Server Workload Model Identification: Monitoring and Control Tools for Linux

• Autorzy: Karpowicz M. , Arabas P.
• Języki publikacji: EN

Abstrakty

EN

Server power control in data centers is a coordinated process carefully designed to reach multiple data center management objectives. The main objectives include avoiding power capacity overloads and system overheating, as well as fulfilling service-level agreements (SLAs). In addition to the primary goals, server control process aims to maximize various energy efficiency metrics subject to reliability constraints. Monitoring of data center performance is fundamental for its efficient management. In order to keep track of how well the computing tasks are processed, cluster control systems need to collect accurate measurements of activities of cluster components. This paper presents a brief overview of performance and power consumption monitoring tools available in the Linux systems.

Słowa kluczowe

EN

cloud computing; energy efficiency; Linux; server performance meterin

• Wydawca: Instytut Łączności - Państwowy Instytut Badawczy
• Czasopismo: Journal of Telecommunications and Information Technology
• Rocznik: 2016
• Tom: nr 2
• Strony: 5--12
• Opis fizyczny: Bibliogr. 47 poz., rys.

Twórcy

autor

Karpowicz M.

• Researchand Academic Computer Network (NASK), Wąwozowa st 18, 02-796 Warsaw, Poland

autor

Arabas P.

• Researchand Academic Computer Network (NASK), Wąwozowa st 18, 02-796 Warsaw, Poland
• Institute of Control and Computation Engineering, Warsaw University of Technology, Nowowiejska st 15/19, 00-665 Warsaw, Poland

