Escalonamento justo em infraestruturas de nuvem com múltiplas classes de serviço
Resumo
Provedores de computação na nuvem oferecem múltiplas classes de serviço para lidar com a heterogeneidade da carga de trabalho. Essas classes são diferenciadas pela Qualidade de Serviço (QoS) esperada, que é definida em termos de Objetivos de Níveis de Serviço (SLO). Um Escalonamento Baseado em Prioridade (EBP) é uma forma comum de permitir que requisições por recursos de diferentes classes alcancem a QoS desejada. No entanto, em períodos de contenção de recursos, a QoS oferecida pode ser injusta para certos usuários. Neste trabalho, apresenta-se o Escalonamento Dirigido por SLO (EDSLO), que escalona com base nos SLOs e na QoS de cada requisição em um dado momento. Experimentos de simulação usando dados de um sistema real em produção foram executados para comparar as diferentes políticas. Em geral, o EDSLO entrega um serviço melhor que o EBP.
Palavras-chave:
Computação em Núvem, Qualidade de Serviço, Políticas de Escalonamento
Referências
Boutin, E., Ekanayake, J., Lin, W., Shi, B., Zhou, J., Qian, Z., Wu, M., and Zhou, L. (2014). Apollo: Scalable and coordinated scheduling for cloud-scale computing. In OSDI.
Burns, B., Grant, B., Oppenheimer, D., Brewer, E., and Wilkes, J. (2016). Borg, omega, and kubernetes. Commun. ACM.
Carvalho, M., Menascé, D., and Brasileiro, F. (2015). Prediction-based admission control for iaas clouds with multiple service classes. In IEEE 7th Int. Conf. on Cloud Computing. IEEE.
Delgado, P., Dinu, F., Kermarrec, A.-M., and Zwaenepoel, W. (2015). Hawk: Hybrid datacenter scheduling. In Proc. of the 2015 USENIX Annual Technical Conf., number EPFL-CONF-208856. USENIX Association.
Delimitrou, C. and Kozyrakis, C. (2013). Paragon: Qos-aware scheduling for heterogeneous datacenters. In ACM SIGPLAN Notices. ACM.
Delimitrou, C. and Kozyrakis, C. (2014). Quasar: resourceefficient and qos-aware cluster management. ACM SIGPLAN Notices.
Delimitrou, C., Sanchez, D., and Kozyrakis, C. (2015). Tarcil: reconciling scheduling speed and quality in large shared clusters. In Proc. of the 6th Symp. on Cloud Computing. ACM.
Gog, I., Schwarzkopf, M., Gleave, A., Watson, R. N., and Hand, S. (2016). Firmament: Fast, centralized cluster scheduling at scale. Usenix.
Goiri, I., Julia, F., Nou, R., Berral, J. L., Guitart, J., and Torres, J. (2010). Energy-aware scheduling in virtualized datacenters. In Cluster Computing (CLUSTER), 2010 IEEE Int. Conf. on.
Hindman, B., Konwinski, A., Zaharia, M., Ghodsi, A., Joseph, A. D., Katz, R. H., Shenker, S., and Stoica, I. (2011). Mesos: A platform for fine-grained resource sharing in the data center. In NSDI.
Isard, M., Prabhakaran, V., Currey, J., Wieder, U., Talwar, K., and Goldberg, A. (2009). Quincy: fair scheduling for distributed computing clusters. In Proc. of the ACM SIGOPS 22nd Symp. on Operating systems principles.
Karanasos, K., Rao, S., Curino, C., Douglas, C., Chaliparambil, K., Fumarola, G. M., Heddaya, S., Ramakrishnan, R., and Sakalanaga, S. (2015). Mercury: Hybrid centralized and distributed scheduling in large shared clusters. In USENIX ATC.
Kong, X., Lin, C., Jiang, Y., Yan, W., and Chu, X. (2011). Efficient dynamic task scheduling in virtualized data centers with fuzzy prediction. Journal of network and Computer Applications.
Ousterhout, K., Wendell, P., Zaharia, M., and Stoica, I. (2013). Sparrow: Distributed, low latency scheduling. In Proc. of the 24th ACM SOSP.
Pan, W., Rowe, J., and Barlaoura, G. (2013). Records in the cloud (ric) user survey report. Technical report.
Reiss, C., Tumanov, A., Ganger, G. R., Katz, R. H., and Kozuch, M. A. (2012). Heterogeneity and dynamicity of clouds at scale: Google trace analysis. In Symp. on Cloud Computing.
Schwarzkopf, M., Konwinski, A., Abd-El-Malek, M., and Wilkes, J. (2013). Omega: Flexible, scalable schedulers for large compute clusters. In Proc. of the 8th ACM European Conf. on Computer Systems.
Shahrad, M. andWentzlaff, D. (2016). Availability knob: Flexible user-defined availability in the cloud. In Proc. of the 7th ACM Symp. on Cloud Computing.
Vavilapalli, V. K., Murthy, A. C., Douglas, C., Agarwal, S., Konar, M., Evans, R., Graves, T., Lowe, J., Shah, H., Seth, S., et al. (2013). Apache hadoop yarn: Yet another resource negotiator. In Proc. of the 4th Symp. on Cloud Computing. ACM.
Verma, A., Korupolu, M., and Wilkes, J. (2014). Evaluating job packing in warehouse-scale computing. In 2014 IEEE Int’l Conf. on Cluster Computing, CLUSTER.
