Incluindo Previsão num Gerenciador Elástico de Memória
Abstract
Cloud computing offers users relatively cheap and easy access to resources and services with sufficient capacity to meet their response times and performance requirements. To be cost effective, cloud provider's have to optimize resource utilization to meet the demands of the largest number of clients using the fewest resources. This work aims to improve decisions made by MEC, a component of the MEMiC tool to manage memory elasticity of VMs. A new predictive performance model is used to adjust VM preemption and scheduling to reduce application execution times and increase resource utilization. Initial experimentation confirms the benefits of the proposed proactive approach.References
Al-Dhuraibi, Y., Zalila, F., Djarallah, N. B., and Merle, P. (2018). Coordinating Vertical Elasticity of both Containers and Virtual Machines. In 8th Int. Conf. Cloud Computing and Service Sciences (CLOSER).
Chen, W., Zhou, X., and Rao, J. (2020). Preemptive and low latency datacenter scheduling via lightweight containers. IEEE Trans. on Par. and Dist. Systems, 31(12):2749–2762.
Farokhi, S., Jamshidi, P., Lakew, E., Brandic, I., and Elmroth, E. (2016). A hybrid cloud controller for vertical memory elasticity: A control-theoretic approach. Future Generation Comp Sys, 65:57–72.
Kotz, S., Read, C. B., Balakrishnan, N., Vidakovic, B., and Johnson, N. L. (2004). Encyclopedia of Statistical Sciences. Wiley-Interscience.
Moltó, G., Caballer, M., Romero, E., and Alfonso, C. (2013). Elastic memory management of virtualized infrastructures for applications with dynamic memory requirements. Procedia Comp Sci., 18:159–168.
Sawamura, R., Boeres, C., and Rebello, V. E. F. (2015). Evaluating the Impact of Memory Allocation and Swap for Vertical Memory Elasticity in VMs. In 2015 27th Int. Sym. on Computer Arc. and High Performance Computing (SBAC-PAD), pages 186–193.
Sawamura, R., Boeres, C., and Rebello, V. E. F. (2016). MEC: The Memory Elasticity Controller. 23rd IEEE Int. Conf. on High Performance Comp. (HiPC), pages 111–120.
Valencia, J., Boeres, C., and Rebello, V. E. F. (2018). Combining VM preemption schemes to improve vertical memory elasticity scheduling in clouds. In IEEE/ACM 11th Int. Conf. on Utility and Cloud Computing (UCC), pages 53–62.
Chen, W., Zhou, X., and Rao, J. (2020). Preemptive and low latency datacenter scheduling via lightweight containers. IEEE Trans. on Par. and Dist. Systems, 31(12):2749–2762.
Farokhi, S., Jamshidi, P., Lakew, E., Brandic, I., and Elmroth, E. (2016). A hybrid cloud controller for vertical memory elasticity: A control-theoretic approach. Future Generation Comp Sys, 65:57–72.
Kotz, S., Read, C. B., Balakrishnan, N., Vidakovic, B., and Johnson, N. L. (2004). Encyclopedia of Statistical Sciences. Wiley-Interscience.
Moltó, G., Caballer, M., Romero, E., and Alfonso, C. (2013). Elastic memory management of virtualized infrastructures for applications with dynamic memory requirements. Procedia Comp Sci., 18:159–168.
Sawamura, R., Boeres, C., and Rebello, V. E. F. (2015). Evaluating the Impact of Memory Allocation and Swap for Vertical Memory Elasticity in VMs. In 2015 27th Int. Sym. on Computer Arc. and High Performance Computing (SBAC-PAD), pages 186–193.
Sawamura, R., Boeres, C., and Rebello, V. E. F. (2016). MEC: The Memory Elasticity Controller. 23rd IEEE Int. Conf. on High Performance Comp. (HiPC), pages 111–120.
Valencia, J., Boeres, C., and Rebello, V. E. F. (2018). Combining VM preemption schemes to improve vertical memory elasticity scheduling in clouds. In IEEE/ACM 11th Int. Conf. on Utility and Cloud Computing (UCC), pages 53–62.
Published
2020-10-21
How to Cite
MOREIRA, Gustavo; GONÇALVES, André; BOERES, Cristina; REBELLO, Vinod.
Incluindo Previsão num Gerenciador Elástico de Memória. In: UNDERGRADUATE RESEARCH WORKSHOP - SYMPOSIUM ON HIGH PERFORMANCE COMPUTING SYSTEMS (SSCAD), 21. , 2020, Evento Online.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2020
.
p. 62-69.
DOI: https://doi.org/10.5753/wscad_estendido.2020.14090.
