Predição de Falhas em Workflows Científicos em Nuvens baseada em Aprendizado de Máquina
Resumo
Os cientistas cada vez mais têm se apoiado em ferramentas computacionais para executar e analisar experimentos científicos, almejando reduzir esforços e custos para comprovar ou refutar hipóteses. Entretanto, recursos podem ser desperdiçados se os parâmetros usados nas aplicações fizerem com que a execução do experimento falhe. Assim, para diminuir a quantidade de execuções que resultam em falha, este artigo propõe a integração de uma técnica de Aprendizado de Máquina com um Sistema de Gerência de Workflows Científicos para induzir um modelo preditivo de falhas, a partir de dados de proveniência. Resultados experimentais mostram que o modelo é capaz de identificar corretamente casos de falha no Workflow Científico SciPhy.
Referências
Amer, M., Goldstein, M., and Abdennadher, S. (2013). Enhancing one-class support vector machines for unsupervised anomaly detection. Proc. of the ACM SIGKDD Workshop on Outlier Detection and Description, pages 8–15.
Bala, A. and Chana, I. (2015). Intelligent failure prediction models for scientific workflows. Expert Systems with Applications, 42(3):980 – 989.
Coates, A. and Ng, A. Y. (2011). The importance of encoding versus training with sparse coding and vector quantization. In Proc. of the 28th Int. Conf. in Machine Learning, pages 921–928.
Cortes, C. and Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3):273–297.
Câmara, R. V., Paes, A., and de Oliveira, D. (2015). Aplicação de Árvores de Decisão para Recomendação de Parâmetros em Workflows Científicos. Brazilian e-Science Workshop.
de Oliveira, D., Ogasawara, E., Baião, F., and Mattoso, M. (2010). SciCumulus: A Lightweight Cloud Middleware to Explore Many Task Computing Paradigm in Scientific Workflows. Proceedings of the 3rd IEEE Int. Conf. on Cloud Computing.
Freire, J., Koop, D., Santos, E., and Silva, C. T. (2008). Provenance for Computational Tasks: A Survey. Computing in Science & Engineering, pages 20–30.
Gaikwad, P., Mandal, A., Ruth, P., Juve, G., Król, D., and Deelman, E. (2016). Anomaly Detection for Scientific Workflow Applications on Networked Clouds. International Conference on High Performance Computing & Simulation (HPCS).
Han, J., Kamber, M., and Pei, J. (2012). Data Mining: Concepts and Techniques, Third Edition. Elsevier.
Mattoso, M., Werner, C., Travassos, G. H., Braganholo, V., Murta, L., Ogasawara, E., de Oliveira, D., da Cruz, S. M. S., and Martinho, W. (2010). Towards Supporting the Life Cycle of Large-scale Scientific Experiments. Int. Journal of Business Process Integration and Management, pages 79–92.
Mitchell, T. M. (1997). Machine Learning. McGraw-Hill Science/Engineering/Math.
Oca˜na, K., de Oliveira, D., Ogasawara, E., Davila, A., Lima, A., and Mattoso, M. (2011).
SciPhy: A Cloud-based ScientificWorkflow for Phylogenetic Analysis of Drug Targets in Protozoan Species. Brazilian Simposium of Bioinformatics.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., et al. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12(Oct):2825–2830. Refaeilzadeh, P., Tang, L., and Liu, H. (2009). Cross-Validation, pages 532–538. Springer US, Boston, MA.
Samak, T., Gunter, D., Goode, M., Deelman, E., Juve, G., Mehta, G., Silva, F., and Vahi, K. (2011). Online Fault and Anomaly Detection for Large-Scale ScientificWorkflows. IEEE Int. Conf. on High Performance Computing and Communications.
Si, Y.-W., Hoi, K.-K., Biuk-Aghai, R. P., Fong, S., and Zhang, D. (2016). Run-based exception prediction for workflows. Journal of Systems and Software, 113:59 – 75.
Steinwart, I. and Christmann, A. (2008). Support vector machines. Springer Science & Business Media.
