Processo de KDD para Auxílio à Reconfiguração de Ambientes Virtualizados
Resumo
Este trabalho propõe um processo de KDD para auxílio à reconfiguração de máquinas virtualizadas Xen. Xen permite a execução simultânea de diversas máquinas virtuais (VM) sobre um mesmo hardware. Inicialmente, são obtidos dados de desempenho de cada VM, compilados de execuções de benchmarks sobre cada uma. Esses dados, após armazenados no data warehouse, podem ser preparados para serem utilizados por técnicas de mineração. Os modelos preditivos gerados podem, então, ser enriquecidos com instruções de reconfiguração. Tais modelos buscam sugerir, dada uma configuração vigente, qual o melhor conjunto de parâmetros de configuração para modificar o ambiente, e alcançar um ganho global de desempenho.
Referências
Barham et al. 2003. Xen and the art of virtualization. In: ACM Symposium on Operating Systems Principles. Proceedings… N.Y.: ACM Press, 2003, p. 164-177.
Becker K.; Ruiz, D.; Cunha, V.; Novello, T.; Souza, F. SPDW : a Software Development Process Performance Data Warehousing Environment. In: 30th Annual IEEE/NASA Software Engineering Workshop. Anais do 30th Annual IEEE/NASA SEW-06. Los Alamitos - CA : IEEE Press, 2006. v. 1. p. 107-118.
Chen, Y. et al. SLA decomposition: translating service level objectives (SLOs) to low-level system thresholds. In: International Conference on Autonimic Cmputing. Proceedings… Washington: IEEE Computer Society, Jun. 2007, p. 3.
Fayyad, U., Piatetsky-S G., Smyth P. The KDD Process for Extracting Useful Knowledge from Volumes of Data. Communications of the ACM, N.Y, v39, n11, p27-34, 1996.
Han, J.; Kamber, M. Data mining: Concepts and Techniques. San Francisco: Morgan Kaufmann, 2001.
Iozone. <http://www.iozone.org>. Acesso out 2007.
Jung, G., et al. Detecting Bottleneck in n-Tier IT applications through analysis. In: International Workshop on Distributed System: operation and management. Proceedings… Dublin: Lecture Notes in Computer Science, Oct. 2006, p. 149-160.
Kimball, R., Ross, M. The data warehouse toolkit. São Paulo: Campus, 2002.
Li, T., Ruan, D. An extended process model of knowledge discovery in databases. Enterprise Information Management, v. 20, n. 2, p.169-177, 2007.
Mc Voy, L. W., Staelin, C. Lmbench: portable tools for performance analysis. In: Annual Technical Conference. Proceedings… San Diego: Usenix, 1996, p279-294.
Mergen, M. F., Uhlig, V., Kreiger, O., Xendis, J. 2006. Virtualization for highperformance computing. SIGOPS. Operating System. Review, New York, v. 40, n. 2, p. 8-11, Apr. 2006.
Parekh, J., et al. Comparison of performance analysis approaches for bottleneck detection in multi-tier enterprise applications. In: International Workshop on Quality of Service. Proceedings… New Haven: IEEE Computer Science, Jun. 2006, p. 302-311.
Quinlan, J. R. C4.5: programs for machine learning. CA: Morgan Kaufman, 1996.
RUBIS: Rice University Bidding System. <rubis.objectweb.org/>. Acesso out 2007.
Tan, P-N., Steinbach, M., Kumar, V. Introduction to data mining. Boston: Addison Wesley, 2006.
Udupi, B., Sahai, A., Sinhal, S. A classification-based approach to policy refinement. In: Symposium on Integrated Management. Proceedings… IEEE Computer Science, May 2007, p. 785-788.
Unixbench. < www.tux.org/pub/tux/niemi/unixbench/>. Acesso out 2007.
Weka: Waikato Environment for Knowledge Analysis. <www.cs.waikato.ac.nz/ml/weka/>. Acesso out 2007.
Becker K.; Ruiz, D.; Cunha, V.; Novello, T.; Souza, F. SPDW : a Software Development Process Performance Data Warehousing Environment. In: 30th Annual IEEE/NASA Software Engineering Workshop. Anais do 30th Annual IEEE/NASA SEW-06. Los Alamitos - CA : IEEE Press, 2006. v. 1. p. 107-118.
Chen, Y. et al. SLA decomposition: translating service level objectives (SLOs) to low-level system thresholds. In: International Conference on Autonimic Cmputing. Proceedings… Washington: IEEE Computer Society, Jun. 2007, p. 3.
Fayyad, U., Piatetsky-S G., Smyth P. The KDD Process for Extracting Useful Knowledge from Volumes of Data. Communications of the ACM, N.Y, v39, n11, p27-34, 1996.
Han, J.; Kamber, M. Data mining: Concepts and Techniques. San Francisco: Morgan Kaufmann, 2001.
Iozone. <http://www.iozone.org>. Acesso out 2007.
Jung, G., et al. Detecting Bottleneck in n-Tier IT applications through analysis. In: International Workshop on Distributed System: operation and management. Proceedings… Dublin: Lecture Notes in Computer Science, Oct. 2006, p. 149-160.
Kimball, R., Ross, M. The data warehouse toolkit. São Paulo: Campus, 2002.
Li, T., Ruan, D. An extended process model of knowledge discovery in databases. Enterprise Information Management, v. 20, n. 2, p.169-177, 2007.
Mc Voy, L. W., Staelin, C. Lmbench: portable tools for performance analysis. In: Annual Technical Conference. Proceedings… San Diego: Usenix, 1996, p279-294.
Mergen, M. F., Uhlig, V., Kreiger, O., Xendis, J. 2006. Virtualization for highperformance computing. SIGOPS. Operating System. Review, New York, v. 40, n. 2, p. 8-11, Apr. 2006.
Parekh, J., et al. Comparison of performance analysis approaches for bottleneck detection in multi-tier enterprise applications. In: International Workshop on Quality of Service. Proceedings… New Haven: IEEE Computer Science, Jun. 2006, p. 302-311.
Quinlan, J. R. C4.5: programs for machine learning. CA: Morgan Kaufman, 1996.
RUBIS: Rice University Bidding System. <rubis.objectweb.org/>. Acesso out 2007.
Tan, P-N., Steinbach, M., Kumar, V. Introduction to data mining. Boston: Addison Wesley, 2006.
Udupi, B., Sahai, A., Sinhal, S. A classification-based approach to policy refinement. In: Symposium on Integrated Management. Proceedings… IEEE Computer Science, May 2007, p. 785-788.
Unixbench. < www.tux.org/pub/tux/niemi/unixbench/>. Acesso out 2007.
Weka: Waikato Environment for Knowledge Analysis. <www.cs.waikato.ac.nz/ml/weka/>. Acesso out 2007.
Publicado
07/04/2008
Como Citar
WINCK, Ana T.; RUIZ, Duncan D..
Processo de KDD para Auxílio à Reconfiguração de Ambientes Virtualizados. In: SIMPÓSIO BRASILEIRO DE SISTEMAS DE INFORMAÇÃO (SBSI), 4. , 2008, Rio de Janeiro.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2008
.
p. 211-222.
DOI: https://doi.org/10.5753/sbsi.2008.5927.
