Abstract
The fault tolerance method currently used in High Performance Computing (HPC) is the rollback-recovery method by using checkpoints. This, like any other fault tolerance method, adds an additional energy consumption to that of the execution of the application. The objective of this work is to determine the factors that affect the energy consumption of the computing nodes on homogeneous cluster, when performing checkpoint and restart operations, on SPMD (Single Program Multiple Data) applications. We have focused on the energetic study of compute nodes, contemplating different configurations of hardware and software parameters. We studied the effect of performance states (states P) and power states (states C) of processors, application problem size, checkpoint software (DMTCP) and distributed file system (NFS) configuration. The results analysis allowed to identify opportunities to reduce the energy consumption of checkpoint and restart operations.
This research has been supported by the Agencia Estatal de Investigación (AEI), Spain and the Fondo Europeo de Desarrollo Regional (FEDER) UE, under contract TIN2017-84875-P and partially funded by a research collaboration agreement with the Fundacion Escuelas Universitarias Gimbernat (EUG).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amrizal, M.A., Takizawa, H.: Optimizing energy consumption on HPC systems with a multi-level checkpointing mechanism. In: International Conference on Networking, Architecture, and Storage (NAS). IEEE (2017)
Mills, B., Grant, R.E., Ferreira, K.B.: Evaluating energy savings for checkpoint/restart. In: Proceedings of the 1st International Workshop on Energy Efficient Supercomputing. ACM (2013)
Mills, B., Znati, T., Melhem, R., Ferreira, K.B., Grant, R.E.: Energy consumption of resilience mechanisms in large scale systems. In: 122nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing. IEEE (2014)
Meneses, E., Sarood, O., Kalé, L.V.: Assessing energy efficiency of fault tolerance protocols for HPC systems. In: 2012 IEEE 24th International Symposium on Computer Architecture and High Performance Computing, pp. 35–42. IEEE (2012)
Diouri, M., Glück, O., Lefevre, L., Cappello, F.: Energy considerations in checkpointing and fault tolerance protocols. In IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN 2012), June, pp. 1–6. IEEE (2012)
Bergman, K., et al.: Exascale computing study: technology challenges in achieving exascale systems. Defense Advanced Research Projects Agency Information Processing Techniques Office (DARPA IPTO), Technical report, 15 (2008)
Dauwe, D., Jhaveri, R., Pasricha, S., Maciejewski, A.A., Siegel, H.J.: Optimizing checkpoint intervals for reduced energy use in exascale systems. In: 2017 Eighth International Green and Sustainable Computing Conference (IGSC), October, pp. 1–8. IEEE (2017)
Chandrasekar, R.R., Venkatesh, A., Hamidouche, K., Panda, D.K.: Power-check: an energy-efficient checkpointing framework for HPC clusters. In: 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, May, pp. 261–270. IEEE (2015)
Cui, X., Znati, T., Melhem, R.: Adaptive and power-aware resilience for extreme-scale computing. In: International IEEE Conferences on Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld), July, pp. 671–679. IEEE (2016)
Saito, T., Sato, K., Sato, H., Matsuoka, S.: Energy-aware I/O optimization for checkpoint and restart on a NAND flash memory system. In: Proceedings of the 3rd Workshop on Fault-Tolerance for HPC at extreme scale, June, pp. 41–48. ACM (2013)
Ferreira, K.B., Ibtesham, D., DeBonis, D., Arnold, D.: Coarse-grained energy modeling of rollback/recovery mechanisms (No. SAND2014-2159C). Sandia National Lab. (SNL-NM), Albuquerque, NM (United States) (2014)
Silveira, D.S., Moro, G.B., Cruz, E.H.M., Navaux, P.O.A., Schnorr, L.M., Bampi, S.: Energy consumption estimation in parallel applications: an analysis in real and theoretical models. In: XVII Simposio em Sistemas Computacionais de Alto Desempenho, pp. 134–145 (2016)
Le Sueur, E., Heiser, G.: Dynamic voltage and frequency scaling: the laws of diminishing returns. In: Proceedings of the 2010 International Conference on Power Aware Computing and Systems, October, pp. 1–8 (2010)
Ansel, J., Arya, K., Cooperman, G.: DMTCP: transparent checkpointing for cluster computations and the desktop. In: 2009 IEEE International Symposium on Parallel & Distributed Processing, May, pp. 1–12. IEEE (2009)
Morán, M., Balladini, J., Rexachs, D., Luque E.: Factores que afectan el consumo energético de operaciones de checkpoint y restart en clusters. In: XIX Workshop Procesamiento Distribuido y Paralelo (WPDP), XXIV Congreso Argentino de Ciencias de la Computación, CACIC 2018, pp. 63–72 (2018). ISBN 978-950-658-472-6
Diouri, M., Glück, O., Lefevre, L., Cappello, F.: ECOFIT: a framework to estimate energy consumption of fault tolerance protocols for HPC applications. In: 13th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing, May, pp. 522–529. IEEE (2013)
ACPI - Advanced Configuration and Power Interface. http://www.acpi.info
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Morán, M., Balladini, J., Rexachs, D., Luque, E. (2019). Checkpoint and Restart: An Energy Consumption Characterization in Clusters. In: Pesado, P., Aciti, C. (eds) Computer Science – CACIC 2018. CACIC 2018. Communications in Computer and Information Science, vol 995. Springer, Cham. https://doi.org/10.1007/978-3-030-20787-8_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-20787-8_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-20786-1
Online ISBN: 978-3-030-20787-8
eBook Packages: Computer ScienceComputer Science (R0)