A Bayesian Approach Based on Bayes Minimum Risk Decision for Reliability Assessment of Web Service Composition
Abstract
:1. Introduction
2. Materials and Methods
2.1. Web Service Composition Reliability Model
2.2. Reliability Assessment Method Based on Bayes Minimum Risk Decision
2.2.1. Reliability Assessment Scenario
2.2.2. Bayes Decision
2.2.3. Bayes Prior Hyperparameter Solving Method Based on MMR
3. Results and Discussion
3.1. Reliability Assessment of Web Service Composition
3.2. Goodness-of-Fit of the Distribution
4. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
Abbreviations
MMR | Method of minimum risk; |
BMCM | Bayesian Monte Carlo method; |
ESM | Expert scoring method; |
SOA | Service-oriented architecture; |
QoS | Quality of service; |
AWS | Amazon web service; |
WSDL | Web service description language; |
Tc | Traffic control; |
BMCM | Bayesian Monte Carlo method; |
ESM | Expert scoring method; |
Abstract web service candidates; | |
A specific web service; | |
Duration; | |
The numbers of invocations of the web service; | |
Failures of the web service; | |
The probability of failure of the web service ; | |
Reliability of the web service | |
The state transition relations between web services; | |
The number of web services; | |
The number of interactions of web service with web service | |
Service composition flow; | |
matrix; | |
The identity matrix of order n; | |
The no-absorption transition probability matrix of order n; | |
Reliability of web service composition; | |
Parameter of distribution; | |
The prior probability density function; | |
The posterior probability density function; | |
The likelihood function | |
Reliability requirement; | |
The probability of failure of the web service composition; | |
The test result meets reliability requirement; | |
Test result does not meet the reliability requirement; | |
Prior distribution of parameter; | |
Accepted ; | |
Rejected ; | |
The actual failure probability of web service composition being smaller than the reliability requirement; | |
The actual failure probability of the web service composition being larger than the reliability requirement; | |
The custom risk corresponding to cases in which the software is accepted despite the failure probability being larger than the reliability requirement; | |
The producer risk corresponding to cases in which the software is rejected despite the failure probability being smaller than the reliability requirement; | |
The risk function of the decision function; | |
Prior probability density function; | |
The total number of test cases; | |
Hyperparameters; | |
Expectation of ESM; | |
Variance of ESM; | |
Marginal distribution of the prior distribution | |
Confidence factor. |
References
- Lee, C.H.; Kim, Y.-H.; Park, D.H. Evaluation of multi-tasking software system performance with consideration of module dependency. J. Softw. Maint. Evol. Res. Pr. 2011, 23, 361–374. [Google Scholar] [CrossRef]
- Straub, D.; Papaioannou, I. Bayesian updating with structural reliability methods. J. Eng. Mech. 2015, 141, 04014134. [Google Scholar] [CrossRef] [Green Version]
- Rausand, M.; Barros, A.; Hoyland, A. System Reliability Theory: Models, Statistical Methods, and Applications; John Wiley & Sons: Hoboken, NJ, USA, 2020. [Google Scholar]
- Sinharay, S.; Stern, H.S. On the sensitivity of Bayes factors to the prior distributions. Am. Stat. 2002, 56, 196–201. [Google Scholar] [CrossRef]
- Li, M.; Meeker, W.Q. Application of Bayesian methods in reliability data analyses. J. Qual. Technol. 2014, 46, 1–23.MLA. [Google Scholar] [CrossRef]
- Anderson-Cook, C.M. Opportunities and issues in multiple data type meta-analyses. Qual. Eng. 2009, 21, 241253. [Google Scholar] [CrossRef]
- Csenki, A. Bayes predictive analysis of a fundamental software reliability model. IEEE Trans. Reliab. 2002, 39, 177–183. [Google Scholar] [CrossRef]
- Moala, F.A.; Rodrigues, J.; Tomazella, V.L.D. A note on the prior distributions of weibull parameters for the reliability function. Commun. Stat. Theory Methods 2009, 38, 1041–1054. [Google Scholar] [CrossRef]
