are to prove the correctness of the ideas and alleviate the deployment in the IDE, integrates rich graphical editors that support the modelingof virtual platform and allocation of components. / Progress

This thesis is focused on the attributes concept of ProCom, a component model developed within The Progress Centre for Predictable Embedded Software Systems. Attributes are pieces of information of various types and levels of abstraction associated with the ProCom entities during the development of a system. Based on the analysis of the development process envisioned by Progress, the requirements for the attributes of ProCom entities are identified, and various alternatives of realizing attributes are analyzed. The chosen solution of highly structured, multi-valued, and extensible attributes is elaborated. The thesis also consists of the design and the prototype implementation of an attribute framework realizing and proving the feasibility of the proposed concepts. The framework addresses the needs of all actors involved in working with attributes throughout the development of a system: It provides an extensible, modular GUI for viewing and editing possibly highly complex information contained in attributes, an interface for the programmatic access to attributes, and well-defined mechanisms for extending the attribute pool by new attributes, new attribute types, and means for their manipulation. The framework is integrated into the main tool supporting the Progress development, the Progress IDE. / Progress

The notion of maximizing software reuse among the family of products has gained considerable attention in the last decade. Lots of research has been done on designing and managing the commonalities and variabilities between the products. However, very few metrics have been developed to assist architects in designing product line architectures. The structure of the product line holds immense importance towards increasing the life span of the product line. Since many of the product line architecture design methodologies follow a component based approach, it seems logical to attempt to adapt the component based metrics to the product line domain. In this thesis, we attempt to derive metrics that quantify the structural quality of product line architecture. / +92-42-5727639

Model-driven and component-based software engineering have been widely recognized as promising paradigms for development of a wide range of systems. Moreover, in the embedded real-time domain, their combination is believed to be helpful in handling the ever-increasing complexity of such systems design.However, in order for these paradigms and their combination to definitely break through at an industrial level for development of embedded real-time systems, both functional and extra-functional properties need to be addressed at each level of abstraction. This research focuses on the preservation of extra-functional properties. More specifically, the aim is to provide support for easing such preservation throughout the entire development process at different abstraction levels.The main outcome of the research work is a round-trip engineering approach aiding the preservation of extra-functional properties by providing code generators, supporting monitoring and analysis of code execution, and then enabling back-propagation of the results to modelling level. In this way, properties that can only be roughly estimated statically are evaluated against runtime values and this consequently allows to optimize the design models for ensuring preservation of analysed extra-functional properties. Moreover, a solution for managing evolution of computational context in which extra-functional properties are defined by means of validity analysis is provided. Such solution introduces a new language for the description of the computational context in which a given property is provided and/or computed by some analysis, enables detection of changes performed to the context description, and analyses the possible impacts on the extra-functional property values based on a precise representation of differences between previous and current version of the model.

Progress

In this thesis we contribute to the Progress IDE, an integrated development enviroment for real-time embedded systems and more precisely to the REMES toolchain, a set of tools to enabling construction and analysis of embedded system behavior models. The contribution aims to facilitate the formal analysis of behavioral models, so that certain extra-functional properties might be verified during early stages of development. Previous work in the field proposes use of the Priced Timed Automata framework for verification of such properties. The thesis outlines the main points where the current toolchain should be extended in order to allow formal analysis of modeled components. Result of the work is a prototype, which minimizes the manual efforts of system designer by model to model transformations and provides seamless integration with existing tools for formal analysis.

In this thesis we contribute to the Progress IDE, an integrated development enviroment for real-time embedded systems and more precisely to the REMES toolchain, a set of tools to enabling construction and analysis of embedded system behavior models. The contribution aims to facilitate the formal analysis of behavioral models, so that certain extra-functional properties might be verified during early stages of development. Previous work in the field proposes use of the Priced Timed Automata framework for verification of such properties. The thesis outlines the main points where the current toolchain should be extended in order to allow formal analysis of modeled components. Result of the work is a prototype, which minimizes the manual efforts of system designer by model to model transformations and provides seamless integration with existing tools for formal analysis.

