Load balancing for cluster systems under heavy-tailed and temporal dependent workloads |
| |
Affiliation: | 1. Sustainable Engineering Asset Management (SEAM) Research Group, University of Sharjah, United Arab Emirates;2. Department of Industrial Engineering and Engineering Management, College of Engineering, University of Sharjah, United Arab Emirates;3. Department of Systems Engineering and Engineering Management, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong;4. Benha Faculty of Engineering, Benha University, Benha, Egypt;1. Postgraduate Program in Biotechnology, Department of Biopharmaceuticals and Biopharmaceutical Engineering, FEMSA Biotechnology Center, National Graduate School of Science, Engineering and Technology, Tecnológico de Monterrey, Nuevo León, Mexico;2. Postgraduate Program in Dentistry, Doctorate School, Universitat Internacional de Catalunya, Barcelona, Spain;3. Medical and Health Sciences Program, Department of Basic Sciences, National School of Medicine, Tecnológico de Monterrey, Nuevo León, Mexico;4. Medical and Surgical Dentist Program, Tecnológico de Monterrey, Nuevo León, Mexico;5. Biomaterials Innovation Research Center, Division of Biomedical Engineering, Department of Medicine, Brigham and Women''s Hospital, Harvard Medical School, Boston 02139, MA, USA;6. Postgraduate Program in Molecular Biology and Genetic Engineering, Department of Virology, Laboratory of Molecular Infectology, Department of Biochemistry and Molecular Medicine, Faculty of Medicine, Universidad Autónoma de Nuevo León, Mexico;1. Korea Advanced Institute of Science and Technology, Republic of Korea;2. State Key Laboratory of Automotive Safety and Energy, Tsinghua University, China |
| |
Abstract: | Large-scaled cluster systems have been employed in various areas by offering pools of fundamental resources. Efficient allocation of the shared resources in a cluster system is a critical but challenging issue, which has been extensively studied in the past few years. Despite the fact that existing load balancing policies, such as Random, Join Shortest Queue and size-based polices, are widely implemented in actual systems due to their simplicity and efficiency, the performance benefits of these policies diminish when workloads are highly variable and temporally correlated. In this paper, we propose a new load balancing policy, named ADuS, which attempts to partition jobs according to their present sizes and further rank the servers based on their loads. By dispatching jobs of similar sizes to the corresponding ranked servers, ADuS can adaptively balance user traffic and system load in a cluster and thus achieve significant performance benefits. Extensive trace-driven simulations using both synthetic and real traces show the effectiveness and robustness of ADuS under many different environments. |
| |
Keywords: | Resource allocation Temporal dependence Heavy tailed workloads Size-based load balancing Cluster systems |
本文献已被 ScienceDirect 等数据库收录! |
|