Using quantitative analysis to implement autonomic IT systems

Călinescu, Radu; Kwiatkowska, Marta

doi:10.1109/icse.2009.5070512

Cited by 75 publications

(71 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second category consists of approaches that consider the behavior of the system and its environment modeled in a monolithic way in terms of more powerful models defined at a lower level of abstraction [2,1,9]. For example, in the approach presented in [2], DTMCs are used to model, for each system configuration, the future state of a system and its environment if that configuration is used.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Impact Models for Architecture-Based Self-adaptive Systems

Moreno

Lopes

Garlan

et al. 2015

Formal Aspects of Component Software

View full text Add to dashboard Cite

Abstract. Self-adaptive systems have the ability to adapt their behavior to dynamic operation conditions. In reaction to changes in the environment, these systems determine the appropriate corrective actions based in part on information about which action will have the best impact on the system. Existing models used to describe the impact of adaptations are either unable to capture the underlying uncertainty and variability of such dynamic environments, or are not compositional and described at a level of abstraction too low to scale in terms of specification effort required for non-trivial systems. In this paper, we address these shortcomings by describing an approach to the specification of impact models based on architectural system descriptions, which at the same time allows us to represent both variability and uncertainty in the outcome of adaptations, hence improving the selection of the best corrective action. The core of our approach is an impact model language equipped with a formal semantics defined in terms of Discrete Time Markov Chains. To validate our approach, we show how employing our language can improve the accuracy of predictions used for decisionmaking in the Rainbow framework for architecture-based self-adaptation.

show abstract

Section: Related Workmentioning

confidence: 99%

“…For example, in the approach presented in [2], DTMCs are used to model, for each system configuration, the future state of a system and its environment if that configuration is used. These models are expressive enough to model variability and the uncertainty underlying adaptation outcomes.…”

Section: Related Workmentioning

confidence: 99%

Impact Models for Architecture-Based Self-adaptive Systems

Moreno

Lopes

Garlan

et al. 2015

Formal Aspects of Component Software

View full text Add to dashboard Cite

show abstract

“…In [26] a model checker executed from the current local state has been used to predict and prevent future inconsistencies in a distributed system. In the quantitative runtime setting, a number of approaches have been proposed for different types of models, to mention the autonomic approach of [2,1] for discrete-and continuous-time Markov chains, parametric techniques of [15,16] for discrete time Markov chains and the incremental approach of [21] for Markov decision processes. Partially observable Markov decision processes are known to be infeasible, but a promising partial approach to adversary generation was recently proposed in [18].…”

Section: Introductionmentioning

confidence: 99%

“…The main culprit is state-space explosion in conjunction with the inherent complexity of the analysis methods that are involved. A quantitative runtime verification approach was recently proposed [2,16,1] as an alternative, complementary analysis method. We adopt this approach, and focus on the following system characteristics:…”

Section: Introductionmentioning

confidence: 99%

Towards Communication-Based Steering of Complex Distributed Systems

Dräger

Kwiatkowska

2012

Large-Scale Complex IT Systems. Development, Operation and Management

Self Cite

View full text Add to dashboard Cite

Abstract. Quantitative verification is an established automated technique that can ensure predictability and dependability of software systems which exhibit probabilistic behaviour. Since offline usage of quantitative verification is infeasible for large-scale complex systems that continuously adapt to the changing environment, quantitative runtime verification was proposed as an alternative. Using an illustrative case study of communicating, distributed probabilistic processes, we formulate the problem of quantitative steering, a runtime technique that involves system monitoring, prediction of future errors, and enforcement of system's behaviour away from the error states. We consider a communicationbased variant of steering where enforcement is achieved by modifying the contents of communication channels. Our approach is based on stochastic games, where one player is the system and the other players assume the role of the controller, and hence steering reduces to finding a controller strategy that meets the given quantitative goal. We discuss the solution to the quantitative steering problem and its extensions inspired by complex real-world scenarios.

show abstract

“…Many of these AI paradigms fall into the category of self-* or autonomous systems. These are Self-Managed systems that are capable of self-configuration, self-adaptation, self-healing, self-monitoring and self-tuning [3].…”

Section: Introductionmentioning

confidence: 99%

Establishing a Framework for Dynamic Risk Management in ‘Intelligent’ Aero-Engine Control

Kurd

Kelly

McDermid

et al. 2009

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. The behaviour of control functions in safety critical software systems is typically bounded to prevent the occurrence of known system level hazards. These bounds are typically derived through safety analyses and can be implemented through the use of necessary design features. However, the unpredictability of real world problems can result in changes in the operating context that may invalidate the behavioural bounds themselves, for example, unexpected hazardous operating contexts as a result of failures or degradation. For highly complex problems it may be infeasible to determine the precise desired behavioural bounds of a function that addresses or minimises risk for hazardous operation cases prior to deployment. This paper presents an overview of the safety challenges associated with such a problem and how such problems might be addressed. A self-management framework is proposed that performs on-line risk management. The features of the framework are shown in context of employing intelligent adaptive controllers operating within complex and highly dynamic problem domains such as Gas-Turbine Aero Engine control. Safety assurance arguments enabled by the framework necessary for certification are also outlined.

show abstract

Using quantitative analysis to implement autonomic IT systems

Cited by 75 publications

References 23 publications

Impact Models for Architecture-Based Self-adaptive Systems

Impact Models for Architecture-Based Self-adaptive Systems

Towards Communication-Based Steering of Complex Distributed Systems

Establishing a Framework for Dynamic Risk Management in ‘Intelligent’ Aero-Engine Control

Contact Info

Product

Resources

About