- Burgman, M.; McBride, M.; Ashton, R.; Speirs-Bridge, A.; Flander, L.; Wintle, B.; Fidler, F.; Rumpff, L.; Twardy, C. Expert status and performance. PLoS ONE 2011, 6, e22998. [Google Scholar] [CrossRef] [PubMed]
- Hgan, A.O.; Buck, C.; Daneshkhah, A.; Eiser, J.; Garthwaite, P.; Jenkinson, D.; Oakley, J.; Rakow, T. Uncertain Judgements: Eliciting Experts Probabilities; John Wiley & Sons: Hoboken, NJ, USA, 2006. [Google Scholar]
- McConway, K. The Combination of Experts Opinions in Probability Assessment: Some Theoretical Considerations. Ph.D. Thesis, University College London, London, UK, 1978. [Google Scholar]
- Li, Z.S.; Guo, J.; Xiao, N.-C.; Huang, W. Multiple priors integration for reliability estimation using the Bayesian melding method. In Proceedings of the 2017 Annual Reliability and Maintainability Symposium (RAMS), Orlando, FL, USA, 23–26 January 2017; pp. 1–6. [Google Scholar]
- Pham, C.; Tran, N.H.; Ren, S.; Saad, W.; Hong, C.S. Traffic-Aware and Energy-Efficient vNF Placement for Service Chaining: Joint Sampling and Matching Approach. IEEE Trans. Serv. Comput. 2020, 13, 172–185. [Google Scholar] [CrossRef]
- Di Mauro, M.; Longo, M.; Postiglione, F.; Tambasco, M. Availability modeling and evaluation of a network service deployed via NFV. In Communications in Computer and Information Science; Springer: Berlin/Heidelberg, Germany, 2017; Volume 766, pp. 31–44. [Google Scholar]
- Bian, S.; Huang, X.; Shao, Z.; Gao, X.; Yang, Y. Service Chain Composition with Failures in NFV Systems: A Game-Theoretic Perspective. In Proceedings of the ICC 2019—2019 IEEE International Conference on Communications (ICC), Shanghai, China, 20–24 May 2019; pp. 1–6. [Google Scholar] [CrossRef]
- di Mauro, M.; Longo, M.; Postiglione, F. Availability Evaluation of Multi-tenant Service Function Chaining Infrastructures by Multidimensional Universal Generating Function. IEEE Trans. Serv. Comput. 2018, 1. [Google Scholar] [CrossRef]
- Fan, T.H.; Balakrishnan, N.; Chang, C. The Bayesian approach for highly reliable electro-explosive devices using one-shot device testing. J. Stat. Comput. Simul. 2009, 79, 1143–1154. [Google Scholar] [CrossRef]
- Kuelbs, J.; Philipp, W. Almost sure invariance principles for partial sums of mixing B-valued random variables. Ann. Probab. 1980, 8, 1003–1036. [Google Scholar] [CrossRef]
- Dai, Y.-S.; Xie, M.; Long, Q.; Ng, S.-H. Uncertainty analysis in software reliability modeling by bayesian analysis with maximum-entropy principle. IEEE Trans. Softw. Eng. 2007, 33, 781–795. [Google Scholar] [CrossRef]
- Guan, Y.; Dy, J. Sparse probabilistic principal component analysis. Artif. Intell. Stat. 2009, 5, 185–192. [Google Scholar]
- Bergman, S.; Schiffer, M. Kernel Functions and Elliptic Differential Equations in Mathematical Physics; Courier Corporation: Chelmsford, MA, USA, 2005. [Google Scholar]
- Singpurwalla, N.D. Reliability and Risk: A Bayesian Perspective; John Wiley & Sons: Hoboken, NJ, USA, 2006. [Google Scholar]
- Sebald, D.J.; Bucklew, J.A. Support vector machines and the multiple hypothesis test problem. IEEE Trans. Signal Process. 2001, 49, 2865–2872. [Google Scholar] [CrossRef]
- Rothermel, G.; Untch, R.H.; Chu, C.; Harrold, M.J. Prioritizing Test Cases for Regression Testing. 2001. Available online: https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1017&context=csearticles (accessed on 3 December 2020).
- Lindley, D.V. The use of prior probability distributions in statistical inference and decision. In Proceedings of the 4th Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, CA, USA, 20–30 July 1960; pp. 453–468. [Google Scholar]
- Sajee, M.; Varia, J. Overview of Amazon Web Services. 2014. Available online: https://docs.aws.amazon.com/whitepapers/latest/aws-overview/aws-overview.pdf (accessed on 3 December 2020).
- Wikipedia. LoadUI. Available online: https://en.wikipedia.org/wiki/LoadUI (accessed on 3 December 2020).
