Evaluation on performance and energy eciency of distributed computing systems

Evaluation on Performance and Energy Efficiency of Distributed Computing Systems Ph.D Dissertation by Tran Thi Xuan (MSc) Supervised by Prof Do Van Tien (DSc) Department of Networked Systems and Services Budapest University of Technology and Economics Hungary, 2020 Abstract The increasing usage of distributed computing systems to serve the growing demand for scientific computation and big data processing comes with the drastic growth of energy consumption in computing clusters Therefore, optimizing the energy consumption of computational clusters has become more crucial than ever The dissertation summarizes a study on the resource allocation problem in distributed systems, motivated by a need of taking into account different resource characteristics and dynamic power management (DPM) techniques First, a generalized model of computational clusters built from heterogeneous types of COTS servers has been introduced to study the resource-aware scheduling A set of scheduling heuristics that consider servers’ performance and power consumption characteristics and the organization of waiting buffers have been investigated We show that the buffering schemes play an important role in ensuring the quality of service parameters in terms of the waiting time and the response time experienced by arriving jobs Moreover, energy efficiency characteristic based scheduling can conserve the system energy and high performance priority based policy yields the best performance Second, new real-time measurement based scheduling algorithms to achieve a trade-off between energy efficiency and the performance capability of computational clusters have been proposed in the thesis Numerical results show that the proposed algorithms attain a balance between the job execution time and energy efficiency Third, the impact of dynamic power management (DPM) in computing systems built from multicore processors has been investigated Numerical results point out that DPM in the core level of processors can play a role in saving energy consumption A resourceaware scheduling solution has been proposed to achieve energy-efficient processing of parallel tasks in multicore systems Obtained results indicate that the proposal reduces energy consumption significantly in comparison to random allocation Last, the energy inefficiency in an ordinary big data scheduler-Hadoop YARN has been investigated Since the resource allocation policy in the Hadoop YARN cluster is data-aware (i.e the allocation strongly depends on the locations of data splits in Hadoop Distributed File System-HDFS), a new data placement scheme for HDFS was proposed to achieve energy efficiency when MapReduce tasks are processed by the cluster Compared to the existing HDFS data layout scheme, the proposal yields above 50% reduction in energy consumption at a small expense of ≈6% increase in job execution time I, the undersigned Tran Thi Xuan, hereby state that I have written this doctoral dissertation myself, and I have used only the sources given in it I have clearly marked all the parts taken from other sources either word for word or reworded but with the same contents, indicating their sources The reviews of the dissertation and the report of the thesis discussion are available at the Dean’s Office of the Electrical Engineering and Informatics Faculty, Budapest University of Technology and Economics Budapest, February 17, 2020 Tran Thi Xuan Acknowledgements I would like to thank all people who have provided invaluable assistance during my study towards the Ph.D degree I would like to express my sincere gratitude to Prof Dr Do Van Tien for his intensive supervision Prof Dr Do Van Tien has guided me on the direction of my