Dynamic QoS Management and Optimization in Service-Based Systems

Călinescu, Radu; Grunske, Lars; Kwiatkowska, Marta; Mirandola, Raffaela; Tamburrelli, Giordano

doi:10.1109/tse.2010.92

Cited by 320 publications

(283 citation statements)

References 81 publications

Supporting

Mentioning

275

Contrasting

Unclassified

Order By: Relevance

“…the taxonomy of uncertainty in [13]), type of adaptation required (cf. the SBS adaptations from, e.g., [2]- [4], [8], [10]), and type(s) of requirements that these adaptations aim to meet. Within these scenarios, we propose the evaluation and comparison of different self-adaptation solutions based on quality attributes and metrics described in [14] and summarised in Table II. …”

Section: Fig 1 Tas Workflowmentioning

confidence: 99%

“…TAS was originally introduced in [2], and has the advantage that it has already been used in the evaluation of several self-adaptation solutions [3], [4], [8], [10], albeit based on ad-hoc implementations, scenarios and evaluation metrics that make the comparison of these solutions and its use to evaluate other solutions very difficult. To address these limitations, we implemented TAS using our new Research Service Platform (ReSeP 1 ), and we propose predefined concrete scenarios for its immediate use in the evaluation of self-adaptation solutions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tele Assistance: A Self-Adaptive Service-Based System Exemplar

Weyns

Călinescu

2015

2015 IEEE/ACM 10th International Symposium on Software Engineering for Adaptive and Self-Managing Systems

Self Cite

View full text Add to dashboard Cite

Abstract-Research on adaptive and self-managing systems is hindered by a lack of prototypical applications that researchers could use to evaluate and compare new methods, techniques and tools. To address this limitation, we introduce a reference implementation of a Tele Assistance System (TAS) for research on self-adaptation in the domain of service-based systems. Our TAS exemplar of service-based systems comes with pre-defined scenarios for comparing the effectiveness of different self-adaptation solutions. Other researchers can easily exploit the underlying service platform, reusable components and development method we devised for TAS to speed up the engineering of additional research exemplars for service-based systems.

show abstract

Section: Fig 1 Tas Workflowmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Tele Assistance: A Self-Adaptive Service-Based System Exemplar

Weyns

Călinescu

2015

2015 IEEE/ACM 10th International Symposium on Software Engineering for Adaptive and Self-Managing Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some approaches have brought state-of-the-art probabilistic model-checkers at runtime [7], providing a suitable infrastructure for many applications. Nonetheless these approaches are not general enough for at least two reasons.…”

Section: Extending Pctl With Rewardsmentioning

confidence: 99%

“…Some components may disappear, or become obsolete, and new ones may be discovered dynamically. This may happen, for example, in the case of Web service-based systems [6,7]. This also happens in pervasive computing scenarios where devices that run application components are mobile [27].…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Verification at Runtime for Self-Adaptive Systems

Filieri

Tamburrelli

2013

Assurances for Self-Adaptive Systems

Self Cite

View full text Add to dashboard Cite

Abstract. An effective design of effective and efficient self-adaptive systems may rely on several existing approaches. Software models and model checking techniques at run time represent one of them since they support automatic reasoning about such changes, detect harmful configurations, and potentially enable appropriate (self-)reactions. However, traditional model checking techniques and tools may not be applied as they are at run time, since they hardly meet the constraints imposed by on-the-fly analysis, in terms of execution time and memory occupation. For this reason, efficient run-time model checking represents a crucial research challenge. This paper precisely addresses this issue and focuses on probabilistic run-time model checking in which reliability models are given in terms of Discrete Time Markov Chains which are verified at run-time against a set of requirements expressed as logical formulae. In particular, the paper discusses the use of probabilistic model checking at run-time for selfadaptive systems by surveying and comparing the existing approaches divided in two categories: state-elimination algorithms and algebra-based algorithms. The discussion is supported by a realistic example and by empirical experiments.

show abstract

“…In [7], we have developed an extensive framework called QoSMOS which can be used to dynamically manage and optimise the performance of service-based systems. The framework has been demonstrated a typical ubiquitous computing healthcare scenario, called TeleAssistance, where patients are remotely monitored, with data being sent to a medical lab for analysis, and there is a requirement to guarantee a certain QoS level of delivering a specific service, for example to change the dosage of a drug.…”

Section: Adaptivenessmentioning

confidence: 99%

Advances in Quantitative Verification for Ubiquitous Computing

Kwiatkowska

2013

Theoretical Aspects of Computing – ICTAC 2013

Self Cite

View full text Add to dashboard Cite

Abstract. Ubiquitous computing, where computers 'disappear' and instead sensor-enabled and software-controlled devices assist us in everyday tasks, has become an established trend. To ensure the safety and reliability of software embedded in these devices, rigorous model-based design methodologies are called for. Quantitative verification, a powerful technique for analysing system models against quantitative properties such as "the probability of a data packet being delivered within 1ms to a nearby Bluetooth device is at least 0.98", has proved useful by detecting and correcting flaws in a number of ubiquitous computing applications. In this paper, we focus on three key aspects of ubiquitous computing: autonomous behaviour, constrained resources and adaptiveness. We summarise recent advances of quantitative verification in relation to these aspects, illustrating each with a case study analysed using the probabilistic model checker PRISM. The paper concludes with an outline of future challenges that remain in this area.

show abstract

Dynamic QoS Management and Optimization in Service-Based Systems

Cited by 320 publications

References 81 publications

Tele Assistance: A Self-Adaptive Service-Based System Exemplar

Tele Assistance: A Self-Adaptive Service-Based System Exemplar

Probabilistic Verification at Runtime for Self-Adaptive Systems

Advances in Quantitative Verification for Ubiquitous Computing

Contact Info

Product

Resources

About