Performance Assessment and Application of Energy Consumption Implemented with OpenMP SIMD

• Gustavo Tamiosso UNIPAMPA
• Gustavo Berned UNIPAMPA
• Marcelo Caggiani Luizelli UNIPAMPA
• Fabio Rossi IFFAR
• Jorji Nonaka RIKEN Center for Computational Science
• Arthur Lorenzon UNIPAMPA

DOI: https://doi.org/10.5753/eradrs.2020.10758

Abstract

Explore the parallelism has become an alternative to reduce energy consumption and improve application performance. This work aims to evaluate the performance statistics and energy of applications implemented using vector operations. Through the implementation of four known applications, we show that the use of vector operations can reduce run time by up to 37% and reduce energy consumption by 35%.

Keywords: Algorithms Parallel and Distributed, Computer architecture and processors, Evaluation, Performance Measurement and Prediction, Languages, Compilers and Tools Parallel and Distributed Computing

References

Abbo, A. A. e. a. (2004). Power consumption of performance-scaled simd processors. In Integrated Circuit and System Design. Power and Timing Modeling, Optimization and Simulation, pages 532–540. Springer Berlin Heidelberg.

Chapman, B., Jost, G., and Pas, R. V. D. (2007). Using OpenMP - Portable Shared Memory Programming.

Che, S., Boyer, M., Meng, J., Tarjan, D., Sheaffer, J. W., Lee, S.-H., and Skadron, K. (2009). Rodinia: A benchmark suite for heterogeneous computing. In IEEE Int. Symp. on Workload Characterization, pages 44–54, DC, USA. IEEE Computer Society.

Inoue, H. (2016). How simd width affects energy efficiency: A case study on sorting. In COOL CHIPS XIX, pages 1–3.

Jakobs, T. and Rünger, G. (2018). On the energy consumption of load/store avx instructions. In FedCSIS, pages 319–327.

Ponte, C., González-Domínguez, J., and Martín, M. J. (2017). Evaluation of openmp simd directives on xeon phi coprocessors. In HPCS, pages 389–395.