research at preliminary time Without his continuous supervision and straight criticisms, I could not accomplish this study and achieve PhD degree I deeply thank Dr Do Hoai Nam, a senior researcher in Analysis, Design and Development of ICT systems laboratory at our department, for his work cooperation and enthusiastic support through my research All members of the Analysis, Design and Development of ICT systems laboratory, other PhD students, and the university staffs are acknowledged Finally, I dedicate my hearty thankfulness to my husband and son Le Linh Bang and Le Minh Anh for their love and encouragement I am also grateful to all family members and friends who have supported me throughout Contents Abstract Acknowledgement List of Figures 14 List of Tables 16 Introduction 17 A generalized model of heterogeneous computing clusters for investigation of scheduling schemes 19 2.1 Introduction 20 2.2 A generalized cluster model and Scheduling algorithms 21 2.3 2.4 2.2.1 Ranking of servers 22 2.2.2 Scheduling algorithms 23 2.2.3 Performance measures and energy metrics 27 Simulation Inputs and Numerical Results 29 2.3.1 Input parameters 29 2.3.2 Numerical results 31 Conclusion 39 CONTENTS New algorithms for balancing energy consumption and performance in computational clusters 40 3.1 Introduction 41 3.2 System description and proposed scheduling algorithms 42 3.2.1 3.3 3.4 Scheduling algorithms 42 Numerical Results 45 3.3.1 The parameters of a computational cluster 46 3.3.2 Job balance 47 3.3.3 System metrics 48 3.3.4 Impacts of DVFS 51 3.3.5 Evaluations with workload traces as input data 52 Conclusion 55 Impact of Dynamic power management techniques in computing systems of multicore processors 56 4.1 Introduction 57 4.2 Dynamic Power Management practices 58 4.3 System descriptions and operation scenarios 59 4.4 4.5 4.3.1 Job assignment scenarios 61 4.3.2 Performance and energy metrics 63 Evaluation on the impact of DPM 65 4.4.1 Simulation inputs 65 4.4.2 Analysis of obtained results 67 A proposal of Resource-aware scheduling algorithm 72 4.5.1 The proposed policy 73 CHAPTER A NEW DATA LAYOUT SCHEME FOR HDFS 5.5.2.2 101 Response times and computation times Figures 5.10 - 5.11 plot the averages of computation time and response time per job versus arrival rate λ in the scenario with Uniform distribution It can be observed that the averages of computation time and response time per job are slightly increased with our proposal Particularly, a configuration with the default HDFS layout takes an average of 282 seconds, while an arrangement with the proposed HDFS algorithm takes approximately 300 seconds to finish the service of a job Figure 5.12 depicts the average response time when data blocks are normally distributed Our proposal results in a slight increase in the average response time of jobs Furthermore, we plot the empirical cumulative distribution function (ECDF) of the execution time of jobs in Figure 5.13 for data sizes of a Normal distribution with the arrival rate of 0.035 It is observed that the computation time of a MapReduce job falls into the boundaries of [74.63, 467.48] s Mean service time s 500 default HDFS biased HDFS 400 300 200 100 0.035 0.052 0.069 0.087 Arrival rate Figure 5.10: Mean computation time - with a Uniform dist Mean response time s 500 default HDFS proposed layout 400 300 200 100 0.035 0.052 0.069 0.087 Arrival rate Figure 5.11: Mean response time - with a Uniform dist CHAPTER A NEW DATA LAYOUT SCHEME FOR HDFS Mean response time s 500 default HDFS proposed layout 400 300 200 100 0.035 0.052 0.069 0.087 Arrival rate Figure 5.12: Mean response time - with a Normal dist Figure 5.13: ECDF of the execution time of jobs - with a Normal dist 102 CHAPTER A NEW DATA LAYOUT SCHEME FOR HDFS 5.6 103 Conclusion