Efficient algorithms of video replication and placement on a cluster of streaming servers |
| |
| Affiliation: | 1. Department of Computer Science, University of Colorado at Colorado Springs, 1420 Austin Bluffs Parkway, Colorado Springs, CO 80918, USA;2. Department of Electrical & Computer Engineering, Wayne State University, 5050 Anthony Wayne Drive, Detroit, MI 48202, USA;1. Laboratory of Alternative Energies, Electrical Eng. Department, Pici Campus, Federal University of Ceará, 60455-760 Fortaleza, Brazil;2. LAESE Laboratory, Computing Department, Federal Institute of Ceará, Industrial District I, 61939-140 Maracanaú, Brazil;1. College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University, No. 30 South Puzhu Road, Nanjing 211816, People’s Republic of China;2. Department of Bioengineering and Imperial College Centre for Synthetic Biology, Imperial College London, London SW7 2AZ, UK;1. State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, People''s Republic of China;2. School of Mechanical and Aerospace Engineering, Queen''s University Belfast, Belfast BT9 5AH, United Kingdom;1. The Research Institute for Information & Culture, Korea University, Republic of Korea;2. School of Media and Communication, Korea University, 145 Anam-Ro, Seongbuk-Gu, Seoul, 02841, Republic of Korea |
| |
| Abstract: | A cost-effective approach to building up scalable video streaming servers is to couple a number of streaming servers together in a cluster so as to alleviate the inherent storage and networking constraints of streaming services. In this article, we investigate a crucial problem of video replication and placement on a distributed storage cluster of streaming servers for high quality and high availability services. We formulate it as a combinatorial optimization problem with objectives of maximizing the encoding bit rate and the number of replicas of each video and balancing the workload of the servers. The objectives are subject to the constraints of the storage capacity and the outgoing network-I/O bandwidth of the servers. Under the assumption of single fixed encoding bit rate for all video objects with different popularity values, we give an optimal replication algorithm and a bounded placement algorithm, respectively. We further present an efficient replication algorithm that utilizes the Zipf-like video popularity distributions to approximate the optimal solutions, which can reduce the complexity of the optimal replication algorithm. For video objects with scalable encoding bit rates, we propose a heuristic algorithm based on simulated annealing. We conduct a comprehensive performance evaluation of the algorithms and demonstrate their effectiveness via simulations over a synthetic workload set. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
