异步I/O连接算法的峰值功率建模

doi:10.3969/j.issn.0372-2112.2019.06.006

PDF(1072 KB)

电子学报 ›› 2019, Vol. 47 ›› Issue (6) : 1220-1229. DOI: 10.3969/j.issn.0372-2112.2019.06.006

学术论文

异步I/O连接算法的峰值功率建模

杨良怀, 戚加欣, 范玉雷, 张璐, 梅建萍

作者信息 +

Peak Power Modeling for Join Algorithms with Asynchronous I/Os

YANG Liang-huai, QI Jia-xin, FAN Yu-lei, ZHANG Lu, MEI Jian-ping

Author information +

文章历史 +

摘要

峰值功率是影响数据中心能效的一个重要因素.本文提出一种功率感知数据库系统中连接算法的峰值功率估算方法，非运行时峰值功率的估算的挑战在于没有运行时的系统信息作为模型的输入.为克服估算困难，提出使用CPU密集度作为CPU功耗指示量，理论上分析了异步I/O连接算法在峰值功率发生阶段的特性，通过模拟连接算法峰值功率发生阶段算法行为来估算该阶段最大CPU密集度，通过CPU密集度与CPU功率的内在联系建立异步I/O连接算法的峰值功率预测模型.实验对数据库系统中采用异步I/O机制的四个连接算法时模型准确性进行了验证，结果表明所提预测方法具有较好的预测准确性，平均相对误差低于7%.

Abstract

Peak power is a critical factor on the power consumption of a data center.This paper proposes a peak power estimation method to predict the peak power to join operations in DBMS.The challenge of non-runtime peak power estimation lies in that there is no runtime system information to use for model construction.To overcome this issue,this paper uses CPU-boundedness as the proxy of CPU power consumption and analyzes the characteristics of the peak power occurring stage of join algorithms with async I/O in theory.By simulating the behavior of this stage,we estimate the maximal CPU-boundedness of join algorithms.By examining the relationship between the CPU-boundedness and CPU power,the peak power models of join algorithms under different CPU execution frequency are hence constructed.Experiments validate the effectiveness of the proposed models on three typical join algorithms in DBMS with async I/O techniques.Results showed that our proposed methods had good accuracy with the average relative error less than 7%.

导出引用

杨良怀, 戚加欣, 范玉雷, 张璐, 梅建萍. 异步I/O连接算法的峰值功率建模[J]. 电子学报, 2019, 47(6): 1220-1229. https://doi.org/10.3969/j.issn.0372-2112.2019.06.006

YANG Liang-huai, QI Jia-xin, FAN Yu-lei, ZHANG Lu, MEI Jian-ping. Peak Power Modeling for Join Algorithms with Asynchronous I/Os[J]. Acta Electronica Sinica, 2019, 47(6): 1220-1229. https://doi.org/10.3969/j.issn.0372-2112.2019.06.006

中图分类号： TP301

参考文献

[1] AGRAWAL R,AILAMAKI A,BERNSTEIN P A,et al.The claremont report on database research[J].ACM Sigmod Record,2008,37(3):9-19.
[2] WANG J,FENG L,XUE W,et al.A survey on energy-efficient data management[J].ACM SIGMOD Record,2011,40(2):17-23.
[3] 金培权,杨濮源,陈恺萌,等.DBPower:面向绿色数据库系统的能耗有效性测试[J].计算机研究与发展,2011(S3):410-413. Jin P Q,Yang P Y,Chen K M,et al.DBPower:measuring energy efficiency for green database systems[J].Journal of Computer Research and Development,2011(S3):410-413.(in Chinese)
[4] LANG W,KANDHAN R,PATEL J M.Rethinking query processing for energy efficiency:Slowing down to win the race[J].IEEE Data Eng Bull,2011,34(1):12-23.
[5] LANG W,HARIZOPOULOS S,PATEL J M,et al.Towards energy-efficient database cluster design[J].Proceedings of the VLDB Endowment,2012,5(11):1684-1695.
[6] NOLL S,FUNKE H,TEUBNER J.Energy efficiency in main-memory databases[J].Datenbank-Spektrum,2017,17(3):223-232.
[7] XU Z,TU Y C,WANG X.Exploring power-performance tradeoffs in database systems[A].Proceedings of 2010 IEEE 26th International Conference on Data Engineering(ICDE 2010)[C].Long Beach:IEEE,2010.485-496.
[8] XU Z,TU Y C,WANG X.Dynamic energy estimation of query plans in database systems[A].Proceedings of 2013 IEEE 33rd International Conference on Distributed Computing Systems[C].Philadelphia:IEEE,2013.83-92.
[9] XU Z,TU Y C,WANG X.Online energy estimation of relational operations in database systems[J].IEEE transactions on computers,2015,64(11):3223-3236.
[10] KORKMAZ M,KARSTEN M,SALEM K,SALIHOGLU S.Workload-aware CPU performance scaling for transactional database systems[A].Proceedings of Special Interest Group on Management of Data[C].Houston:ACM,2018.291-306.
[11] TSIROGIANNIS D,HARIZOPOULOS S,SHAH M A.Analyzing the energy efficiency of a database server[A].Proceedings of the 2010 ACM SIGMOD International Conference on Management of data[C].Indianapolis:ACM,2010.231-242.
[12] KUNJIR M,BIRWA P K,HARITSA J R.Peak power plays in database engines[A].Proceedings of the 15th International Conference on Extending Database Technology[C].Berlin:ACM,2012.444-455.
[13] YANG L H,ZHAO Y,FAN Y,et al.Peak power modeling for join algorithms in DBMS[J].Journal of Computer and System Sciences,2015,81(3):599-614.
[14] CHOI K,SOMA R,PEDRAM M.Fine-grained dynamic voltage and frequency scaling for precise energy and performance tradeoff based on the ratio of off-chip access to on-chip computation times[J].IEEE Transactions on Computer-Aided Design of Integrated Circuits And Systems,2005,24(1):18-28.