We have proposed a layout scheme that gains the application of an energy management procedure in a resource-heterogeneous Hadoop cluster Our algorithm sorts servers into three sets according to computed rankings of the characteristics of the cluster servers The proposed layout algorithm places data blocks to the high-performance set and the energy-efficient set based on the data size, and keeps a partition of replicas of data blocks in inefficient servers Numerical results showed that our solution outperforms the default layout scheme in the term of the energy consumption The proposed data layout for HDFS can be implemented in Hadoop clusters to reduce the energy consumptions In our future work, we may evaluate the performance of our proposed data layout with other job scheduling frameworks as Mesos [10] Chapter Summary This research investigated the job scheduling problem and energy consumption in various contexts of computational clusters The contributions can be summarized in three thesis groups in the following The first result group is a set of scheduling heuristics applicable for computational clusters of heterogeneous machines (Chapters 2-3) This result also emphasizes a guideline that is helpful to select an appropriate scheduling algorithm from a system operator’s perspective The second result group points out the possibility of energy savings in computing systems if dynamic power management (DPM) could be applied in cores of processors (Chapter 4) It suggests a prospective trend of processor design and manufacture to enable switching on/off cores dynamically and individually The proposed Resourceaware algorithm can be used straightforwardly in the field of resource management and allocation in computational clusters of multicore servers The third result provides the insight of how a big data job scheduler-Hadoop YARN allocates resources among tasks and how a data layout scheme affects on system performance in terms of energy consumption and job completion time This result is applicable in any context of processing datasets that are stored in Hadoop Distributed File System (HDFS) Moreover, the proposal of taking account of compute resources and a new data layout scheme should be considered in operating any big data processing system to attain energy efficiency 104 Own Publications Journal Papers [J1] Tien V Do, Binh T Vu, Xuan T Tran, and Anh P Nguyen A generalized model for investigating scheduling schemes in computational clusters Simulation Modelling Practice and Theory, 37(0):30–42, 2013 (Impact Factor = 2.426* ) [J2] Xuan, T T and Tien, V D and Binh, T V New algorithms for balancing an energy consumption and performance in computational clusters Journal of Computing and Informatics, 36(2):307–330, 2017 (Impact Factor = 0.524) [J3] Xuan, T T and Tien, V D and Chakka, R The Impact of Dynamic Power Management in Computational Clusters with Multi-Core Processors Journal of Scientific and Industrial Research (JSIR) , 75:339–343, June 2016 (Impact Factor = 0.735* ) [J4] Xuan, T T and Tien, V D and Csaba, R and Dosam, H A New Data Layout Scheme for Energy-Efficient MapReduce Processing Tasks Journal of Grid Computing, Feb 2018 (Impact Factor = 3.288*) * Impact Factor of 2018 105 Conference Papers [C1] N H Do, T Van Do,X Thi Tran, L Farkas, and C Rotter A scalable routing mechanism for stateful microservices, 20th Conference on Innovations in Clouds, Internet and Networks (ICIN),pages 72–78 March 2017 [C2] Xuan T Tran Resource-Aware Scheduling in Heterogeneous, Multi-core Clusters for Energy Efficiency, Advances in Information and Communication Technology: Proceedings of the International Conference, ICTA 2016, pages 520–529 Springer International Publishing, Cham, 2017 [C3] N H Do, T V Do, X T Tran, L Farkas, and