虚拟计算环境下基于改进二分K均值的任务聚类模型
详细信息 查看官网全文
- 英文论文题名：Task Clustering Model based on the Improved Bisect K-means under iVCE
- 论文作者：秦童 ; 孙斌 ; 刘欣然 ; 刘悦 ; 杨榆
- 年：2016
- 论文关键词：任务聚类 ; 二分K均值 ; 统计特征 ; 虚拟计算环境
- 英文论文关键词：task clustering ; bisect k-means ; statistical characteristics ; iVCE
- 会议召开时间：2016-06-24
- 会议录名称：2016年全国通信软件学术会议程序册与交流文集
- 语种：中文
- 分类号：TP181
- 学会代码：ZGTH
- 会议名称：2016年全国通信软件学术会议
- 会议地点：中国陕西西安
- 主办单位：中国通信学会
- 学会名称：中国通信学会
- 页数：6
- 文件大小：1933k
- 原文格式：O
- 会议级别：全国
摘要
针对虚拟计算环境中任务具有数量庞大、需求模糊、种类多样等特征,本文通过虚拟计算实验床平台公布的一周数据,提出了一种对任务特征进行聚类的分析模型,通过分析任务特征,从运行需求、服务指标、任务状态与消耗情况等方面进行分析,并基于机器学习理论对任务特征进行聚类研究,经实验结果验证,应用与任务消耗之间的关系不是绝对的。根据不同任务特征进行聚类,其聚类结果也各不相同。此外,取值半径越小,簇类数量越少;当取值半径的数值不变时,合并因子越小,簇类数量越多。这些实验结果为构建任务仿真模型与解决虚拟计算环境下资源匹配面临的挑战提供新思路。
Under iVCE(Internet-based Virtual Computing Environment),tasks has a large number,ambiguous requirements,variety types and other features.In this paper,according with one-week data published by iVCE platfonn,we analysis on task features from running demand,service index,task status and consumption.In addition,we adopt machine learning theory to study task clustering.It has been verified by experimental results that the relationship between application and task consumption is not absolute.According with various task characteristics to cluster,the results are different from each other.Besides,the value of radius is small,the cluster number is less.When radius is fixed,the smaller merge factor can lead to more number of clusters.These results can be to build the simulation model and solve the problems of resource allocation.
Under iVCE(Internet-based Virtual Computing Environment),tasks has a large number,ambiguous requirements,variety types and other features.In this paper,according with one-week data published by iVCE platfonn,we analysis on task features from running demand,service index,task status and consumption.In addition,we adopt machine learning theory to study task clustering.It has been verified by experimental results that the relationship between application and task consumption is not absolute.According with various task characteristics to cluster,the results are different from each other.Besides,the value of radius is small,the cluster number is less.When radius is fixed,the smaller merge factor can lead to more number of clusters.These results can be to build the simulation model and solve the problems of resource allocation.
引文
[1]Lu XC,Wang HM,Wang J.Internet-based Virtual Computing Environment:Beyond the data center as a computer.Future generation computer system-the international journal of grid computing and science,2013,29(l):309-322.
[2]Sun JJ,Wang XW,Gao CX,Huang M.Resource allocation scheme based on neural network and group search optimization in cloud environment.Journal of Software,2014,25(8):1858-1873.
[3]Zhu Cliunge,Liu Xinran,Yang Yixian,Zhang Hong.Application-oriented resource matching model based on trust for internet-based virtual computing environment.Journal on Communications.2013,34(9):24-32.
[4]Abbadi IM,Ruan A.Towards trustworthy resource scheduling in clouds.IEEE Transactions on Information Forensics and Security,2013,8(6):973-984.
[5]Xu Dayu.Yang Shanlin,Liu Renping.A mixture of HMM,GA and Elman networks for load predication in cloud-oriented data centers.Journal of Zhejiang University-Science C(Computer&Electronics),2013,14(11):845-858.
[6]Guo Fengyu,Yu Long,Tian Shengwei,Yu Jiong.A workflow task scheduling algorithm based on the resources'fuzzy clustering in cloud computing environment.International Journal of Communication Systems,2015.28:1053-1067.
[7]Zhenhua Liu,Adam Wierman,Yuan Chen,et al.Data center demand response:Avoiding the coincident peak via workload shifting and local generation.Performance Evaluation,2013,70:770-791.
[8]Stillwell M,Vivien F,Casanova H.Virtual machine resource allocation for sen-ice hosting on heterogeneous distributed platforms.In:Proceedings of IEEE 26th international conference on parallel distributed processing symposium(IPDPS'12),2012,786-797.
[9]Li Xuelong,Gong Haigang.Review on big data system.Science China(Information Science),2015,45(l):l-44.