Component-based software engineering (CBSE) and model-driven development (MDD) are two approaches for handling software development complexity. In essence, while CBSE focuses on the construction of systems from existing software modules called components; MDD promotes the usage of system models which after a series of transformations result with an implementation of the desired system. Even though they are different, MDD and CBSE are not mutually exclusive. However, there has not been any substantial research about what their similarities and differences are and how they can be combined. In this respect, the main goal of this thesis is to summarize the theoretical background of MDD and CBSE, and to propose and apply a systematic method for their comparison. The method takes into account the different effects that these development paradigms have on a wide range of development aspects. The comparison results are then summarized and analyzed. The thesis also enriches the theoretical discussion with a practical case study comparing CBSE and MDD with respect to ProCom, a component model designed for the development of component-based embedded systems in the vehicular-, automation- and telecommunication domains. The aforementioned comparison method is refined and applied for this purpose. The comparison results are again summarized, analyzed and proposals about future work on ProCom are made.

A software component model is the cornerstone of any Component-based Software Development (CBSD) methodology. Such a model defines the modelling elements for constructing software systems. In software system modelling, it is necessary to capture the three elements of a system's behaviour: (i) control (ii) computation and (iii) data. Within a system, computations are performed according to the flow of control or the flow of data, depending on whether computations are control-driven or data-driven. Computations are function evaluations, assignments, etc., which transform data when invoked by control or data flow. Therefore a component model should be able to model control flow, data flow as well as computations. Current component models all model computations, but beside computations tend to model either control flow only or data flow only, but not both. In this thesis, we present a new component model which can model both control flow and data flow. It contains modelling elements that capture control flow and data flow explicitly. Furthermore, the modelling of control flow is separate from that of data flow; this enables the modelling of both control-driven and data-driven computations. The feasibility of the model is shown by means of an implementation of the model, in the form of a prototype tool. The usefulness of the model is then demonstrated for a specific domain, the embedded systems domain, as well as a generic domain. For the embedded systems domain, unlike current models, our model can be used to construct systems that are both control-driven and data-driven. In a generic domain, our model can be used to construct domain models, by constructing control flows and data flows which together define a domain model.

For a long time, the development of component-based systems has been a crucial part of real-time software development required for embedded and automotive domains. However, most of the existing tools used in these fields are not only proprietary, but also expensive and not research-friendly. Open-source tools in this domain are so far quite limited in terms of the features supported, especially, code generation. In this thesis, we demonstrate how we can improve the development of real-time and concurrent systems by the introduction of component-based modelling into Umple, an open-source modelling tool. Our work enables component-based modelling to be performed both textually and visually, as is the case with other Umple features. We introduce a number of major features into Umple. First, we introduce support for real-time C++ code generation. This includes supporting all Umple features, such as class diagrams, associations, state machines, and attributes. In order to achieve this, we also introduce Umple Template Language (Umple-TL), which helps Umple developers to use Umple itself to emit text using easy-to-use constructs, such that the text emitted can be in different target languages such Java and C++. Umple-TL provides additional capabilities relying on Umple being a model-oriented and object-oriented language. Umple-TL has become the technology for all code generation in Umple, not just our real-time C++ generators. Umple-TL also plays a vital role easing writing component descriptions Second, we support concurrency, which is crucial for the underlying architecture of composite structure. We have to avoid relying on any third-party libraries in order to make sure that the code generated will be deployable on embedded devices, which are limited and do not provide a lot of options. The concurrency pattern we follow extends the active object pattern aiming to enhance communication among active objects. Concurrency development in general, even if a programming language used is not real-time, is not easy. Hence, we simplify active object concepts, such as future, promise, and delay, using new Umple keywords. We also add composite structure support to Umple, we believe that our syntax and language constructs are comprehensive, and do not require a wide knowledge of modelling and UML concepts. Additionally, we introduce a novel protocol-free approach that dynamically extracts communication protocols from ports, bindings, and active objects as a way to simplify development, and to lead to concise and optimized code generation. We demonstrate the effectiveness of our work using cases studies, in which we implement Umple models using our new composite structure and concurrency constructs. We show that the amount of code required to specify complex concepts is reduced, and the generated systems are effective.