摘要
为了吸引更多的用户使用云存储以获得更大的经济效益,云提供商为潜在用户提供免费体验,为收费用户提供质量更高的存储服务。融合免费服务和收费服务给出一种云存储架构,考虑潜在用户的不耐烦行为和收费用户源有限,建立双队列多服务台排队模型。采用稳态解析法,导出系统中潜在用户和收费用户的平均时延的性能表达式,并通过系统实验揭示系统参数对系统性能的影响。综合用户的响应性能和云提供商的投资规模,建立系统成本函数,改进万有引力智能寻优算法,给出云服务器的部署方案,进行云存储资源的有效管理。
In order to appeal more users to the cloud storage and to obtain greater economic benefits,cloud vendors offer a free experience to potential users and provide a higher quality storage service for chargeable users.By integrating free service and chargeable service,a cloud storage architecture is given.Considering impatience behavior of potential users and finite number of chargeable users,a queueing model with two cloud buffers and multiple servers is built.With the help of steady-state analysis method,the average latency of potential users and the average latency of chargeable users are derived.The influence of system parameters on system performance is investigated.Considering both the response performance of users and the investment of cloud vendors,a cost function of system is established.Through improving gravitational searching algorithm,a deployment plan of cloud servers is provided,and an effective management of cloud storage resource is presented.
引文
[1]LIU C,ZHANG X,YANG C.Authorized public auditing of dynamic big data storage on cloud with efficient verifiable fine-grained updates [J].IEEE Transactions on Parallel and Distributed Systems,2014,25(9):2234-2244.
[2]SOOKHAK M,GANI A,KHAN M.Dynamic remote data auditing for securing big data storage in cloud computing [J].Information Science,2016,380(20):101-116.
[3]KIM H,PARK J,JEONG Y.Human-centric storage resource mechanism for big data on cloud service architecture [J].Journal of Supercomputing,2016,72(7):2437-2452.
[4]PAPIER F,THONEMANN U.Capacity rationing in rental systems with two customer classes and batch arrivals [J].Omega,2011,39(1):73-85.
[5]ZHAO N,LIAN Z.A queueing-inventory system with two classes of customers [J].International Journal of Production Economics,2011,129(1):225-231.
[6]LIU Z,GAO S.Discrete-time Geo1,GeoX2/G1,G2/1 retrial queue with two classes of customers and feedback [J].Mathematical and Computer Modelling,2011,53(5/6):1208-1220.
[7]SUDHESH R.Transient analysis of a queue with system disasters and customer impatience [J].Queueing Systems,2010,66(1):95-105.
[8]LAXMI P,JYOTHSNA K.Impatient customer queue with Bernoulli schedule vacation interruption [J].Computers & Operations Research,2015,56:1-7.
[9]AMMAR S.Transient analysis of a two-heterogeneous servers queue with impatient behavior [J].Journal of the Egyptian Mathematical Society,2014,22(1):90-95.
[10]金顺福,史国英.认知无线电网络中动态频谱分配策略的收费方案 [J].燕山大学学报,2014,38(4):327-331.JIN S F,SHI G Y.Pricing policy of dynamic spectrum allocation strategy in cognitive radio networks [J].Journal of Yanshan University,2014,38(4):327-331.
[11]刘赛,李绪蓉,万麟瑞.基于排队论的云计算资源池模型研究 [J].计算机技术与发展,2012,22(12):87-89.LIU S,LI X R,WAN L R.Computing clouds resources pool model research based on queue theory [J].Computer Technology and Development,2012,22(12):87-89.
[12]GANDOMI A,YANG X.Chaotic bat algorithm [J].Journal of Computational Science,2014,5(2):224-232.
[13]YAZDANI S,NEZAMABADI-POUR H,KAMYAB S.A gravitational search algorithm for multimodal optimization [J].Swarm Evolutionary Computation,2014,14:1-14.