[10]Mohamed Ben Belgacem,Bastien Chopard.A hybrid HPC/cloud distributed infrastructure:Coupling EC2 cloud resources with HPC clusters to run large tightly coupled multi-scale applications.Future Generation Computer Systems,2015,42:11-21.
[11]Hameed Hussain,Saif Ur Rehman Malik,etc.A survey on resource allocation in high performance distributed computing systems.Parallel Computing(Systems&Applications),2013,39:709-736.
[12]Sahar Adabi,Ali Movaghar,Amir Masoud Rahmani.Bi-level fuzzy based advanced reservation of Cloud workflow applications on distributed Grid resources.Journal of Supercomputing,2014,67:175-218.
[13]Sheng Di,Derrick Kondo,Walfredo Cime.Google hostload predication based on Bayesian model with optimized feature combination.Journal of Parallel and Distributed Computing,2014,74:1820-1832.
[14]Fahimeh Ramezani,Jie Lu,Farookh Khadeer Hussain.Task-based system load balancing in Cloud Computing using particle swarm optimization.International Journal of Parallel and Programming,2014,42:739-754.
[15]Marc Bux,Ulf Leser.DynamicCloudSim:Simulationg heterogeneity in computational clouds.Future Generation Computer Systems.2015,46:85-99.
[16]Liu Z,Cho S.Characterizing machines and workloads on a Google cluster.In:8th international workshop on scheduling and resource management for parallel and distributed systems(SRMPDS 12),2012,397-403.
[17]Sheng Di,Derrick Kondo,Franck Cappello.Characterizing and modeling cloud applications/jobs on a Google data center.Journal of Supercomputing,2014,69:139-160.
[2]Sun JJ,Wang XW,Gao CX,Huang M.Resource allocation scheme based on neural network and group search optimization in cloud environment.Journal of Software,2014,25(8):1858-1873.
[3]Zhu Cliunge,Liu Xinran,Yang Yixian,Zhang Hong.Application-oriented resource matching model based on trust for internet-based virtual computing environment.Journal on Communications.2013,34(9):24-32.
[4]Abbadi IM,Ruan A.Towards trustworthy resource scheduling in clouds.IEEE Transactions on Information Forensics and Security,2013,8(6):973-984.
[5]Xu Dayu.Yang Shanlin,Liu Renping.A mixture of HMM,GA and Elman networks for load predication in cloud-oriented data centers.Journal of Zhejiang University-Science C(Computer&Electronics),2013,14(11):845-858.
[6]Guo Fengyu,Yu Long,Tian Shengwei,Yu Jiong.A workflow task scheduling algorithm based on the resources'fuzzy clustering in cloud computing environment.International Journal of Communication Systems,2015.28:1053-1067.
[7]Zhenhua Liu,Adam Wierman,Yuan Chen,et al.Data center demand response:Avoiding the coincident peak via workload shifting and local generation.Performance Evaluation,2013,70:770-791.
[8]Stillwell M,Vivien F,Casanova H.Virtual machine resource allocation for sen-ice hosting on heterogeneous distributed platforms.In:Proceedings of IEEE 26th international conference on parallel distributed processing symposium(IPDPS'12),2012,786-797.
[9]Li Xuelong,Gong Haigang.Review on big data system.Science China(Information Science),2015,45(l):l-44.
[10]Mohamed Ben Belgacem,Bastien Chopard.A hybrid HPC/cloud distributed infrastructure:Coupling EC2 cloud resources with HPC clusters to run large tightly coupled multi-scale applications.Future Generation Computer Systems,2015,42:11-21.
[11]Hameed Hussain,Saif Ur Rehman Malik,etc.A survey on resource allocation in high performance distributed computing systems.Parallel Computing(Systems&Applications),2013,39:709-736.
[12]Sahar Adabi,Ali Movaghar,Amir Masoud Rahmani.Bi-level fuzzy based advanced reservation of Cloud workflow applications on distributed Grid resources.Journal of Supercomputing,2014,67:175-218.
[13]Sheng Di,Derrick Kondo,Walfredo Cime.Google hostload predication based on Bayesian model with optimized feature combination.Journal of Parallel and Distributed Computing,2014,74:1820-1832.
[14]Fahimeh Ramezani,Jie Lu,Farookh Khadeer Hussain.Task-based system load balancing in Cloud Computing using particle swarm optimization.International Journal of Parallel and Programming,2014,42:739-754.
[15]Marc Bux,Ulf Leser.DynamicCloudSim:Simulationg heterogeneity in computational clouds.Future Generation Computer Systems.2015,46:85-99.
[16]Liu Z,Cho S.Characterizing machines and workloads on a Google cluster.In:8th international workshop on scheduling and resource management for parallel and distributed systems(SRMPDS 12),2012,397-403.
[17]Sheng Di,Derrick Kondo,Franck Cappello.Characterizing and modeling cloud applications/jobs on a Google data center.Journal of Supercomputing,2014,69:139-160.