A MOPSO-Based Optimization Strategy Applied to Electrical Subsystems of Data Centers

• F. M. Sousa Sobrinho UFRPE
• G. R. A. Callou UFRPE
• W. S. Leonardo UFRPE
• B. C. S. Nogueira UFAL

DOI: https://doi.org/10.5753/wperformance.2022.223081

Abstract

Data center infrastructures must have high availability, low cost, and high energy efficiency. However, these objectives are often conflicting. This paper presents a strategy based on the MOPSO, a multi-objective particle swarm optimization, to improve the design of electrical data center architectures. To show the applicability of the proposed strategy, we present a comparative study between a brute force algorithm and the proposed strategy. Six models of electrical architectures were defined, the results showed that the application of the proposed strategy reduces the execution time by up to 870 times and showed that the algorithm can generate the approximate Pareto frontier with a difference of 3% .

Keywords: Availability, Data Center, Modeling, Optimization

References

Abualigah, L. and Diabat, A. (2021). A novel hybrid antlion optimization algorithm for multi-objective task scheduling problems in cloud computing environments. Cluster Computing, pages 205–223.

Alvarez-Benitez, J., Everson, R., and Fieldsend, J. (2005). A mopso algorithm based exclusively on pareto dominance concepts. volume 3410, pages 459–473.

Association, T. I. (2005). Telecommunications Infrastructure Standard for Data Centers ANSI/TIA-942. Telecommunication Industry Association.

Callou, G., Ferreira, J., Maciel, P., Tutsch, D., and Souza, R. (2014). An integrated modeling approach to evaluate and optimize data center sustainability, dependability and cost. Energies, 7(1):238–277.

Deb, K. and Goel, T. (2001). Controlled elitist non-dominated sorting genetic algorithms for better convergence. In International conference on evolutionary multi-criterion optimization, pages 67–81. Springer.

Guimarães, A. P. and Pereira, A. (2020). Análise de aspectos de dependabilidade em sistemas de data centers integrando as infraestruturas de comunicação, de potência e de refrigeração. Revista Brasileira de Administração Científica.

Isaak, P. (2021). Architecture: Data center rack floor plan and facility layout design. Data Center Handbook: Plan, Design, Build, and Operations of a Smart Data Center.

Koot, M. and Wijnhoven, F. (2021). Usage impact on data center electricity needs: A system dynamic forecasting model. Applied Energy, 291:116798.

Li, Y., Wen, Y., Tao, D., and Guan, K. (2020). Transforming cooling optimization for green data center via deep reinforcement learning. IEEE Transactions on Cybernetics.

Marsan, M. A. (1990). Stochastic Petri Nets: An Elementary Introduction, page 1–29. Springer-Verlag, Berlin, Heidelberg.

Masanet, E., Shehabi, A., Lei, N., Smith, S., and Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science, 367(6481):984–986.

Nguyen, T. A., Min, D., Choi, E., and Tran, T. D. (2019). Reliability and availability evaluation for cloud data center networks using hierarchical models. IEEE Access.

Oliveira, D., Matos, R., Dantas, J., Ferreira, J. a., Silva, B., Callou, G., Maciel, P., and Brinkmann, A. (2017). Advanced stochastic petri net modeling with the mercury scripting language. New York, NY, USA. Association for Computing Machinery.

Souza Leonardo, W. and Callou, G. (2021). Stars: um ambiente integrado para avaliação de disponibilidade, custo e consumo de energia de sistemas. Revista Brasileira de Computação Aplicada, 13(3):10–21.

Sun, J., Deng, J., and Li, Y. (2020). Indicator & crowding distance-based evolutionary algorithm for combined heat and power economic emission dispatch. Applied Soft Computing, 90:106158.

Wang, W., Loman, J. M., Arno, R. G., Vassiliou, P., Furlong, E. R., and Ogden, D. (2004). Reliability block diagram simulation techniques applied to the ieee std. 493 standard network. IEEE Transactions on Industry Applications, 40(3):887–895.