Abstract
Verification algorithms are among the most resource-intensive computation tasks. Saving energy is important for our living environment and to save cost in data centers. Yet, researchers compare the efficiency of algorithms still in terms of consumption of CPU time (or even wall time). Perhaps one reason for this is that measuring energy consumption of computational processes is not as convenient as measuring the consumed time and there is no sufficient tool support. To close this gap, we contribute CPU Energy Meter, a small tool that takes care of reading the energy values that Intel CPUs track inside the chip. In order to make energy measurements as easy as possible, we integrated CPU Energy Meter into BenchExec, a benchmarking tool that is already used by many researchers and competitions in the domain of formal methods. As evidence for usefulness, we explored the energy consumption of some state-of-the-art verifiers and report some interesting insights, for example, that energy consumption is not necessarily correlated with CPU time.
Keywords: Energy Measurement, RAPL, Benchmarking, BenchExec
Contributor Information
Armin Biere, Email: biere@jku.at.
David Parker, Email: d.a.parker@cs.bham.ac.uk.
Philipp Wendler, Email: uni@philippwendler.de.
References
- 1.Bekas, C., Curioni, A.: A new energy aware performance metric. Computer Science - R&D 25(3-4), 187–195 (2010). 10.1007/s00450-010-0119-z
- 2.Beyer, D.: Reliable and reproducible competition results with BenchExec and witnesses (Report on SV-COMP 2016). In: Proc. TACAS. pp. 887–904. LNCS 9636, Springer (2016). 10.1007/978-3-662-49674-9_55
- 3.Beyer, D.: Automatic verification of C and Java programs: SV-COMP 2019. In: Proc. TACAS (3). pp. 133–155. LNCS 11429, Springer (2019). 10.1007/978-3-030-17502-3_9
- 4.Beyer, D., Löwe, S., Wendler, P.: Reliable benchmarking: Requirements and solutions. Int. J. Softw. Tools Technol. Transfer 21(1), 1–29 (2019). 10.1007/s10009-017-0469-y
- 5.Beyer, D., Wendler, P.: Replication package for article ‘CPU Energy Meter: A tool for energy-aware algorithms engineering’ in Proc. TACAS ’20. Zenodo (2020). 10.5281/zenodo.3679337
- 6.Desrochers, S., Paradis, C., Weaver, V.M.: A validation of DRAM RAPL power measurements. In: Proc. Int. Symposium on Memory Systems (MEMSYS). pp. 455–470. ACM (2016). 10.1145/2989081.2989088
- 7.Dongarra, J.J., Ltaief, H., Luszczek, P., Weaver, V.M.: Energy footprint of advanced dense numerical linear algebra using tile algorithms on multicore architectures. In: Proc. Int. Conference on Cloud and Green Computing (CGC). pp. 274–281. IEEE (2012). 10.1109/CGC.2012.113
- 8.Ge, R., Feng, X., Song, S., Chang, H., Li, D., Cameron, K.W.: PowerPack: Energy profiling and analysis of high-performance systems and applications. IEEE Trans. Parallel Distrib. Syst. 21(5), 658–671 (2010). 10.1109/TPDS.2009.76
- 9.Hackenberg, D., Ilsche, T., Schöne, R., Molka, D., Schmidt, M., Nagel, W.E.: Power measurement techniques on standard compute nodes: A quantitative comparison. In: Proc. Int. Symposium on Performance Analysis of Systems & Software (ISPASS). pp. 194–204. IEEE (2013). 10.1109/ISPASS.2013.6557170
- 10.Hackenberg, D., Schöne, R., Ilsche, T., Molka, D., Schuchart, J., Geyer, R.: An energy efficiency feature survey of the Intel Haswell processor. In: Proc. Int. Parallel and Distributed Processing Symposium (IPDPS). pp. 896–904. IEEE (2015). 10.1109/IPDPSW.2015.70
- 11.Hähnel, M., Döbel, B., Völp, M., Härtig, H.: Measuring energy consumption for short code paths using RAPL. SIGMETRICS Performance Evaluation Review 40(3), 13–17 (2012). 10.1145/2425248.2425252
- 12.Hindle, A.: Green software engineering: The curse of methodology. Tech. Rep. 4:e1470v2, PeerJ PrePrints (2016). 10.7287/peerj.preprints.1470v2
- 13.Ilsche, T., Hackenberg, D., Graul, S., Schöne, R., Schuchart, J.: Power measurements for compute nodes: Improving sampling rates, granularity and accuracy. In: Proc. Int. Green and Sustainable Computing Conference (IGSC). pp. 1–8. IEEE (2015). 10.1109/IGCC.2015.7393710
- 14.Intel: Intel 64 and IA-32 architectures software developer’s manual, vol. 3B, chap. 14.9 (December 2017), available at https://software.intel.com/en-us/articles/intel-sdm
- 15.Khan, K.N., Ou, Z., Hirki, M., Nurminen, J.K., Niemi, T.: How much power does your server consume? Estimating wall socket power using RAPL measurements. Computer Science - R&D 31(4), 207–214 (2016). 10.1007/s00450-016-0325-4
- 16.Scaramella, J., Eastwood, M.: Solutions for the data center’s thermal challenges. Tech. rep., IDC (2007), available at https://www-935.ibm.com/services/fr/igs/pdf/idc_opinion_coolblue_wp.pdf
- 17.Schuchart, J., Hackenberg, D., Schöne, R., Ilsche, T., Nagappan, R., Patterson, M.K.: The shift from processor power consumption to performance variations: Fundamental implications at scale. Computer Science - R&D 31(4), 197–205 (2016). 10.1007/s00450-016-0327-2
- 18.Venkatesh, A., Kandalla, K.C., Panda, D.K.: Evaluation of energy characteristics of MPI communication primitives with RAPL. In: Proc. Int. Symposium on Parallel & Distributed Processing (IPDPSW). pp. 938–945. IEEE (2013). 10.1109/IPDPSW.2013.243
- 19.Weaver, V.M., Johnson, M., Kasichayanula, K., Ralph, J., Luszczek, P., Terpstra, D., Moore, S.: Measuring energy and power with PAPI. In: Proc. Int. Conference on Parallel Processing Workshops (ICPPW). pp. 262–268. IEEE (2012). 10.1109/ICPPW.2012.39