We present a new method for the accurate analysis of the quality-of-service (QoS) properties of component-based systems. Our method takes as input a QoS property of interest and a high-level continuous-time Markov chain (CTMC) model of the analysed system, and refines this CTMC based on observations of the execution times of the system components. The refined CTMC can then be analysed with existing probabilistic model checkers to accurately predict the value of the QoS property. The paper describes the theoretical foundation underlying this model refinement, the tool we developed to automate it, and two case studies that apply our QoS analysis method to a service-based system implemented using public web services and to an IT support system at a large university, respectively. Our experiments show that traditional CTMC-based QoS analysis can produce highly inaccurate results and may lead to invalid engineering and business decisions. In contrast, our new method reduced QoS analysis errors by 84.4–89.6 percent for the service-based system and by 94.7–97 percent for the IT support system, significantly lowering the risk of such invalid decisions.

/pdf/observation-enhanced-qos-analysis-of-component-based-systems-1n8e99we2h.pdf

Observation-Enhanced QoS Analysis of Component-Based Systems

We present a new method for the accurate analysis of the quality-of-service (QoS) properties of component-based systems. Our method takes as input a QoS property of interest and a high-level continuous-time Markov chain (CTMC) model of the analysed system, and refines this CTMC based on observations of the execution times of the system components. The refined CTMC can then be analysed with existing probabilistic model checkers to accurately predict the value of the QoS property. The paper describes the theoretical foundation underlying this model refinement, the tool we developed to automate it, and two case studies that apply our QoS analysis method to a service-based system implemented using public web services and to an IT support system at a large university, respectively. Our experiments show that traditional CTMC-based QoS analysis can produce highly inaccurate results and may lead to invalid engineering and business decisions. In contrast, our new method reduced QoS analysis errors by 84.4-89.6% for the service-based system and by 94.7-97% for the IT support system, significantly lowering the risk of such invalid decisions.

We introduce a tool-supported method for the automated refinement of continuous-time Markov chains (CTMCs) used to assess quality properties of component-based software. Existing research focuses on improving the efficiency of CTMC analysis and on identifying new applications for this analysis. As such, ensuring that the analysis is accurate by using CTMCs that closely model the behaviour of the analysed software has received relatively little attention. Our new method addresses this gap by refining the high-level CTMC model of a component-based software system based on observations of the execution times of its components. Our refinement method reduced analysis errors by 77–90.3% for a service-based system implemented using six public web services from three different providers, improving the accuracy of the analysis and significantly reducing the risk of invalid software engineering decisions.

/pdf/accurate-analysis-of-quality-properties-of-software-with-4jhv6ui8av.pdf

Accurate Analysis of Quality Properties of Software with Observation-Based Markov Chain Refinement

Approximation problems are of great importance in applied mathematics. The approximation itself consists in changing the initial mathematical model to another one which is more preferable to consider (e.g. from a computational point of view). At most the arising problems are implied by the problem of receiving quantitative estimates of the approximation accuracy.

S. A. Anichkin

Papers

Hypererlang approximation of probability distributions on (0, ∞) and its application