期刊文献+

Openstack平台资源负载预测方法研究 预览

Research on Resource Load Prediction Method of Openstack Platform
在线阅读 下载PDF
收藏 分享 导出
摘要 Openstack中的资源一般是静态分配,利用率较低,因此动态资源分配成为了人们研究的热点。论文针对动态资源分配中的核心步骤之一的负载预测进行了研究,主要采用动态一次指数平滑作为系统的负载预测算法,预测下一时刻系统的负载情况。通过以误差平方和SSE为衡量目标,分别采用Fibonacci法和黄金分割法计算最优权系数,使得SSE最小的权系数即为预测时采用的权系数,然后将选取的权系数代入归一后的系数公式中进行计算预测。 Resources in Openstack are normally statically allocated with low utilization. Therefore, dynamic resource allocation has become a hot topic of research. This paper focuses on load prediction,which is one of the key steps of dynamic resource allocation.The dynamic single exponential smoothing is used as the load prediction algorithm to predict the load of the system at the next moment. Based on the measurement goal of error square and SSE, the optimal weight coefficient is calculated by using Fibonacci method and golden section method respectively, so that the minimum weight coefficient of SSE is the weight coefficient used in the prediction, and then the weight coefficient is substituted into the coefficient formula after normalization for calculation and prediction.
作者 黄秀 蔡全旺 王庆年 HUANG Xiu;CAI Quanwang;WANG Qingnian(No.722 Research Institute, CSIC, Wuhan 430205)
出处 《舰船电子工程》 2019年第3期117-121,172共6页 Ship Electronic Engineering
关键词 OPENSTACK 负载预测 动态一次指数平滑 FIBONACCI 黄金分割 Openstack load forecasting dynamic single exponential smoothing Fibonacci golden section
作者简介 黄秀,女,硕士研究生,研究方向:云计算;蔡全旺,男,硕士,研究员,研究方向:云计算;王庆年,男,硕士,工程师,研究方向:云计算。
  • 相关文献

参考文献4

二级参考文献30

  • 1唐炎森.指数平滑预测公式与平滑系数[J].统计与信息论坛,1998(1):39-44. 被引量:17
  • 2张贤达.现代信号处理[M].北京:清华大学出版社,2001.. 被引量:9
  • 3高仁祥,张世英.非线性经济建模与预测新思路探讨[J].系统工程理论与实践,1997,17(7):7-11. 被引量:14
  • 4Von LASZEWSKI G, WANG Li-zhe, YOUNGE A J, et al. Power-a- ware scheduling of -irtual machines in DVFS-enahled clusters [ C ]// Proc of IEEE International Conference on Cluster Computing and Workshops. [ S. 1. ] : IEEE Press ,2009 : 1 - 10. 被引量:1
  • 5BELOGLAZOV A, ABAWAJY J, BUYYA R. Energy aware resource allocation heuristics for efficient management of data centers for cloud computing[ J ]. Future Generation Computer Systems, 2012,28(5) :755-768. 被引量:1
  • 6OpenStack[ EB/OL]. [2013-05-21 ]. http://www, openstack, org. 被引量:1
  • 7BRET P. OpenStaek tutorial[ R].[ S. 1. ] :IEEE Cloud Corn, 2010. 被引量:1
  • 8ECP Project, Intel cloud builder[ EB/OL]. [ 2013-05-21 ]. http:// www. enomaly, eom/fileadmin/docs/Intel_Cloud_Builder, pdf. 被引量:1
  • 9BELOGLZOV A, BUYYA R. Energy efficient resource management in virtualized cloud data centers[ C ]//Proc of the 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing. [ S. 1. ] : IEEE Press, 2010:826-831. 被引量:1
  • 10URGAONKAR R, KOZAT U, IGARASHI K, et al. Dynamic re- source allocation and power management in virtualized data centers [ C ]//Proc of IEEE Network Operations and Management Sympo- sium. [ S. 1. ] : IEEE Press, 2010:479-486. 被引量:1

共引文献40

投稿分析

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部 意见反馈