能耗与可靠性兼顾的云工作流成本优化方法研究
发布时间：2022-01-06 07:31
在云上执行复杂的科学工作流应用程序通常会涉及大量的虚拟机（VM）,这使得成本和能耗成为人们关注的焦点。为了缓解该问题,一些云服务提供商（例如CloudSigma和ElasticHosts）引入了新的定价策略,根据分配的CPU频率以及虚拟机配置和价格的各种组合对用户收费。然而,可定制的CPU频率使资源分配和调度变得更加困难,难以实现成本最优的调度方案。较高的CPU频率会带来高能耗并增强可靠性,而为了降低能耗,降低CPU频率会产生软错误问题,从而导致工作流应用程序在规定时间内完成的失败率很高。因此,非常需要一种频率调节方法来获得成本优化的工作流调度解决方案。为了解决上述挑战,本研究提出了一种基于遗传算法的新方法。该方法采用新引入的遗传算子（即交叉和变异）,在能耗、可靠性、最大完工时间和内存约束下,将任务分配到具有特定工作频率的虚拟机中,为云工作流快速找到成本最优的资源供应和任务调度解决方案。本论文主要有以下三点贡献:1.在考虑带回滚恢复的检查点开销的情况下,将最大完工时间、能耗、内存和可靠性约束下的云工作流任务调度的成本优化问题形式化。2.基于遗传算法,我们引入了一种新的遗传算子,即染色体...
【文章来源】：华东师范大学上海市 211工程院校 985工程院校 教育部直属院校
【文章页数】：87 页
【学位级别】：硕士
【文章目录】：
摘要
ABSTRACT
Notations used in this thesis
1 Introduction
1.1 Research Background and Significance
1.2 Research Challenges and Contributions
1.3 Relevant Work
1.3.1 Cost Optimization in Cloud Computing
1.3.2 Energy Saving Technologies of Data Center
1.3.3 Cloud Service Fault Tolerance Technology
1.4 Thesis Organization
2 Introduction to Relevant Theories
2.1 Cloud Computing
2.1.1 Classification Based on Service Type
2.1.2 Classification Based on Deployment Mode
2.2 Soft Error
2.3 Checkpointing with Rollback Recovery
2.4 Summary
3 System Models and Problem Description
3.1 System Models
3.1.1 Virtual Machine Model
3.1.2 Workflow Model
3.1.3 Fault Tolerant Task Model
3.1.4 Energy Consumption Model
3.2 Problem Description
3.3 Summary
4 Cost Optimizing Workflow Scheduling Algorithm
4.1 Algorithm Overview
4.2 Detailed Design of Algorithm
4.2.1 Chromosome Encoding
4.2.2 Frequency Selection
4.2.3 Chromosome Regularization
4.2.4 Fitness Function
4.2.5 Crossover
4.2.6 Mutation
4.2.7 Chromosome Modification
4.3 Case Study
4.4 Summary
5 Experimental Results and Analysis
5.1 Experimental Setup
5.1.1 Experimental Parameters
5.1.2 Workflow Data Set
5.1.3 Comparison Algorithms
5.2 Results and Analysis
5.2.1 Cost Optimization with Different Makespan Goals
5.2.2 Energy Optimization with Different Makespan Goals
5.2.3 Cost Optimization with Different Reliability Goals
5.2.4 Energy Optimization with Different Reliability Goals
5.3 Summary
6 Conclusion and Future Work
6.1 Conclusion
6.2 Future Work
Bibliography
Acknowledgements
Research Outcomes
本文编号：3572027
【文章来源】：华东师范大学上海市 211工程院校 985工程院校 教育部直属院校
【文章页数】：87 页
【学位级别】：硕士
【文章目录】：
摘要
ABSTRACT
Notations used in this thesis
1 Introduction
1.1 Research Background and Significance
1.2 Research Challenges and Contributions
1.3 Relevant Work
1.3.1 Cost Optimization in Cloud Computing
1.3.2 Energy Saving Technologies of Data Center
1.3.3 Cloud Service Fault Tolerance Technology
1.4 Thesis Organization
2 Introduction to Relevant Theories
2.1 Cloud Computing
2.1.1 Classification Based on Service Type
2.1.2 Classification Based on Deployment Mode
2.2 Soft Error
2.3 Checkpointing with Rollback Recovery
2.4 Summary
3 System Models and Problem Description
3.1 System Models
3.1.1 Virtual Machine Model
3.1.2 Workflow Model
3.1.3 Fault Tolerant Task Model
3.1.4 Energy Consumption Model
3.2 Problem Description
3.3 Summary
4 Cost Optimizing Workflow Scheduling Algorithm
4.1 Algorithm Overview
4.2 Detailed Design of Algorithm
4.2.1 Chromosome Encoding
4.2.2 Frequency Selection
4.2.3 Chromosome Regularization
4.2.4 Fitness Function
4.2.5 Crossover
4.2.6 Mutation
4.2.7 Chromosome Modification
4.3 Case Study
4.4 Summary
5 Experimental Results and Analysis
5.1 Experimental Setup
5.1.1 Experimental Parameters
5.1.2 Workflow Data Set
5.1.3 Comparison Algorithms
5.2 Results and Analysis
5.2.1 Cost Optimization with Different Makespan Goals
5.2.2 Energy Optimization with Different Makespan Goals
5.2.3 Cost Optimization with Different Reliability Goals
5.2.4 Energy Optimization with Different Reliability Goals
5.3 Summary
6 Conclusion and Future Work
6.1 Conclusion
6.2 Future Work
Bibliography
Acknowledgements
Research Outcomes
本文编号：3572027
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/3572027.html
