Elena Gómez-Martínez

ArgoSPE: model-based software performance engineering

Stochastic Petri nets (SPNs) have been proved useful for the quantitative analysis of systems. This paper introduces ArgoSPE, a tool for the performance evaluation of software systems in the first stages of the life-cycle. ArgoSPE implements a performance evaluation process that builds on the principles of the software performance engineering (SPE). The theory behind the tool, i.e. the underlying SPE process, has been presented in previous papers and consists in translating some performance annotated UML diagrams into SPN models. Therefore, ArgoSPE prevents software engineers to model with SPN since they are obtained as a by-product of their UML models. The design of the tool follows the architecture proposed by OMG in the UML Profile for Schedulability, Performance and Time specification.

Impact of SOAP implementations in the performance of a web service-based application

This article recalls, from the literature, a performance study of a web service. That study, based on the layered queuing network (LQN) paradigm, is now addressed following the PUMA approach to obtain a new performance model, in this case in terms of Petri nets, for the target web service. Such Petri net model is used to extend the previous LQN results with respect to some key web service performance aspects: the SOAP toolkit and the XML parsers. Actually, this paper aims to explore through a case study some of the main concerns of web services performance at the middleware layer. The acquired background is meant to start to develop a methodology, based on the SPE principles, useful to analyze web services performance.

A software performance engineering tool based on the UML-SPT

Software performance engineering by C. U. Smith. (1990) (SPE) proposes methods to evaluate performance of software systems early in the development process. A SPE accepted approach consists in deriving performance models from UML specifications, usually annotated according to the OMG profile for schedulability, performance and time specification. We introduce here a new SPE tool that fits in the OMG framework and implements most of the features given by J.P.Lopez-Grao et al. (2004). The tool allows to design UML diagrams annotated according to the UML-SPT, and automatically generates a performance model in terms of generalized stochastic Petri nets (GSPN), using the GreatSPN file format. The input of the tool constitutes a software model designed as a set of UML state machines, whose activities can be modelled using UML activity diagrams. The class diagram specifies system population and the deployment diagram models some inter-nodes characteristics (i.e. network speed).

Performance analysis of mobile agents tracking

Mobile agents have arisen as an interesting paradigm to build distributed applications, due to the unparalleled advantages they offer. However, along with the advantages they also present new challenges. One of the most relevant is that it is not easy to ensure efficient communication among agents that move continually from one computer to another.In this paper, we apply SPE techniques to model and analyze, how a mobile agent tracking approach addresses the highly dynamic movement problem in a distributed computing environment.

Performance assessment of an architecture with adaptative interfaces for people with special needs

People in industrial societies carry more and more portable electronic devices (e.g., smartphone or console) with some kind of wireless connectivity support. Interaction with auto-discovered target devices present in the environment (e.g., the air conditioning of a hotel) is not so easy since devices may provide inaccessible user interfaces (e.g., in a foreign language that the user cannot understand). Scalability for multiple concurrent users and response times are still problems in this domain. In this paper, we assess an interoperable architecture, which enables interaction between people with some kind of special need and their environment. The assessment, based on performance patterns and antipatterns, tries to detect performance issues and also tries to enhance the architecture design for improving system performance. As a result of the assessment, the initial design changed substantially. We refactorized the design according to the Fast Path pattern and The Ramp antipattern. Moreover, resources were correctly allocated. Finally, the required response time was fulfilled in all system scenarios. For a specific scenario, response time was reduced from 60 seconds to less than 6 seconds.

Model-Based Verification of Safety Contracts

The verification of safety becomes crucial in critical systems where human lives depend on the correct functioning of such systems. Formal methods have often been advocated as necessary to ensure the reliability of software systems, albeit with a considerable effort. In any case, such an effort is cost-effective when verifying safety-critical systems. Safety requirements are usually expressed using safety contracts, in terms of assumptions and guarantees. To facilitate the adoption of formal methods in the safety-critical software industry, we propose the use of well-known modelling languages, such as UML, to model a software system, and the use of OCL to express the system safety contracts within UML. A UML model enriched with OCL constraints is then transformed to a Petri net model that enables to formally verify such safety contracts. We apply our approach to an industrial case study that models a train doors controller in charge of the opening and closing of train doors. Our approach allows to perform an early safety verification, which increases the confidence of software engineers while designing the system.

Performance modeling and analysis of the universal control hub

People with special needs may find difficulties using electronic consumer devices, user interfaces limit their chances of having full control on them. The Universal Remote Control (URC) is an ISO standard that promotes pluggable and interoperable user interfaces to remotely operate electronic devices. The Universal Control Hub (UCH) is the software architecture that materialises URC, and several implementations are currently available. However, users and developers wonder about UCH feasibility to respond to future needs regarding performance. In this paper, we conduct a study to analyze whether UCH can face multiple concurrent users. Serious problems are exposed at this regard in this paper, they may contribute to question a solution that initially and from the interoperability point of view was very-well suited.

Deriving Safety Case Fragments for Assessing MBASafe’s Compliance with EN 50128

According to EN 50129, manufacturers of rail vehicles shall justify via a safety case that their vehicles are adequately safe for their intended applications. MBASafe is a recently proposed and potentially innovative design and verification process. In the presence of compelling arguments concerning its adequacy as process evidence, MBASafe could support the safety claims within the required safety cases. In this paper, we contribute to partially justify the adequacy of MBASafe to act as process evidence. To do that, we first manually check if MBASafe includes EN 50128-compliant process elements, then we model MBASafe in compliance with Software Process Engineering Meta-model 2.0, then, we derive process-based arguments from the MBASafe process model by using MDSafeCer, the recently introduced Model Driven Safety Certification method. By doing so, we provide a twofold contribution: we further validate MDSafeCer in the rail domain and we strengthen MBASafe.

Model-Based Safety Assessment Using OCL and Petri Nets

Safety becomes a primordial assessment in safety-related systems where human lives can be somehow put in risk, needing to comply with safety requirements defined by industry standards such as IEC 61508, ISO 26262 or DO-178C. Safety contracts are useful to specify these requirements (as assumptions and guarantees), thus assuring an expected level of confidence. To verify the safety requirements is measured to represent more than a half of the overall system development costs. In this paper, we propose a model-based verification that addresses safety verification from the early beginning of system development, thus saving costs. Namely, we use UML for system design and Object Constraint Language (OCL) for specifying safety contracts, while its verification is carried out using Petri nets. As case study, we assess the safety of an embedded system that models a fire prevention system in a hospital building.

A semantic approach for designing Assistive Software Recommender systems

Assistive Software Recommender (ASR) systems in the context of interoperability architectures.Requirements and design of ASR systems.Implementation and deployment of an ASR system.We report the non-functional assessment of our ASR system: performance and scalability evaluation. Assistive Software offers a solution for people with disabilities to manage specialized hardware, devices or services. However, these users may have difficulties in selecting and installing Assistive Software in their devices for managing smart environments. This paper addresses the requirements of these kinds of systems and their design in the context of interoperability architectures. Our solution follows a semantic approach, for which ontologies are a key. The paper also presents an implementation of our design proposal, i.e., a real and usable system which is evaluated according to a set of functional and non-functional requirements here proposed.