Bibliografia

• [1] J. Koomey, Growth in Data Center Electricity Use 2005 to 2010. Oakland, CA: Analytical Press, 2011.
• [2] B. Subramaniam and Wu-chun Feng, “Towards energy-proportional computing for enterprise-class server workloads”, in Proc. 4th ACM/SPEC Int. Conf. Perform. Engin. ICPE 2013, Prague, Czech Republic, 2013, pp. 15-26.
• [3] J. Dongarra et al., “The international exascale software project roadmap”, Int. J. High Perform. Comput. Appl., vol. 25, no. 1, pp. 3-60, 2011.
• [4] S.-Y. Jing, S. Ali, K. She, and Y. Zhong, “State-of-the-art research study for green cloud computing”, The J. of Supercomput., vol. 65, no. 1, pp. 445-468, 2013.
• [5] V. K. Vavilapalli et al., “Apache hadoop yarn: yet another resource negotiator”, in Proc. 4th Ann. Symp.on Cloud Comput. SOCC'13, Santa Clara, CA, USA, 2013, pp. 5:1-5:16, 2013.
• [6] S. Jha, J. Qiu, A. Luckow, P. Mantha, and G. C. Fox, “A tale of two data-intensive paradigms: applications, abstractions, and architectures”, in Proc. 3rd IEEE Int. Congr. on Big Data BigData 2014, Anchorage, AK, USA, 2014, pp. 645-652.
• [7] Y. Georgiou, T. Cadeau, D. Glesser, D. Auble, M. Jette, and M. Hautreux, “Energy accounting and control with SLURM resource and job management system”, in Distributed Computing and Networking, M. Chatterjee et al., Eds., LNCS, vol. 8314, pp. 96-118. Springer, 2014.
• [8] E. Niewiadomska-Szynkiewicz, A. Sikora, P. Arabas, M. Kamola, M. Mincer, and J. Kołodziej, “Dynamic power management in energy-aware computer networks and data intensive computing systems”, Future Gener. Comp. Syst., vol. 37, pp. 284-296, 2014(doi: 10.1016/j.future.2013.10.002).
• [9] M. P. Karpowicz, P. Arabas, and E. Niewiadomska-Szynkiewicz, ”Energy-aware multilevel control system for a network of Linux software routers: design and implementation”, IEEE Syst. J., vol. PP, no. 99, pp. 1-12, 2015(doi: 10.1109/JSUST.20152489244).
• [10] P. Jaskóła, P. Arabas, and A. Karbowski, “Simultaneous routing and flow rate optimization in energy aware computer networks”, Int. J. Appl. Mathem. & Comp. Sci., vol. 26, no. 1, pp. 231-243, 2016.
• [11] A. Karbowski and P. Jaskóła, “Two approaches to dynamic power management in energy-aware computer networks – methodological considerations”, in Proc. of Feder. Conf. Comp. Science and Inform. Syst. FedCSIS 2015, Łódź, Poland, 2015, vol. 5, pp. 1177-1182.
• [12] ACPI Specification Document [Online]. Available: www.acpi.info
• [13] J. H. Laros III, D. DeBonis, R. Grant, S. M. Kelly, M. Levenhagen, S. Olivier, and K. Pedretti, “High performance computing power application programming interface specification”, Tech. Rep. SAND2014-17061, Sandia National Laboratories, 2014.
• [14] D. DeBonis et al., “A power API for the HPC community”, Sandia Report SAND2014-17061, Sandia National Laboratories, 2014.
• [15] R. Bolla et al., “Green Abstraction Layer (GAL): power management capabilities of the future energy telecommunication fixed network nodes”, Techn. Rep. ES 203 237, ETSI, 2014.
• [16] C. Lefurgy, K. Rajamani, F. Rawson, W. Felter, M. Kistler, and T. W. Keller, “Energy management for commercial servers”, Computer, vol. 36, no. 12, pp. 39-48, 2003.
• [17] L. A. Barroso, J. Clidaras, and U. Hölzle, The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines, 2nd ed. Morgan & Claypool Publ., 2013.
• [18] L. Wang and S. U. Khan, “Review of performance metrics for green data centers: a taxonomy study”, The J. Supercomput., vol. 63, no. 3, pp. 639-656, 2013.
• [19] T. Mastelic, A. Oleksiak, H. Claussen, I. Brandic, J.-M. Pierson, and A. V. Vasilakos, “Cloud computing: survey on energy efficiency”, ACM Comput. Surveys, vol. 47, no. 2, pp. 33:1.33:36, 2015.
• [20] B. Subramaniam, W. Saunders, T. Scogland, and Wu-chun Feng, “Trends in energy-efficient computing: A perspective from the Green500”, in 4th Int. Green Comput. Conf. IGCC 2013, Arlington, VA, USA, 2013, pp. 1-8.
• [21] Standard Performance Evaluation Corporation (SPEC), SPEC Power and Performance Benchmark Methodology [Online]. Available: www.spec.org/power ssj2008/
• [22] K.-D. Lange, “Identifying Shades of Green: The SPECpower Benchmarks”, IEEE Computer, vol. 42, no. 3, pp. 95-97, 2009.
• [23] D. Molka, D. Hackenberg, R. Schöne, T. Minartz, and W. E. Nagel, “Flexible workload generation for HPC cluster efficiency benchmarking”, Comp. Science-Res. & Develop., vol. 27, no. 4, pp. 235-243, 2012.
