Skip to main content

Advertisement

SpringerLink
Log in

Power-aware throughput control for containerized relational operation

  • Regular Paper
  • Published:
CCF Transactions on High Performance Computing Aims and scope Submit manuscript

Abstract

As the high performance computing (HPC) moves towards AI-driven applications, these AIHPC applications demand steady and high-throughput relational data acquisition from data sites, which gradually dominates the whole computing process. In order to provide an energy-efficient service in a multi-tenant environment, we have to reduce the power consumption from serving relational data, such that the energy cost can be best amortized. While modern hardware provide multiple power modes, these power performance tradeoffs cannot be directly applied to data services due to the lack of understanding from the operation behavior under different power modes. To address this challenge, we provide an system identification study on learning the power behavior under serving database operations under different hardware modes, and propose a Control framework for Relational Operations to save Power. In contrast to today’s heuristic-based power tuning techniques, our solution achieves the goal via two facilities: (1) a control-theory based controller design that minimizes overshoot and guarantees the minimum settling time, thus control accuracy and system stability; (2) a fuzzy classifier inside database engine that helps to understand software behavior, in order to tune the sensitivity of the whole system control. We prototyped Crop as wrapping these functions in a container hierarchy, and evaluate it with workloads generated from various database benchmarks. The results show that Crop achieves up to 51.3% additional energy savings despite runtime workload dynamics and model errors, as compared to other competing methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others

Notes

  1. We use the active power of the whole system for the measurement throughout this paper. Any power data, if without specification, is the active power of the system.

  2. Details of the experiment setup can be found in Sect. 6.

  3. Sensitivity is defined as the change in performance in response to changes in CPU frequency as \((\frac{throughout}{CPU frequency})\)

  4. The value of \(\beta\) needs to be calibrated when Crop is applied to a different system environment.

  5. The performance overshoot is measured by Pmax/Rs, where \(P_{max}\) is the maximum performance and \(R_s\) is the set point.

References

  • Bernstein, D.: Containers and cloud: from lxc to docker to kubernetes. IEEE Cloud Comput. 1(3), 81–84 (2014)

    Article  Google Scholar 

  • Chen, M., Wang, X., Gunasekaran, R., Qi, H., Shankar, M.: Control-based real-time metadata matching for information dissemination. In: 2008 14th IEEE International Conference on embedded and real-time computing systems and applications, pp. 133–142 (2008)

  • Conci, A., Kubrusly, C.: Distance between sets-a survey (2018). arXiv:1808.02574

  • Council, T.P.P.: Transaction processing performance council (2005). http://www.tpc.org

  • Deng, Q., Meisner, D., Ramos, L., Wenisch, T.F., Bianchini, R.: MemScale: active low-power modes for main memory. SIGPLAN Not. 46(3), 225–238 (2011)

    Article  Google Scholar 

  • Franklin, G.F., Powell, J.D., Workman, M.L.: Digital control of dynamic systems. Addison and Wesley (1990)

  • Graefe, G.: Database servers tailored to improve energy efficiency. In: Proceedings of the 2008 EDBT workshop on Software engineering for tailor-made data management—SETMDM ’08. ACM Press, Nantes (2008)

  • Harizopoulos, S., Shah, M., Meza, J., Ranganathan, P.: Energy efficiency: the new holy grail of data management systems research (2009). arXiv:0909.1784 [cs]

  • Hellerstein, J.L., Diao, Y., Parekh, S., Tilbury, D.M.: Feedback control of computing systems (2004)

  • Hirst, J.M., Miller, J.R., Kaplan, B.A., Reed, D.D.: Watts up? Pro AC power meter for automated energy recording: a product review. Behav. Anal. Pract. 6(1), 82–95 (2013)

    Article  Google Scholar 

  • Holland, D., Zhang, W.: Distributing an SQL query over a cluster of containers. In: 2019 IEEE 12th International Conference on cloud computing (CLOUD), pp. 457–464. IEEE. IEEE (2019)

  • Jain, A.K., Murty, M.N., Flynn, P.J.: Data clustering: a review. ACM Comput. Surv. 31(3), 264–323 (1999)

    Article  Google Scholar 

  • Kang, K.-D., Oh, J., Son, S.H.: Chronos: feedback control of a real database system performance. In: 28th IEEE International real-time systems symposium (RTSS 2007), pp. 267–276 (2007)

  • Kunjir, M., Birwa, P.K., Haritsa, J.R.: Peak power plays in database engines. In: Proceedings of the 15th International conference on extending database technology—EDBT ’12. ACM Press, Berlin (2012)

  • Lang, W., Kandhan, R., Patel, J.M.: Rethinking query processing for energy efficiency: slowing down to win the race. IEEE Data Eng. Bull. 34(1), 12–23 (2011)

    Google Scholar 

  • Lefurgy, C., Wang, X., Ware, M.: Server-level power control. In: Fourth International conference on autonomic computing (ICAC’07) (2007)

  • Mahajan, D., Blakeney, C., Zong, Z.: Improving the energy efficiency of relational and nosql databases via query optimizations. Sustain. Comput Inf. Syst. 22, 120–133 (2019)

    Google Scholar 

  • Meisner, D., Wenisch, T.F.: DreamWeaver: architectural support for deep sleep. SIGPLAN Not. 47(4), 313–324 (2012)

    Article  Google Scholar 

  • Meisner, D., Sadler, C.M., Barroso, L.A., Weber, W.-D., Wenisch, T.F.: Power management of online data-intensive services. In: Proceeding of the 38th annual international symposium on Computer architecture—ISCA ’11. ACM Press, San Jose (2011)

