Sensibilidade de políticas cientes de carbono para VMs: efeitos da especificação, do cenário e da comparabilidade da carga
Resumo
Avalia-se a sensibilidade da comparação entre 11 políticas de alocação e migração de máquinas virtuais (VMs) em 30 configurações, com 6.820 execuções pareadas por semente. Primeiro, a comparação entre duas especificações de ordenação mostrou alterações no posicionamento das VMs e nas métricas resultantes. Segundo, o cenário altera magnitudes dos contrastes e ordenações: sob PUE invertido contrafactual, a política de menor consumo de energia resultou em até 3,2× as emissões estimadas da política de menor emissão. Terceiro, em 10 das 26 configurações distintas, a admissão difere entre políticas e três métricas de emissão produzem ordenações incompatíveis, impedindo uma ordenação escalar. Assim, conclusões sobre o desempenho em carbono de uma política dependem conjuntamente de sua especificação, do cenário experimental e da comparabilidade da carga admitida. Esses elementos devem, portanto, integrar o desenho de avaliações comparativas de políticas cientes de carbono.Referências
Acun, B., Lee, B., Kazhamiaka, F., Maeng, K., Gupta, U., Chakkaravarthy, M., Brooks, D., and Wu, C.-J. (2023). Carbon explorer: A holistic framework for designing carbon aware datacenters. In ASPLOS 2023: Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2, pages 118–132. ACM.
Ali, H., Zakarya, M., Rahman, I. U., Khan, A. A., and Buyya, R. (2021). Followme@ls: Electricity price and source aware resource management in geographically distributed heterogeneous datacenters. Journal of Systems and Software, 175:110907.
Casanova, H., Giersch, A., Legrand, A., Quinson, M., and Suter, F. (2025). Lowering entry barriers to developing custom simulators of distributed applications and platforms with simgrid. Parallel Computing, 123:103125.
Clark, C., Fraser, K., Hand, S., Hansen, J. G., Jul, E., Limpach, C., Pratt, I., and Warfield, A. (2005). Live migration of virtual machines. In NSDI’05: Proceedings of the 2nd conference on Symposium on Networked Systems Design & Implementation - Volume 2, pages 273–286. USENIX Association.
Electricity Maps (2026). AE, AU-NSW, BR-CS, FR, IN-WE, KR, SG, US-MIDA-PJM and ZA 2021 Hourly Carbon Intensity Data (Version February 10, 2026). Electricity Maps. [link].
Hadary, O., Marshall, L., Menache, I., Pan, A., Greeff, E. E., Dion, D., Dorminey, S., Joshi, S., Chen, Y., Russinovich, M., and Moscibroda, T. (2020). Protean: VM allocation service at scale. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20), pages 845–861. USENIX Association.
Hanafy, W. A., Bostandoost, R., Bashir, N., Irwin, D., Hajiesmaili, M., and Shenoy, P. (2023). The war of the efficiencies: Understanding the tension between carbon and energy optimization. In HotCarbon ’23: Proceedings of the 2nd Workshop on Sustainable Computer Systems, pages 1–7. ACM.
Masanet, E., Shehabi, A., Lei, N., Smith, S., and Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science, 367(6481):984–986.
Saraiva, G., Vasconcelos, M., Bruschi, S., Carastan-Santos, D., and Cordeiro, D. (2025). Estimating CO2 emissions of distributed applications and platforms with SimGrid/Batsim. In Anais do XXVI Simpósio em Sistemas Computacionais de Alto Desempenho, pages 109–120, Porto Alegre, RS, Brasil. SBC.
Vasconcelos, M., Cordeiro, D., Da Costa, G., Dufossé, F., Nicod, J.-M., and Rehn-Sonigo, V. (2023). Optimal sizing of a globally distributed low carbon cloud federation. In 2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing (CCGrid), pages 203–215. IEEE.
Vasconcelos, M. F. S., Cordeiro, D., and Dufossé, F. (2022). Indirect network impact on the energy consumption in multi-clouds for follow-the-renewables approaches. In Proceedings of the 11th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS), pages 44–55.