C Rotter Data I/O provision for Spark applications in a Mesos cluster, 19th IEEE International Conference on Innovation in Cloud Internet and Networking: ICIN 2016, pages 45–52 March 2016 [C4] Tran Thi Xuan and Tien Van Do Job Scheduling in a Computational Cluster with Multicore Processors, Advanced Computational Methods for Knowledge Engineering (ICSAMA 2016), vol 453, pages 75–84 Springer International Publishing, Cham, 2016 [C5] X T Tran, T V Do, N H Do, L Farkas, and C Rotter Provision of Disk I/O Guarantee for MapReduce Applications, 2015 IEEE Trustcom/BigDataSE/ISPA, volume 2, pages 161–166 Aug 2015 [C6] Xuan T Tran and Binh T Vu A New Approach for Buffering Space in Scheduling Unknown Service Time Jobs in a Computational Cluster with Awareness of Performance and Energy Consumption, Advanced Computational Methods for Knowledge Engineering (ICSAMA 2014), vol 282, pages 129–139 Springer International Publishing, Cham, 2014 106 Bibliography [1] Yeo, Chee Shin and Buyya, Rajkumar and Pourreza, Hossein and Eskicioglu, Rasit and Graham, Peter and Sommers, Frank Cluster Computing: High-Performance, HighAvailability, and High-Throughput Processing on a Network of Computers Handbook of Nature-Inspired and Innovative Computing: Integrating Classical Models with Emerging Technologies, pages 521–551, 2006 [2] Ejaz Ahmed, Ibrar Yaqoob, Ibrahim Abaker Targio Hashem, Imran Khan, Abdelmuttlib Ibrahim Abdalla Ahmed, Muhammad Imran, and Athanasios V Vasilakos The role of big data analytics in internet of things Computer Networks, 129:459 – 471, 2017 Special Issue on 5G Wireless Networks for IoT and Body Sensors [3] Dutta K Distributed Computing Technologies in Big Data Analytics Distributed Computing in Big Data Analytics Scalable Computing and Communications Springer, Cham,pages 57–82, 2017 [4] Albert Reuther, Chansup Byun, William Arcand, David Bestor, Bill Bergeron, Matthew Hubbell, Michael Jones, Peter Michaleas, Andrew Prout, Antonio Rosa, Jeremy Kepner Scalable System Scheduling for HPC and Big Data Journal of Parallel and Distributed Computing, Vol 111, January 2018, Pages 76–92, 2018 [5] Rajkumar Buyya High Performance Cluster Computing: Architectures and Systems, Vol Prentice Hall, 1999 [6] Tom White Hadoop: The Definitive Guide O’Reilly Media, Inc., 4th edition, 2015 [7] Apache Spark - Unified Analytics Engine for Big Data http://www.spark.apache org/ [8] Condor Team, University of Wisconsin-Madison Condor Version 8.8.0 Manual 2019 107 BIBLIOGRAPHY 108 [9] Robert L Henderson Job Scheduling Under the Portable Batch System Proceeding IPPS ’95 Proceedings of the Workshop on Job Scheduling Strategies for Parallel Processing, Pages 279–294, 1995 [10] Hindman, Benjamin and Konwinski, Andy and Zaharia, Matei and Ghodsi, Ali and Joseph, Anthony D and Katz, Randy and Shenker, Scott and Stoica, Ion Mesos: A Platform for Fine-grained Resource Sharing in the Data Center Proceedings of the 8th USENIX Conference on Networked Systems Design and Implementation, pages: 295–308, 2011 [11] Yang, Y., Liu, K., Chen, J., Liu, X., Yuan, D., Jin, H An algorithm in SwinDeW-C for scheduling transaction- intensive cost-constrained cloud workflows In IEEE Fourth International Conference on eScience, 374–375, 2008 [12] Van den Bossche, R., Vanmechelen, K., Broeckhove, J Online cost-efficient scheduling of deadline-constrained workloads on hybrid clouds Futur Gener Comput Syst., 29(4): 973–985, 2013 [13] Reig, G., Alonso, J., Guitart, J Prediction of job resource requirements for deadline schedulers to manage high-level SLAs on the cloud In 2010 9th IEEE International Sym- posium on Network Computing and Applications (NCA), 162–167, 2010 [14] L Benini, A Bogliolo, and G De Micheli A survey of design techniques for systemlevel dynamic power management