Verma, A., Pedrosa, L., Korupolu, M., Oppenheimer, D., Tune, E., and Wilkes, J. (2015). Large-scale cluster management at google with borg. In Proc. of the 10th European Conf. on Computer Systems.
Wilkes, J. (2011). More Google cluster data. Google research blog.
Burns, B., Grant, B., Oppenheimer, D., Brewer, E., and Wilkes, J. (2016). Borg, omega, and kubernetes. Commun. ACM.
Carvalho, M., Menascé, D., and Brasileiro, F. (2015). Prediction-based admission control for iaas clouds with multiple service classes. In IEEE 7th Int. Conf. on Cloud Computing. IEEE.
Delgado, P., Dinu, F., Kermarrec, A.-M., and Zwaenepoel, W. (2015). Hawk: Hybrid datacenter scheduling. In Proc. of the 2015 USENIX Annual Technical Conf., number EPFL-CONF-208856. USENIX Association.
Delimitrou, C. and Kozyrakis, C. (2013). Paragon: Qos-aware scheduling for heterogeneous datacenters. In ACM SIGPLAN Notices. ACM.
Delimitrou, C. and Kozyrakis, C. (2014). Quasar: resourceefficient and qos-aware cluster management. ACM SIGPLAN Notices.
Delimitrou, C., Sanchez, D., and Kozyrakis, C. (2015). Tarcil: reconciling scheduling speed and quality in large shared clusters. In Proc. of the 6th Symp. on Cloud Computing. ACM.
Gog, I., Schwarzkopf, M., Gleave, A., Watson, R. N., and Hand, S. (2016). Firmament: Fast, centralized cluster scheduling at scale. Usenix.
Goiri, I., Julia, F., Nou, R., Berral, J. L., Guitart, J., and Torres, J. (2010). Energy-aware scheduling in virtualized datacenters. In Cluster Computing (CLUSTER), 2010 IEEE Int. Conf. on.
Hindman, B., Konwinski, A., Zaharia, M., Ghodsi, A., Joseph, A. D., Katz, R. H., Shenker, S., and Stoica, I. (2011). Mesos: A platform for fine-grained resource sharing in the data center. In NSDI.
Isard, M., Prabhakaran, V., Currey, J., Wieder, U., Talwar, K., and Goldberg, A. (2009). Quincy: fair scheduling for distributed computing clusters. In Proc. of the ACM SIGOPS 22nd Symp. on Operating systems principles.
Karanasos, K., Rao, S., Curino, C., Douglas, C., Chaliparambil, K., Fumarola, G. M., Heddaya, S., Ramakrishnan, R., and Sakalanaga, S. (2015). Mercury: Hybrid centralized and distributed scheduling in large shared clusters. In USENIX ATC.
Kong, X., Lin, C., Jiang, Y., Yan, W., and Chu, X. (2011). Efficient dynamic task scheduling in virtualized data centers with fuzzy prediction. Journal of network and Computer Applications.
Ousterhout, K., Wendell, P., Zaharia, M., and Stoica, I. (2013). Sparrow: Distributed, low latency scheduling. In Proc. of the 24th ACM SOSP.
Pan, W., Rowe, J., and Barlaoura, G. (2013). Records in the cloud (ric) user survey report. Technical report.
Reiss, C., Tumanov, A., Ganger, G. R., Katz, R. H., and Kozuch, M. A. (2012). Heterogeneity and dynamicity of clouds at scale: Google trace analysis. In Symp. on Cloud Computing.
Schwarzkopf, M., Konwinski, A., Abd-El-Malek, M., and Wilkes, J. (2013). Omega: Flexible, scalable schedulers for large compute clusters. In Proc. of the 8th ACM European Conf. on Computer Systems.
Shahrad, M. andWentzlaff, D. (2016). Availability knob: Flexible user-defined availability in the cloud. In Proc. of the 7th ACM Symp. on Cloud Computing.
Vavilapalli, V. K., Murthy, A. C., Douglas, C., Agarwal, S., Konar, M., Evans, R., Graves, T., Lowe, J., Shah, H., Seth, S., et al. (2013). Apache hadoop yarn: Yet another resource negotiator. In Proc. of the 4th Symp. on Cloud Computing. ACM.
Verma, A., Korupolu, M., and Wilkes, J. (2014). Evaluating job packing in warehouse-scale computing. In 2014 IEEE Int’l Conf. on Cluster Computing, CLUSTER.
Verma, A., Pedrosa, L., Korupolu, M., Oppenheimer, D., Tune, E., and Wilkes, J. (2015). Large-scale cluster management at google with borg. In Proc. of the 10th European Conf. on Computer Systems.
Wilkes, J. (2011). More Google cluster data. Google research blog.
Publicado
06/05/2019
Como Citar
FARIAS, Giovanni; LOPES, Raquel; BRASILEIRO, Francisco; CARVALHO, Marcus; MORAIS, Fabio; MAFRA, João; TURULL, Daniel.
Escalonamento justo em infraestruturas de nuvem com múltiplas classes de serviço. In: SIMPÓSIO BRASILEIRO DE REDES DE COMPUTADORES E SISTEMAS DISTRIBUÍDOS (SBRC), 37. , 2019, Gramado.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2019
.
p. 636-649.
ISSN 2177-9384.
DOI: https://doi.org/10.5753/sbrc.2019.7392.
