DOI QR코드

DOI QR Code

Performance Improvement of an Energy Efficient Cluster Management Based on Autonomous Learning

자율학습기반의 에너지 효율적인 클러스터 관리에서의 성능 개선

  • 조성철 (숭실대학교 전자공학과) ;
  • 정규식 (숭실대학교 스마트시스템소프트웨어학과)
  • Received : 2015.07.06
  • Accepted : 2015.08.27
  • Published : 2015.11.30

Abstract

Energy aware server clusters aim to reduce power consumption at maximum while keeping QoS(quality of service) compared to energy non-aware server clusters. They adjust the power mode of each server in a fixed or variable time interval to activate only the minimum number of servers needed to handle current user requests. Previous studies on energy aware server cluster put efforts to reduce power consumption or heat dissipation, but they do not consider energy efficiency well. In this paper, we propose an energy efficient cluster management method to improve not only performance per watt but also QoS of the existing server power mode control method based on autonomous learning. Our proposed method is to adjust server power mode based on a hybrid approach of autonomous learning method with multi level thresholds and power consumption prediction method. Autonomous learning method with multi level thresholds is applied under normal load situation whereas power consumption prediction method is applied under abnormal load situation. The decision on whether current load is normal or abnormal depends on the ratio of the number of current user requests over the average number of user requests during recent past few minutes. Also, a dynamic shutdown method is additionally applied to shorten the time delay to make servers off. We performed experiments with a cluster of 16 servers using three different kinds of load patterns. The multi-threshold based learning method with prediction and dynamic shutdown shows the best result in terms of normalized QoS and performance per watt (valid responses). For banking load pattern, real load pattern, and virtual load pattern, the numbers of good response per watt in the proposed method increase by 1.66%, 2.9% and 3.84%, respectively, whereas QoS in the proposed method increase by 0.45%, 1.33% and 8.82%, respectively, compared to those in the existing autonomous learning method with single level threshold.

에너지 절감형 서버 클러스터에서는 에너지 절감을 고려하지 않는 기존 서버 클러스터에 비해 서비스 품질을 보장하면서 전력소비를 절감하는 것을 목표로 하며, 현재의 부하를 처리하는 데 필요한 최소수의 서버들만 ON 하도록 고정 주기 또는 가변 주기로 서버들의 전원모드를 조정한다. 이에 대한 기존 연구들은 전력을 절감하거나 열을 낮추는데 노력해왔지만 에너지 효율성을 잘 고려하지 못했다. 본 논문에서는 기존 자율학습기반의 서버 전원 모드 제어 방법의 단위전력당 성능과 QoS를 높이기 위한 에너지 효율적인 클러스터 관리기법을 제안한다. 제안 방법은 다중임계기반의 자율학습 방법과 전력소모 예측 방법을 결합한 서버 전원 모드 제어이다. 일반적인 부하 상황에서는 다중임계 학습기반의 서버 전원 모드 제어를 적용하고, 급변하는 부하 상황에서는 예측기반의 서버 전원 모드 제어가 적용된다. 일반적 상황과 급변하는 상황의 구별은 현재의 사용자 요청과 관찰된 과거 몇 분의 사용자 요청의 비율에 따라 이루어진다. 또한, 동적종료 기법을 추가로 적용해 서버가 OFF 하는 데 소요되는 시간을 단축한다. 제안 방법은 16대 서버로 구성된 클러스터 환경에서 3가지 부하 패턴을 이용하여 실험을 수행한다. 다중임계 학습, 예측, 동적종료를 함께 이용한 실험에서 단위전력당 성능(유효응답 수)과 표준화된 QoS 측면에서 가장 우수한 결과를 보여준다. 제안하는 방법과 파라미터 로드된 단일임계 학습을 비교할 때 뱅킹 부하패턴, 실제 부하패턴, 가상 부하패턴에서 단위전력당 유효응답 수가 각각 1.66%, 2.9%, 3.84% 향상되고, QoS 관점에서는 각각 0.45%, 1.33%, 8.82% 향상되었다.