Very Large Scale Integration (VLSI) Systems, IEEE Transactions on, 8(3):299 –316, june 2000 [15] B Callaghan, B Pawlowski, and P Staubach NFS Version Protocol Specification RFC 1813 (Informational), June 1995 [16] Jeffrey Dean and Sanjay Ghemawat MapReduce: Simplified Data Processing on Large Clusters Commun ACM, 51(1):107–113, January 2008 [17] Tien V Do, Binh T Vu, Xuan T Tran, and Anh P Nguyen A generalized model for investigating scheduling schemes in computational clusters Simulation Modelling Practice and Theory, 37(0):30–42, 2013 [18] Do, Tien Van and Rotter, Csaba Comparison of Scheduling Schemes for On-demand IaaS Requests J Syst Softw., 85(6):1400–1408, June 2012 [19] Brad Ellision and Lauri Minas Energy Efficiency for Information Technology: How to Reduce Power Consumption in Servers and Data Centers Intel press, 2009 BIBLIOGRAPHY 109 [20] Boliang Feng, Jiaheng Lu, Yongluan Zhou, and Nan Yang Energy Efficiency for MapReduce Workloads: An In-depth Study In Proceedings of the Twenty-Third Australasian Database Conference - Volume 124, ADC ’12, pages 61–70, Darlinghurst, Australia, Australia, 2012 Australian Computer Society, Inc [21] Paul Fishwick Simulation toolkit implement/simpack/implement.shtml http://www.cs.sunysb.edu/~algorith/ [22] Ian Foster What is the grid? a three point checklist 2002 [23] Ian Foster,C Kesselman, S Tuecke The anatomy of the Grid: Enabling scalable virtual organization The Intl Jrnl of High Performance Computing Applications, 15(3):200–222, 2001 [24] Anshul Gandhi, Mor Harchol-Balter, and Michael A Kozuch Are Sleep States Effective in Data Centers? In Proceedings of the 2012 International Green Computing Conference (IGCC), IGCC ’12, pages 1–10, Washington, DC, USA, 2012 IEEE Computer Society [25] Kyriaki Gkoutioudi and Helen D Karatza Multi-criteria job scheduling in grid using an accelerated genetic algorithm J Grid Comput., 10(2):311–323, 2012 [26] Goiri, Íđigo and Le, Kien and Nguyen, Thu D and Guitart, Jordi and Torres, Jordi and Bianchini, Ricardo GreenHadoop: Leveraging Green Energy in Data-processing Frameworks In Proceedings of the 7th ACM European Conference on Computer Systems, EuroSys ’12, pages 57–70, New York, NY, USA, 2012 ACM [27] Tyler Harter, Dhruba Borthakur, Siying Dong, Amitanand Aiyer, Liyin Tang, Andrea C Arpaci-Dusseau, and Remzi H Arpaci-Dusseau Analysis of HDFS Under HBase: A Facebook Messages Case Study In Proceedings of the 12th USENIX Conference on File and Storage Technologies, FAST’14, pages 199–212, Berkeley, CA, USA, 2014 USENIX Association [28] Y He, W.J Hsu, and C.E Leiserson Provably efficient online non-clairvoyant adaptive scheduling In Parallel and Distributed Processing Symposium, 2007 IPDPS 2007 IEEE International, pages –10, march 2007 [29] Y He, W.J Hsu, and C.E Leiserson Provably efficient online non-clairvoyant adaptive scheduling In Parallel and Distributed Processing Symposium, 2007 IPDPS 2007 IEEE International, pages –10, march 2007 BIBLIOGRAPHY 110 [30] C Hertel Implementing CIFS - The Common Internet File System Prentice Hall, 2003 [31] Ladislav Hluchý From grid to cloud computing In Huynh Quyet Thang and Dinh Khang Tran, editors, SoICT, page ACM, 2011 [32] Ladislav Hluchý, Nataliia Kussul, Andrii Shelestov, Serhiy Skakun, Oleksii M Kravchenko, Y Gripich, P Kopp, and E Lupian The data fusion grid infrastructure: Project objectives and achievements Computing and Informatics, 29(2):319– 334, 2010 [33] Rini T Kaushik and Milind Bhandarkar GreenHDFS: Towards an Energyconserving, Storage-efficient, Hybrid Hadoop Compute Cluster In Proceedings of the 2010 International Conference on Power Aware Computing and Systems, HotPower’10, pages 1–9, Berkeley, CA, USA, 2010 USENIX Association [34] Rini T Kaushik, Ludmila Cherkasova, Roy Campbell, and