基于用户行为模型的TVOS资源分配算法

doi:10.3969/j.issn.0253-2778.2014.01.002

中国科学技术大学学报 ›› 2014, Vol. 44 ›› Issue (1): 12-18.DOI: 10.3969/j.issn.0253-2778.2014.01.002

• 论著 • 上一篇

基于用户行为模型的TVOS资源分配算法

陈磊，王嵩，吴刚

中国科学技术大学自动化系网络传播系统与控制安徽省重点实验室，安徽合肥 230027

收稿日期:2012-10-15 修回日期:2013-03-16 接受日期:2013-03-16 出版日期:2013-03-16 发布日期:2013-03-16
通讯作者: 吴刚
作者简介:陈磊,男，1986年生，博士生. 研究方向：多媒体资源系统. E-mail: chl123@mail.ustc.edu.cn

User behavior model based TVOS resource allocation

CHEN Lei, WANG Song, WU Gang

University of Science and Technology of China, Department of Automation, Network Communication System and Control, Hefei 230027

Received:2012-10-15 Revised:2013-03-16 Accepted:2013-03-16 Online:2013-03-16 Published:2013-03-16

摘要/Abstract

摘要： 现有的智能电视操作系统（TVOS）资源分配多依赖于操作系统本身对任务的资源分配方案，而系统对任务的调度是尽力而为（best-effort）的，以最大化系统的吞吐量为目的，这种资源调度分配方案在实时或多媒体应用系统存在不能保障应用的服务质量（QoS）.为此，在研究TVOS用户行为模型的基础上，量化了用户对应用的偏好，并结合应用QoS模型提出了两种资源分配算法RA_DP和RA_PLSH.实验结果表明，基于动态规划的RA_DP算法能够求出问题的最优解，可作为算法间比较的参考，但算法时间复杂度很高；基于资源定价的局部搜索启发式RA_PLSH算法可在短时间内求出问题的近似最优解，与其他启发式算法相比更适合于智能电视资源的实时分配.

关键词: TVOS, 资源分配, 用户行为模型, 服务质量

Abstract: The resource allocation of the existing smart TV operating system (TVOS) depends on the operating system task resource allocation scheme, which tries to maximize the throughput of the system. However, this scheme cannot guarantee the quality of service (QoS) of applications in the real-time or multimedia system. In order to solve this problem, the user preferences of applications based on the user behavior model of TVOS were quantified, and two resource allocation algorithms named RA_DP and RA_PLSH were proposed based on the application QoS model. The experimental result shows that the algorithm based on dynamic programming (RA_DP) ensures optimal solution with high time complexity, and can be used as a reference for other algorithms. However, the other algorithm based on resource pricing local search heuristic (RA_PLSH) can obtain near-optimal solution in a very short time, and is thus more suitable for smart TV real-time resource allocation.

Key words: TVOS, resource allocation, user behavior model, QoS

中图分类号:

TP393

陈磊，王嵩，吴刚. 基于用户行为模型的TVOS资源分配算法[J]. 中国科学技术大学学报, 2014, 44(1): 12-18.

CHEN Lei, WANG Song, WU Gang. User behavior model based TVOS resource allocation[J]. Journal of University of Science and Technology of China, 2014, 44(1): 12-18.

参考文献

［1］ Rajkumar R, Lee C, Lehoczky J, et al. A resource allocation model for QoS management［C］// Proceedings of the 18th IEEE Real-Time Systems Symposium. San Francisco, USA: IEEE Press, 1997: 298-307.
［2］ Rajkumar R, Lee C, Lehoczky J P, et al. Practical solutions for QoS-based resource allocation problems［C］// Proceedings of the 19th IEEE Real-Time Systems Symposium. Madrid, Spain: IEEE Computer Society, 1998: 296-306.
［3］ Lee C, Lehoczky J, Rajkumar R, et al. On quality of service optimization with discrete QoS options［C］// Proceedings of the 5th IEEE Real-Time Technology and Applications Symposium. Vancouver, Canada: IEEE Computer Society, 1999: 276-286.
［4］ Lee C, Lehoczky J, Siewiorek D, et al. A scalable solution to the multi-resource QoS problem［C］// Proceedings of the 20th IEEE Real-Time Systems Symposium. Phoenix, USA: IEEE Computer Society, 1999: 315-326.
［5］ Gertphol S, Prasanna V K. MIP formulation for robust resource allocation in dynamic real-time systems［J］. Journal of Systems and Software, 2005, 77(1): 55-65.
［6］ Harada F, Ushio T, Nakamoto Y. Adaptive resource allocation control for fair QoS management［J］. IEEE Transactions on Computers, 2007, 56(3): 344-357.
［7］ Khan S. Quality adaptation in a multisession multimedia system: Model, algorithms and architecture［D］. Department of ECE, University of Victoria, 1998.
［8］ Toyoda Y. A simplified algorithm for obtaining approximate solution to zero-one programming problems［J］. Management Science, 1975, 21(12): 1 417-1 427.
［9］ Akbar M M, Manning E G, Shoja G C, et al. Heuristic solutions for the multiple-choice multi-dimension knapsack problem［C］// Proceedings of the International Conference on Computational Science. San Francisco: Springer-Verlag, 2001: 659-668.
［10］ Shahriar M, Akbar M M, Rahman M S, et al. A multiprocessor based heuristic for multi-dimensional multiple-choice knapsack problem［J］. The Journal of Supercomputing, 2008, 43(3): 257-280.
［11］伍之昂, 罗君舟, 宋爱波，等. 具有QoS保证的服务资源联合分配与管理［J］. 软件学报, 2009, 20(12): 3 150-3 162.
［12］ Verkasalo H. Analysis of smartphone user behavior［C］// Proceedings of the 9th International Conference on Mobile Busines/9th Global Mobility Roundtable. Athens, Greece: IEEE Computer Society, 2010: 258-263.
［13］ Gerber S, Fry M, Kay J, et al. PersonisJ: Mobile, client-side user modelling［C］// Proceedings of the18th International Conference on User Modeling, Adaptation, and Personalization. Big Island, USA: Springer, 2010: 6075: 111-122.
［14］维基百科. 凸包［EB/OL］. http://zh.wikipedia.org/wiki/%E5%87%B8%E5%8C%85.

[1]	郑晓杰，李京. 微云环境下请求分发和容器部署成本优化[J]. 中国科学技术大学学报, 2019, 49(10): 820-827.
[2]	王悦宸，苏醒，贾熹滨，郭黎敏，丁治明. 应急救援中基于线性规划的多目标多资源分配模型[J]. 中国科学技术大学学报, 2018, 48(6): 458-466.
[3]	袁伟，范进，赵定涛，张忞娴. 第三方物流服务质量对顾客忠诚影响的实证研究[J]. 中国科学技术大学学报, 2017, 47(6): 508-513.
[4]	李读，何平. 婚车租赁平台服务标准制定和定价决策[J]. 中国科学技术大学学报, 2016, 46(12): 1022-1029.
[5]	柴磊，赵定涛. 基于DEA的并联生产系统效率评价与资源分配[J]. 中国科学技术大学学报, 2016, 46(11): 939-945.
[6]	李建东，张琰，盛敏，田野，姚俊良. 有效利用协同空间复用的“绿色”资源分配策略[J]. 中国科学技术大学学报, 2009, 39(10): 1022-1026.
[7]	陈晓明，张朝阳，王超. 多用户分布式与集中式天线系统传输容量的分析与比较[J]. 中国科学技术大学学报, 2009, 39(10): 1097-1101.

基于用户行为模型的TVOS资源分配算法

User behavior model based TVOS resource allocation

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

本文评价