Keywords

References

  1. Fanxin Kong and Xue Liu, "A Survey on Green-Energy- Aware Power Management for Datacenters," in ACM Computing Surveys(CSUR), 2014.
  2. Chenguang Liu, Jianzhong Huang, Qiang Cao, Shenggang Wan, and Changsheng Xie, "Evaluating Energy and Performance for Server-Class Hardware Configurations," 6th IEEE International Conference on Networking, Architecture and Storage, 2011.
  3. J. Mair, K. Leung, Z. Huang, "Metrics and task scheduling policies for energy saving in multicore computers," 11th IEEE/ACM International Conference on Grid Computing (GRID), 2010.
  4. G. Chen et. al., "Energy-Aware Server Provisioning and Load Dispatching for Connection-Intensive Internet Services," NSDI'08 Proceedings of the 5th USENIX Symposium on Networked Systems Design and Implementation, 2008.
  5. A Krioukov, et al., "NapSAC: design and implementation of a power-proportional web cluster," ACM SIGCOMM computer communication overview, 2011.
  6. Abdul Hameed, Alireza Khoshkbarforoushha, Rajiv Ranjan, Prem Prakash Jayaraman, Joanna Kolodziej, Pavan Balaji, Sherali Zeadally, Qutaibah Marwan Malluhi, Nikos Tziritas, Abhinav Vishnu, Samee U. Khan, and Albert Zomaya, "A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems," Springer Computing, Jun., 2014.
  7. Anton Beloglazov, Rajkumar Buyya, Young Choon Lee, and Albert Zomaya, "A Taxonomy and Survey of Energy- Efficient Data Centers and Cloud Computing Systems," The University of Melbourne, Australia, The University of Sydney, Australia, 2010.
  8. Sungchul Cho, Hukeun Kwak, and Kyusik Chung, "An Energy Efficient Cluster Management Method based on Autonomous Learning in a Server Cluster Environment," KIPS Transactions on Computer and Communication Systems, Vol.4, No.6, pp.185-196, 2015. https://doi.org/10.3745/KTCCS.2015.4.6.185
  9. Taejune Ahn, Sungchul Cho, Seokkoo Kim, Kyongho Chun, and Kyusik Chung, "A Flexible Multi-Threshold Based Control of Server Power Mode for Handling Rapidly Changing Loads in an Energy Aware Server Cluster," KIPS Transactions on Computer and Communication Systems, Vol.3, No.9, pp.279-292, 2014. https://doi.org/10.3745/KTCCS.2014.3.9.279
  10. Hoyeon Kim, Chihwan Ham, Hukeun Kwak, and Kyusik Chung, "Dynamic Shutdown of Server Power Mode Control for Saving Energy in a Server Cluster Environment," KIPS Transactions on Computer and Communication Systems, 2013.
  11. LVS(Linux Virtual Server) [Internet], http://www. linuxvirt ualserver.org.
  12. Sungchul Cho, Sanha Kang, Heungsik Moon, Hukeun Kwak, and Kyusik Chung, "Prediction of Power Consumption for Improving QoS in an Energy Saving Server Cluster Environment," KIPS Transactions on Computer and Communication Systems, Vol.4, No.2, pp.47-56, 2015. https://doi.org/10.3745/KTCCS.2015.4.2.47
  13. Hoyeon Kim, Chihwan Ham, Hukeun Kwak, Hulung Kwon, Youngjoung Kim, Kyusik Chung, "A Dynamic Server Power Mode Control for Saving Energy in a Server Cluster Environment," The KIPS Transactions: PartC, Vol.19, No.3, pp.135-144, 2012.
  14. SPECweb [Internet], http://www.spec.org/benchmarks.html/.
  15. Apache [Internet], http://www.apache.org/.
  16. InternetTrend [Internet], http://www.internettrend.co.kr.
  17. Direct Routing [Internet], http://www.linuxvirtualserver.org /VS-DRouting.html.
  18. H. Kwak, A. Sohn and K. Chung, "Autonomous Learning of Load and Traffic Patterns to Improve Cluster Utilization," Cluster Computing, Vol.14, Issue 4, Dec., 2011.
  19. Hukeun Kwak, Kyusik Chung, Hyung Won Choi, and Andrew Sohn "Enabling Scalabe Cloud Infrastructure using Autonomous VM Migration," 2012 IEEE 14th International Conference on High Performance Computing and Communications, 2012.