  • Melhem, R., Mosse, D., Elnozahy, E.: The interplay of power management and fault recovery in real-time systems. IEEE Trans. Comput. 53(2), 217–231 (2004)

    Article  Google Scholar 

  • Padala, P., Hou, K.-Y., Shin, K.G., Zhu, X., Uysal, M., Wang, Z., Singhal, S., Merchant, A.: Automated control of multiple virtualized resources. In: Proceedings of the fourth ACM European conference on Computer systems—EuroSys ’09. ACM Press, Nuremberg (2009)

  • Poess, M., Nambiar, R.O.: Energy cost, the key challenge of today’s data centers: a power consumption analysis of TPC-C results. Proc. VLDB Endow. 1(2), 1229–1240 (2008)

    Article  Google Scholar 

  • Poess, M., Nambiar, R.O.: Tuning servers, storage and database for energy efficient data warehouses. In: 2010 IEEE 26th International Conference on Data Engineering (ICDE 2010), pp. 1006–1017 (2010)

  • Rareshide, M.: Power in the Data Center and its Cost Across the U.S., Technical report, Site Selection Group (2020)

  • Seybold, D., Hauser, C.B., Eisenhart, G., Volpert, S., Domaschka, J.: The impact of the storage tier: a baseline performance analysis of containerized DBMS. In: European Conference on Parallel Processing, pp. 93–105. Springer, New York (2018)

  • Shehabi, A., Smith, S., Sartor, D., Brown, R., Herrlin, M., Koomey, J., Masanet, E., Horner, N., Azevedo, I., Lintner, W.: United States Data Center Energy Usage Report, Technical Report LBNL-1005775, Lawrence Berkeley National Lab. (LBNL), Berkeley (2016). https://doi.org/10.2172/1372902. https://www.osti.gov/biblio/1372902/

  • Takagi, T., Sugeno, M.: Fuzzy identification of systems and its applications to modeling and control. IEEE Trans. Syst. Man Cybern. SMC 15(1), 116–132 (1985)

  • Tsirogiannis, D., Harizopoulos, S., Shah, M.A.: Analyzing the energy efficiency of a database server. In: Proceedings of the 2010 international conference on Management of data—SIGMOD ’10. ACM Press, Indianapolis (2010)

  • Tu, Y.-C., Liu, S., Prabhakar, S., Yao, B.: Load shedding in stream databases: a control-based approach (2006)

  • Wang, L., Xu, J., Zhao, M., Tu, Y., Fortes, J.A.B., Management, F.M.B.R., for Virtualized Database Systems. In: IEEE 19th Annual International Symposium on modelling, analysis, and simulation of computer and telecommunication systems, pp. 32–42 (2011)

  • Wang, X., Chen, M., Lefurgy, C., Keller, T.W.: SHIP: scalable hierarchical power control for large-scale data centers. In: 2009 18th International Conference on Parallel Architectures and Compilation Techniques, pp. 91–100 (2009)

  • Wang, Y., Wang, X., Chen, M., Zhu, X.: PARTIC: power-aware response time control for virtualized web servers. IEEE Trans. Parallel Distrib. Syst. 22(2), 323–336 (2011)

    Article  Google Scholar 

  • Wu, Q., Juang, P., Martonosi, M., Peh, L.-S., Clark, D.W.: Formal control techniques for power-performance management. IEEE Micro 25(5), 52–62 (2005)

    Article  Google Scholar 

  • Xu, Z., Tu, Y.-C., Wang, X.: Exploring power-performance tradeoffs in database systems. In: 2010 IEEE 26th International Conference on Data Engineering (ICDE 2010), pp. 485–496 (2010)

  • Xu, Z., Wang, X., Tu, Y.-c.: Power-aware throughput control for database management systems. In: Proceedings of the 10th International Conference on Autonomic Computing (ICAC 13), pp. 315–324 (2013)

  • York, D.G., Adelman, J., Anderson, J., et al.: The Sloan digital sky survey: technical summary. Astron. J. 120(3), 1579–1587 (2000)

    Article  Google Scholar 

  • Zhan, C., Su, M., Wei, C., Peng, X., Lin, L., Wang, S., Chen, Z., Li, F., Pan, Y., Zheng, F., et al.: Analyticdb: real-time olap database system at Alibaba cloud. Proc. VLDB Endow. 12(12), 2059–2070 (2019)

    Article  Google Scholar 

Download references

Acknowledgements

This work is supported by Core Electronics High End General Chips and Infrastructural Software funding No. 2018ZX 01035-101, Young Scientists Fund No. 61702250, Major Research Plan No. 2018YFB14043033, and research grant CARCHB202017. We would like to express our gratitude to all those who helped us during the writing of this paper. Also, we would like to thank all anonymous reviewers for their insightful comments.

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, Nanchang University, Nanchang, China

    Zichen Xu, Ao Cui & Shasha Wang

  2. China Development Bank Inner Mongolia Branch, Hohhot, China

    Gele Bai

Authors
  1. Zichen Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gele Bai
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ao Cui
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Shasha Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Zichen Xu.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xu, Z., Bai, G., Cui, A. et al. Power-aware throughput control for containerized relational operation. CCF Trans. HPC 3, 70–84 (2021). https://doi.org/10.1007/s42514-020-00061-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s42514-020-00061-6

Keywords

Search

Navigation