Wiesner, P., Behnke, I., Scheinert, D., Gontarska, K., and Thamsen, L. (2021). Let’s wait awhile: how temporal workload shifting can reduce carbon emissions in the cloud. In Middleware ’21: Proceedings of the 22nd International Middleware Conference, pages 260–272. ACM.
Wilkes, J. (2011). Google cluster-usage traces: format + schema. Technical report, Google Inc.
Xu, M. and Buyya, R. (2020). Managing renewable energy and carbon footprint in multi-cloud computing environments. Journal of Parallel and Distributed Computing, 135:191–202.
Ali, H., Zakarya, M., Rahman, I. U., Khan, A. A., and Buyya, R. (2021). Followme@ls: Electricity price and source aware resource management in geographically distributed heterogeneous datacenters. Journal of Systems and Software, 175:110907.
Casanova, H., Giersch, A., Legrand, A., Quinson, M., and Suter, F. (2025). Lowering entry barriers to developing custom simulators of distributed applications and platforms with simgrid. Parallel Computing, 123:103125.
Clark, C., Fraser, K., Hand, S., Hansen, J. G., Jul, E., Limpach, C., Pratt, I., and Warfield, A. (2005). Live migration of virtual machines. In NSDI’05: Proceedings of the 2nd conference on Symposium on Networked Systems Design & Implementation - Volume 2, pages 273–286. USENIX Association.
Electricity Maps (2026). AE, AU-NSW, BR-CS, FR, IN-WE, KR, SG, US-MIDA-PJM and ZA 2021 Hourly Carbon Intensity Data (Version February 10, 2026). Electricity Maps. [link].
Hadary, O., Marshall, L., Menache, I., Pan, A., Greeff, E. E., Dion, D., Dorminey, S., Joshi, S., Chen, Y., Russinovich, M., and Moscibroda, T. (2020). Protean: VM allocation service at scale. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20), pages 845–861. USENIX Association.
Hanafy, W. A., Bostandoost, R., Bashir, N., Irwin, D., Hajiesmaili, M., and Shenoy, P. (2023). The war of the efficiencies: Understanding the tension between carbon and energy optimization. In HotCarbon ’23: Proceedings of the 2nd Workshop on Sustainable Computer Systems, pages 1–7. ACM.
Masanet, E., Shehabi, A., Lei, N., Smith, S., and Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science, 367(6481):984–986.
Saraiva, G., Vasconcelos, M., Bruschi, S., Carastan-Santos, D., and Cordeiro, D. (2025). Estimating CO2 emissions of distributed applications and platforms with SimGrid/Batsim. In Anais do XXVI Simpósio em Sistemas Computacionais de Alto Desempenho, pages 109–120, Porto Alegre, RS, Brasil. SBC.
Vasconcelos, M., Cordeiro, D., Da Costa, G., Dufossé, F., Nicod, J.-M., and Rehn-Sonigo, V. (2023). Optimal sizing of a globally distributed low carbon cloud federation. In 2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing (CCGrid), pages 203–215. IEEE.
Vasconcelos, M. F. S., Cordeiro, D., and Dufossé, F. (2022). Indirect network impact on the energy consumption in multi-clouds for follow-the-renewables approaches. In Proceedings of the 11th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS), pages 44–55.
Wiesner, P., Behnke, I., Scheinert, D., Gontarska, K., and Thamsen, L. (2021). Let’s wait awhile: how temporal workload shifting can reduce carbon emissions in the cloud. In Middleware ’21: Proceedings of the 22nd International Middleware Conference, pages 260–272. ACM.
Wilkes, J. (2011). Google cluster-usage traces: format + schema. Technical report, Google Inc.
Xu, M. and Buyya, R. (2020). Managing renewable energy and carbon footprint in multi-cloud computing environments. Journal of Parallel and Distributed Computing, 135:191–202.
Publicado
03/11/2026
Como Citar
SILVA, Guilherme Fernandes Moraes da; CORDEIRO, Daniel.
Sensibilidade de políticas cientes de carbono para VMs: efeitos da especificação, do cenário e da comparabilidade da carga. In: SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 27. , 2026, Natal/RN.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 247-260.
DOI: https://doi.org/10.5753/sscad.2026.34925.