- Dai, Z.; Wang, Z.; Jiao, Y. Bayes Monte-Carlo assessment method of protection systems reliability based on small failure sample data. IEEE Trans. Power Deliv. 2014, 29, 1841–1848. [Google Scholar] [CrossRef]
- Walls, L.; Quigley, J. Building prior distributions to support Bayesian reliability growth modelling using expert judgement. Reliab. Eng. Syst. Saf. 2001, 74, 117–128. [Google Scholar] [CrossRef]
- Hamada, M.; Martz, H.F.; Reese, C.S.; Graves, T.; Johnson, V.; Wilson, A.G. A fully bayesian approach for combining multilevel failure information in fault tree quantification and optimal follow-on resource allocation. Reliab. Eng. Syst. Saf. 2004, 86, 297–305. [Google Scholar] [CrossRef]
- Christensen, R.; Johnson, W.; Branscum, A.; Hanson, T.E. Bayesian Ideas and Data Analysis: An Introduction for Scientists and Statisticians; CRC Press: Boca Raton, FL, USA, 2011. [Google Scholar]
- Wang, H.; Ouni, A.; Kessentini, M.; Maxim, B.; Grosky, W.I. Identification of web service refactoring opportunities as a multi-objective problem. In Proceedings of the 2016 IEEE International Conference on Web Services (ICWS), San Francisco, CA, USA, 27 June–2 July 2016; pp. 586–593. [Google Scholar]
Authors | Year | Approaches |
---|---|---|
Anderson-Cook [6] | 2009 | By considering the totality of data simultaneously, instead of performing analyses on each data type separately, for decision-making; however, it does not apply the service composition. |
Csenki, A [7] | 2009 | The authors studied an averaging method that combines the opinions of different experts. However, it does not adequately explain the deviations of different opinions and require a more complex model such as the Bayesian model |
Moala F A, Rodrigues J, Tomazella V L D [8] | 2009 | The authors compare the posterior densities of the reliability function by using several examples. |
M. Burgman, M. McBride, R. Ashton [9] | 2011 | The study proposed a method of multiple prior information integration is to average all multiple priors. However, the deviations of the different opinions are not properly quantified in the averaging approach. |
A. OHgan, C. Buck, A. Daneshkhah [10] | 2006 | Linear and geometric pooling methods allow unequal weights for prior information and focus on multiple priors; however, it over depends the experts’ judgement. |
K. McConway [11] | 1978 | The study considered system configuration and structure; the system prior can be derived from subsystem priors. |
Li Z S, Guo J, Xiao N C, et al. [12] | 2017 | The study proposed an integration-based method to estimate and pool the weighted priors; it utilizes Bayes’ theorem to integrate heterogeneous priors. |
Pham, N. H. Tran, S. Ren, W. Saad and C. S. Hong [13] | 2020 | A sampling-based Markov approximation (MA) approach is proposed to solve the combinatorial NP-hard problem, to overcome this issue of requires a long convergence time. |
Di Mauro, M. Longo, F. Postiglione, and M. Tambasco [14] | 2017 | For a chain of network nodes in Service Function Chains, a double-layer model is adopted, where Reliability Block Diagram describes the high-level dependencies among the architecture components, availability analysis is carried out to characterize the minimal configuration of the overall system. |
Bian, X. Huang, Z. Shao, X. Gao and Y. Yang [15] | 2015 | The authors proposed the DISCCA algorithm that guides the service nodes towards the Nash Equilibrium with short latency and low congestion, through decision making by individual users with local information to improve the reliability of system. |
M. Di Mauro, M. Longo, and F. Postiglione [16] | 2018 | For Service Function Chaining, propose a Universal Generating Function (UGF) approach, which minimizes deployment cost while respecting a given availability requirement. |
Web Service | Failure Probability |
---|---|
0.081 | |
0.024 | |
0.096 | |
0.038 |
Number of Failures | Number of Test Cases | ||
---|---|---|---|
BMCM | ESM | MMR | |
0 | 4602 | 3932 | 3692 |
1 | 7635 | 5478 | 4573 |
2 | 9402 | 7123 | 6432 |
3 | 10,041 | 8232 | 8923 |
4 | 11,900 | 10,345 | 10,401 |
5 | 16,104 | 13,335 | 11,452 |
Stage in Growth Testing | Expert 1 | Expert 2 | Expert 3 | Expert 4 | Beta Distribution | Hyperparameter | ||
---|---|---|---|---|---|---|---|---|
Expected | Variance | b | ||||||
1 | (0.003, 0.0072) | (0.0035, 0.0075) | (0.0040, 0.0074) | (0.0038, 0.0072) | 0.0054 | 1.17 × 10−6 | 17.58 | 140.41 |
2 | (0.0080, 0.0088) | (0.0060, 0.0086) | (0.0065, 0.0090) | (0.0063, 0.0085) | 0.0077 | 3.42 × 10−6 | 21.22 | 270.78 |
3 | (0.0075, 0.0095) | (0.0077, 0.0093) | (0.0076, 0.0092) | (0.0078, 0.95) | 0.0085 | 2.48 × 10−6 | 33.27 | 311.92 |
Web Services | Failure Probability |
---|---|
0.011 | |
0.034 | |
0.096 | |
0.018 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Song, Y.; Wang, Y.; Jin, D. A Bayesian Approach Based on Bayes Minimum Risk Decision for Reliability Assessment of Web Service Composition. Future Internet 2020, 12, 221. https://doi.org/10.3390/fi12120221
Song Y, Wang Y, Jin D. A Bayesian Approach Based on Bayes Minimum Risk Decision for Reliability Assessment of Web Service Composition. Future Internet. 2020; 12(12):221. https://doi.org/10.3390/fi12120221
Chicago/Turabian StyleSong, Yang, Yawen Wang, and Dahai Jin. 2020. "A Bayesian Approach Based on Bayes Minimum Risk Decision for Reliability Assessment of Web Service Composition" Future Internet 12, no. 12: 221. https://doi.org/10.3390/fi12120221