Klara Nahrstedt Lightning: Self-adaptive, Energy-conserving, Multi-zoned, Commodity Green Cloud Storage System In Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing, HPDC ’10, pages 332–335, New York, NY, USA, 2010 ACM [35] K.R Krish, A Anwar, and A.R Butt hatS: A Heterogeneity-Aware Tiered Storage for Hadoop In Cluster, Cloud and Grid Computing (CCGrid), 2014 14th IEEE/ACM International Symposium on, pages 502–511, May 2014 [36] Willis Lang and Jignesh M Patel Energy Management for MapReduce Clusters Proc VLDB Endow., 3(1-2):129–139, September 2010 [37] Charles Lefurgy, Karthick Rajamani, Freeman Rawson, Wes Felter, Michael Kistler, and Tom W Keller Energy management for commercial servers Computer, 36(12):39–48, December 2003 [38] Jacob Leverich and Christos Kozyrakis On the Energy (in)Efficiency of Hadoop Clusters SIGOPS Oper Syst Rev., 44(1):61–65, March 2010 [39] Bin Liao, Jiong Yu, Tao Zhang, Guo Binglei, Sun Hua, and Changtian Ying Energyefficient algorithms for distributed storage system based on block storage structure reconfiguration Journal of Network and Computer Applications , 48(0):71 – 86, 2015 BIBLIOGRAPHY 111 [40] Nitesh Maheshwari, Radheshyam Nanduri, and Vasudeva Varma Dynamic energy efficient data placement and cluster reconfiguration algorithm for MapReduce framework "Future Generation Computer Systems ", 28(1):119 – 127, 2012 [41] L Mashayekhy, M.M Nejad, D Grosu, Dajun Lu, and Weisong Shi Energy-Aware Scheduling of MapReduce Jobs In Big Data (BigData Congress), 2014 IEEE International Congress on, pages 32–39, June 2014 [42] Song Jie , He HongYan, Wang Zhi , Yu Ge, and Pierson Jean-Marc Modulo Based Data Placement Algorithm for Energy Consumption Optimization of MapReduce System Journal of Grid Computing, pages 32–39, June 2016 [43] Arun C Murthy, Vinod Kumar Vavilapalli, Doug Eadline, Joseph Niemiec, and Jeff Markham Apache Hadoop YARN: Moving Beyond MapReduce and Batch Processing with Apache Hadoop Addison-Wesley Professional, 1st edition, 2014 [44] Shekhar Srikantaiah, Aman Kansal, and Feng Zhao Energy aware consolidation for cloud computing Cluster Computing, 12:1–15, 2009 [45] Jiankang, D., Hongbo, W., Yangyang, L., Shiduan, C Virtual machine scheduling for improving energy efciency in IaaS cloud Communications, 11:1–12, 2014 [46] Tchernykh, Andrei and Ramírez, Juan Manuel and Avetisyan, Arutyun and Kuzjurin, Nikolai and Grushin, Dmitri and Zhuk, Sergey Two level job-scheduling strategies for a computational grid In Proceedings of the 6th international conference on Parallel Processing and Applied Mathematics, pages 774–781, 2006 [47] Beloglazov, A., Abawajy, J., Buyya, R Energy-aware resource allocation heuristics for efficient management of data centers for cloud computing Futur Gener Comput Syst., 28:755–768, 2012 [48] Yan Ma, Bin Gong, and Lida Zou Energy-efficient scheduling algorithm of task dependent graph on DVS-Unable cluster system In proceedings of 10th IEEE/ACM International Conference on Grid Computing (GRID), 2009 [49] Chia-Ming Wu, Ruay-Shiung Chang, Hsin-Yu Chan A green energy-efficient scheduling algorithm using the DVFS technique for cloud datacenters Future Generation Comp Syst, 37: 141-147, 2014 [50] Brian Guenter, Navendu Jain, and Charles Williams Managing cost, performance, and reliability tradeoffs for energy-aware server provisioning In Proceedings IEEE INFOCOM, 2011 BIBLIOGRAPHY 112 [51] Sergio Nesmachnow, BernabĂŠ Dorronsoro, JohnatanE Pecero, and Pascal Bouvry Energy-aware scheduling on multicore heterogeneous grid computing systems Journal of Grid Computing, pages 1–28, 2013 [52] Binh Minh Nguyen, Viet D Tran, and Ladislav Hluchý A generic development and deployment framework for cloud computing and distributed applications Computing and Informatics, 32(3):461–485, 