Wozniak, J. M., Armstrong, T. G., Wilde, M., Katz, D. S., Lusk, E., and Foster, I. T. (2013). Swift/t: Scalable data flow programming for many-task applications. SIGPLAN Not., 48(8):309–310.
Yeo, P. and Abidi, S. S. (2013). Dataflow oriented similarity matching for scientific workflows. IEEE 27th International, page 2091–2100.
Bala, A. and Chana, I. (2015). Intelligent failure prediction models for scientific workflows. Expert Systems with Applications, 42(3):980 – 989.
Coates, A. and Ng, A. Y. (2011). The importance of encoding versus training with sparse coding and vector quantization. In Proc. of the 28th Int. Conf. in Machine Learning, pages 921–928.
Cortes, C. and Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3):273–297.
Câmara, R. V., Paes, A., and de Oliveira, D. (2015). Aplicação de Árvores de Decisão para Recomendação de Parâmetros em Workflows Científicos. Brazilian e-Science Workshop.
de Oliveira, D., Ogasawara, E., Baião, F., and Mattoso, M. (2010). SciCumulus: A Lightweight Cloud Middleware to Explore Many Task Computing Paradigm in Scientific Workflows. Proceedings of the 3rd IEEE Int. Conf. on Cloud Computing.
Freire, J., Koop, D., Santos, E., and Silva, C. T. (2008). Provenance for Computational Tasks: A Survey. Computing in Science & Engineering, pages 20–30.
Gaikwad, P., Mandal, A., Ruth, P., Juve, G., Król, D., and Deelman, E. (2016). Anomaly Detection for Scientific Workflow Applications on Networked Clouds. International Conference on High Performance Computing & Simulation (HPCS).
Han, J., Kamber, M., and Pei, J. (2012). Data Mining: Concepts and Techniques, Third Edition. Elsevier.
Mattoso, M., Werner, C., Travassos, G. H., Braganholo, V., Murta, L., Ogasawara, E., de Oliveira, D., da Cruz, S. M. S., and Martinho, W. (2010). Towards Supporting the Life Cycle of Large-scale Scientific Experiments. Int. Journal of Business Process Integration and Management, pages 79–92.
Mitchell, T. M. (1997). Machine Learning. McGraw-Hill Science/Engineering/Math.
Oca˜na, K., de Oliveira, D., Ogasawara, E., Davila, A., Lima, A., and Mattoso, M. (2011).
SciPhy: A Cloud-based ScientificWorkflow for Phylogenetic Analysis of Drug Targets in Protozoan Species. Brazilian Simposium of Bioinformatics.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., et al. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12(Oct):2825–2830. Refaeilzadeh, P., Tang, L., and Liu, H. (2009). Cross-Validation, pages 532–538. Springer US, Boston, MA.
Samak, T., Gunter, D., Goode, M., Deelman, E., Juve, G., Mehta, G., Silva, F., and Vahi, K. (2011). Online Fault and Anomaly Detection for Large-Scale ScientificWorkflows. IEEE Int. Conf. on High Performance Computing and Communications.
Si, Y.-W., Hoi, K.-K., Biuk-Aghai, R. P., Fong, S., and Zhang, D. (2016). Run-based exception prediction for workflows. Journal of Systems and Software, 113:59 – 75.
Steinwart, I. and Christmann, A. (2008). Support vector machines. Springer Science & Business Media.
Wozniak, J. M., Armstrong, T. G., Wilde, M., Katz, D. S., Lusk, E., and Foster, I. T. (2013). Swift/t: Scalable data flow programming for many-task applications. SIGPLAN Not., 48(8):309–310.
Yeo, P. and Abidi, S. S. (2013). Dataflow oriented similarity matching for scientific workflows. IEEE 27th International, page 2091–2100.
Publicado
22/07/2017
Como Citar
DA SILVA JUNIOR, Daniel; PAES, Aline; DE OLIVEIRA, Daniel.
Predição de Falhas em Workflows Científicos em Nuvens baseada em Aprendizado de Máquina. In: BRAZILIAN E-SCIENCE WORKSHOP (BRESCI), 11. , 2017, São Paulo.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2017
.
p. 61-68.
ISSN 2763-8774.
DOI: https://doi.org/10.5753/bresci.2017.9923.