• [24] J. J. Dongarra, P. Luszczek, and A. Petitet, “The LINPACK Benchmark: past, present and future”, Concurr. & Comput.: Practice and Experience, vol. 15, no. 9, pp. 803-820, 2003.
• [25] A. Iosup, S. Ostermann, M. N. Yigitbasi, R. Prodan, T. Fahringer, and D. H. J. Epema, “Performance analysis of cloud computing services for many-tasks scientific computing”, IEEE Trans. Parall. & Distrib. Syst., vol. 22, no. 6, pp. 931-945, 2011.
• [26] S. Bradner and J. McQuaid, “RFC 2544: Benchmarking methodology for network interconnect devices”, Mar. 1999.
• [27] P. Arabas and M. Karpowicz, “Server power consumption: measurements and modeling with MSRs”, in Challenges in Automation, Robotics and Measurement Techniques, R. Szewczyk, C. Zieliński, and M. Kaliczyńska, Eds., Advances in Intelligent Systems and Computing, vol. 440, pp. 233-244. Springer, 2016.
• [28] M. P. Karpowicz and P. Arabas, “Preliminary results on the Linux libpcap model identification., in Proc. 20th Int. Conf. Methods & Models Autom Robot MMAR 2015, Międzyzdroje, Poland, 2015, pp. 1056-1061.
• [29] SNIA Emerald, SNIA Emerald Power Efficiency Measurement Specification [Online]. Available: www.snia.org
• [30] Storage Performance Council (SPC), Storage Performance Council SPC Benchmark 2/Energy Extension [Online]. Available: www.storageperformance.org
• [31] D. Hackenberg et al., “Power measurement techniques on standard compute nodes: A quantitative comparison”, in Proc. IEEE Int. Symp. on Perform. Anal. Syst. & Software ISPASS 2013, Austin, TX, USA, 2013, pp. 194-204.
• [32] J. Mair, D. Eyers, Z. Huang, and H. Zhang, “Myths in power estimation with performance monitoring counters”, Sustain. Comput.: Inform. & Syst., vol. 4, no. 2, pp. 83-93, 2014.
• [33] M. E. Mehdi Diouri et al., “Assessing power monitoring approaches for energy and power analysis of computers”, Sustainable Comput.: Informatics and Syst., vol. 4, no. 2, pp. 68-82, 2014.
• [34] Intel Intelligent Power Node Manager [Online]. Available: www.intel.com
• [35] D. Hackenberg et al., “HDEEM: high definition energy efficiency monitoring”, in Proc. IEEE Energy Efficient Supercomput. Worksh. SC14, New Orleans, LA, USA, 2014, pp. 1-10.
• [36] M. F. Dolz, M. R. Heidari, M. Kuhn, T. Ludwig, and G. Fabregat, “ARDUPOWER: A low-cost wattmeter to improve energy efficiency of HPC applications”, in Proc. 6th Int. Green Comput. Conf. & Sustain. Comput. Conf. IGSC 2015, Las Vegas, NV, USA, 2015, pp. 1-8.
• [37] Intel IA-64 and IA-32 Architectures Software Developer's Manual [Online]. Available: www.intel.com
• [38] NVML API Reference Manual, 2012 [Online]. Available: http://developer.nvidia.com
• [39] T. Ilsche, D. Hackenberg, S. Graul, R. Schöne, and J. Schuchart, “Power measurements for compute nodes: improving sampling rates, granularity and accuracy”, in Proc. 6th Int. Green Comput. Conf. and Sustain. Comput. Conf. IGSC 2015, Las Vegas, NV, USA, 2015.
• [40] V. M. Weaver, M. Johnson, K. Kasichayanula, J. Ralph, P. Luszczek, D. Terpstra, and S. Moore, “Measuring energy and power with PAPI”, in Proc. 41st Int. Conf. Parallel Proces. Worksh. ICPPW 2012, Pittsburgh, PA, USA, 2012, pp. 262-268.
• [41] L. Taniça, A. Ilic, P. Tomás, and L. Sousa, “Schedmon: A performance and energy monitoring tool for modern multi-cores”, in Euro-Par 2014: Parallel Processing Workshops, LNCS, vol. 8806, pp. 230.241. Springer, 2014.
• [42] Performance Application Programming Interface (PAPI) [Online]. Available: icl.cs.utk.edu/papi
• [43] Unofficial Linux Perf. Events Performance Counter Weg Page [Online]. Available: web.eece.maine.edu/~vweaver/projects/perf_events
• [44] M. Gerndt, E. César, and S. Benkner, Eds., Automatic Tuning of HPC Applications. Shaker Verlag, 2015.
• [45] M. P. Karpowicz, “Energy-efficient CPU frequency control for the Linux system”, Concur. & Computat.: Pract. & Exper., vol. 28, no. 2, pp. 420-437, 2016 (cpe.3476).
• [46] D. Brandewie, “Intel P-state driver” [Online]. Available: www.kernel.org/doc/
• [47] M. P. Karpowicz, P. Arabas, and E. Niewiadomska-Szynkiewicz, „Design and implementation of energy-aware application-specific CPU frequency governors for the heterogeneous distributed computing systems”, Future Generation Computer Systems, available online, 2016 (doi: 10.1016/j.future.2016.05.011).