2013 [53] NVIDIA Variable SMP - A Multi-Core CPU Architecture for Low Power and High Performance Technical report, NVIDIA’s Project Kal-El, NVIDIA Corporation, 2012 [54] Xuan Qi and Da-Kai Zhu Energy Efficient Block-Partitioned Multicore Processors for Parallel Applications Journal of Computer Science and Technology, 26(3):418–433, 2011 [55] Aysan Rasooli and Douglas G Down Guidelines for selecting hadoop schedulers based on system heterogeneity Journal of Grid Computing, 12(3):499–519, Sep 2014 [56] A Roy, Jingye Xu, and M.H Chowdhury Multi-core processors: A new way forward and challenges In Microelectronics, 2008 ICM 2008 International Conference on, pages 454–457, 2008 [57] Shieh, Wann-Yun and Pong, Chin-Ching Energy and transition-aware runtime task scheduling for multicore processors J Parallel Distrib Comput., 73(9):1225–1238, September 2013 [58] Mark T Chapman The benefits of dual-core Processors in High-Performance computing IBM System and Technology Group, 2005 [59] Grochowski, E and Ronen, R and Shen, J and Wang, P Best of both latency and throughput Computer Design: VLSI in Computers and Processors ICCD 2004, pages 236–243, 2004 [60] Kolpe, T and Zhai, A and Sapatnekar, S.S Enabling improved power management in multicore processors through clustered DVFS Design, Automation Test in Europe Conference Exhibition (DATE), page 1–6, 2011 [61] Shvachko, Konstantin and Kuang, Hairong and Radia, Sanjay and Chansler, Robert The Hadoop Distributed File System In Proceedings of the 2010 IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST), MSST ’10, pages 1–10, Washington, DC, USA, 2010 IEEE Computer Society BIBLIOGRAPHY 113 [62] SPEC Acer AW2000h-Aw170h f2(intel xeon e5-2660) machine http://www.spec org/power_ssj2008/results/res2012q4/power_ssj2008-20120918-00546.html, October 2012 [63] SPEC Acer Incorporated Acer ar380 f2 (intel xeon e5-2640) http://www.spec.org/ power_ssj2008/results/res2012q3/power_ssj2008-20120525-00481.html, July 2012 [64] SPEC Acer Incorporated Acer ar380 f2 (intel xeon e5-2665) http://www.spec.org/ power_ssj2008/results/res2012q3/power_ssj2008-20120525-00479.html, July 2012 [65] SPEC Fujitsu primergy tx100 s3p (intel xeon e3-1240v2) http://www.spec.org/ power_ssj2008/results/res2012q3/power_ssj2008-20120726-00519.html, August 2012 [66] SPEC Hitachi ha8000/rs110-hhm (intel xeon e5-2470) https://www.spec.org/ power_ssj2008/results/res2012q3/power_ssj2008-20120724-00515.html, August 2012 [67] SPEC PowerEdge r820 (intel xeon e5-4650l) machine http://www.spec org/power_ssj2008/results/res2012q4/power_ssj2008-20121113-00586.html, November 2012 [68] SPEC Acer AW2000h-Aw170h f2 (intel xeon e5-2670) machine http://www.spec org/power_ssj2008/results/res2013q1/power_ssj2008-20121212-00590.html, February 2013 [69] SPEC Fujitsu PRIMERGY rx100 s8 (intel xeon e3-1265lv3) https://www.spec org/power_ssj2008/results/res2013q4/power_ssj2008-20131018-00643.html, November 2013 [70] SPEC Fujitsu FUJITSU Server PRIMERGY tx1330 m1 https://www.spec.org/ power_ssj2008/results/res2015q1/power_ssj2008-20150116-00684.html, Jul 2015 [71] SPEC Fujitsu FUJITSU Server PRIMERGY tx1330 m2 https://www.spec.org/ power_ssj2008/results/res2016q1/power_ssj2008-20151214-00707.html, Jan 2016 [72] Xiaoyong Tang, Kenli Li, Meikang Qiu, and Edwin H.-M Sha A hierarchical reliability-driven scheduling algorithm in grid systems Journal of Parallel and Distributed Computing, 72(4):525 – 535, 2012 BIBLIOGRAPHY 114 [73] Andrei Tchernykh, Juan Manuel Ramírez, Arutyun Avetisyan, Nikolai Kuzjurin, Dmitri Grushin, and Sergey Zhuk Two Level Job-Scheduling Strategies for a Computational Grid, pages 774–781 Springer Berlin Heidelberg, Berlin, Heidelberg, 2006 [74] George Terzopoulos and Helen D Karatza Performance evaluation of a real-time grid system using power-saving capable processors The Journal of Supercomputing, 61(3):1135–1153, 2012 [75] George Terzopoulos and Helen D Karatza Performance evaluation of a real-time grid system using power-saving capable processors The Journal of Supercomputing, 61(3):1135–1153, 2012 [76] U Karthick Kumar A dynamic load balancing algorithm in computational Grid using Fair scheduling International Journal of Computer Science Issues - IJCSI, 8(1):123–129 [77] Vinod Kumar Vavilapalli, Arun C Murthy, Chris Douglas, Sharad Agarwal, Mahadev Konar, Robert Evans, Thomas Graves, Jason Lowe, Hitesh Shah, Siddharth Seth, Bikas Saha, Carlo Curino, Owen O’Malley, Sanjay Radia, Benjamin Reed, and Eric Baldeschwieler Apache Hadoop YARN: Yet Another Resource Negotiator In Proceedings of the 4th Annual Symposium on Cloud Computing, SOCC ’13, pages 5:1–5:16, New York, NY, USA, 2013 ACM [78] Verma, Abhishek and Cherkasova, Ludmila and Campbell, Roy H Orchestrating an Ensemble of MapReduce Jobs for Minimizing Their Makespan IEEE Trans Dependable Secur Comput., 10(5):314–327, September 2013 [79] Lizhe Wang, Jie Tao, and Gregor von Laszewski Multicores in Cloud Computing: Research Challenges for Applications Journal of Computers, 5(6), 06/2010 2010 [80] Fatos Xhafa and Ajith Abraham Computational models and heuristic methods for grid scheduling problems Future Generation Computer Systems, 26(4):608 – 621, 2010 [81] Xuan, T T and Tien, V D and Chakka, R The Impact of Dynamic Power Management in Computational Clusters with Multi-Core Processors Journal of Scientific and Industrial Research (JSIR) , 75:339–343, June 2016 [82] B Yagoubi and Y Slimani Dynamic load balancing strategy for grid computing Transactions on Engineering, Computing and Technology, 13:260–265, 2006 BIBLIOGRAPHY 115 [83] B Yagoubi and Y Slimani Dynamic load balancing strategy for grid computing Transactions on Engineering, Computing and Technology, 13:260–265, 2006 [84] Yi Yao, Jiayin Wang, Bo Sheng, Jason Lin, and Ningfang Mi HaSTE: Hadoop YARN Scheduling Based on Task-Dependency and Resource-Demand In Proceedings of the 2014 IEEE International Conference on Cloud Computing, CLOUD ’14, pages 184–191, Washington, DC, USA, 2014 IEEE Computer Society [85] Nezih Yigitbasi, Kushal Datta, Nilesh Jain, and Theodore Willke Energy Efficient Scheduling of MapReduce Workloads on Heterogeneous Clusters In Green Computing Middleware on Proceedings of the 2Nd International Workshop, GCM ’11, pages 1:1– 1:6, New York, NY, USA, 2011 ACM [86] Matei Zaharia, Andy Konwinski, Anthony D Joseph, Randy Katz, and Ion Stoica Improving MapReduce Performance in Heterogeneous Environments In Proceedings of the 8th USENIX Conference on Operating Systems Design and Implementation, OSDI’08, pages 29–42, Berkeley, CA, USA, 2008 USENIX Association [87] S Zikos and H D Karatza Resource allocation strategies in a 2-level hierarchical grid system In 41st Annual Simulation Symposium (anss-41 2008), pages 157–164, April 2008 [88] Stylianos Zikos and Helen D Karatza Communication cost effective scheduling policies of nonclairvoyant jobs with load balancing in a grid Journal of Systems and Software, 82(12):2103–2116, 2009 [89] Stylianos Zikos and Helen D Karatza The impact of service demand variability on resource allocation strategies in a grid system ACM Trans Model Comput Simul., 20(4):19:1–19:29, November 2010 [90] Stylianos Zikos and Helen D Karatza A clairvoyant site allocation policy based on service demands of jobs in a computational grid Simulation Modelling Practice and Theory, 19(6):1465–1478, 2011 [91] Stylianos Zikos and Helen D Karatza Performance and energy aware cluster-level scheduling of compute-intensive jobs with unknown service times Simulation Modelling Practice and Theory, 19(1):239–250, 2011