为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改...为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改进人工鱼群算法在迭代中的收敛速度和种群多样性.其次,将自适应全局-局部记忆机制引入到标准AFSA算法的觅食行为中,以进一步提高勘探能力.最后,增加了基于平均适应度的行为选择机制,以提供更合理的行为选择,减少算法的复杂性.通过使用CloudSim平台进行实验验证,分别测试在不同任务规模下IAFSA的算法效能.实验结果表明,改进人工鱼群算法在降低系统任务最大完工时间和成本上均表现出了显著的优势.展开更多
In the US Federal government, an analysis of alternatives (AoA) is required for a significant investment of resources. The AoA yields the recommended alternative from a set of viable alternatives for the investment de...In the US Federal government, an analysis of alternatives (AoA) is required for a significant investment of resources. The AoA yields the recommended alternative from a set of viable alternatives for the investment decision. This paper presents an integrated AoA and project management framework for analyzing new or emerging alternatives (e.g., Cloud computing), as may be driven by an information system strategy that incorporates a methodology for analyzing the costs, benefits, and risks of each viable alternative. The case study in this paper, about a business improvement project to provide public health and safety services to citizens in a US Federal agency, is a practical application of this integrated framework and reveals the benefits of this integrated approach for an investment decision. The decision making process in the framework—as an integrated, organized, and adaptable set of management and control practices—offers a defensible recommendation and provides accountability to stakeholders.展开更多
Cloud computing has created a paradigm shift that affects the way in which business applications are developed. Many business organizations use cloud infrastructures as platforms on which to deploy business applicatio...Cloud computing has created a paradigm shift that affects the way in which business applications are developed. Many business organizations use cloud infrastructures as platforms on which to deploy business applications. Increasing numbers of vendors are supplying the cloud marketplace with a wide range of cloud products. Different vendors offer cloud products in different formats. The cost structures for consuming cloud products can be complex. Finding a suitable set of cloud products that meets an application’s requirements and budget can be a challenging task. In this paper, an ontology-based resource mapping mechanism is proposed. Domain-specific ontologies are used to specify high-level application’s requirements. These are then translated into high-level infrastructure ontologies which then can be mapped onto low-level descriptions of cloud resources. Cost ontologies are proposed for cloud resources. An exemplar media transcoding and delivery service is studied in order to illustrate how high-level requirements can be modeled and mapped onto cloud resources within a budget constraint. The proposed ontologies provide an application-centric mechanism for specifying cloud requirements which can then be used for searching for suitable resources in a multi-provider cloud environment.展开更多
为了解决IaaS(Infrastructure as a Service)云的工作流调度优化问题,提出基于预算约束的工作流调度算法。以最小化工作流调度时长为目标,算法分调度任务选择和虚拟机实例选择两阶段进行。第一阶段将工作流任务依据依赖关系作层次划分,...为了解决IaaS(Infrastructure as a Service)云的工作流调度优化问题,提出基于预算约束的工作流调度算法。以最小化工作流调度时长为目标,算法分调度任务选择和虚拟机实例选择两阶段进行。第一阶段将工作流任务依据依赖关系作层次划分,同层次组成包任务,以Min-Max方法对层次任务估算时间作标准化处理,定义最迟完成时间与最早完成时间差值最大者为调度任务;第二阶段在期望预算下以最早完成时间最小为标准选择资源,实现任务与资源间的映射。利用算例阐述了算法实现过程,并通过仿真实验测试了算法性能。结果证实,改进算法执行效率与调度成功率优于同类算法。展开更多
文摘为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改进人工鱼群算法在迭代中的收敛速度和种群多样性.其次,将自适应全局-局部记忆机制引入到标准AFSA算法的觅食行为中,以进一步提高勘探能力.最后,增加了基于平均适应度的行为选择机制,以提供更合理的行为选择,减少算法的复杂性.通过使用CloudSim平台进行实验验证,分别测试在不同任务规模下IAFSA的算法效能.实验结果表明,改进人工鱼群算法在降低系统任务最大完工时间和成本上均表现出了显著的优势.
文摘In the US Federal government, an analysis of alternatives (AoA) is required for a significant investment of resources. The AoA yields the recommended alternative from a set of viable alternatives for the investment decision. This paper presents an integrated AoA and project management framework for analyzing new or emerging alternatives (e.g., Cloud computing), as may be driven by an information system strategy that incorporates a methodology for analyzing the costs, benefits, and risks of each viable alternative. The case study in this paper, about a business improvement project to provide public health and safety services to citizens in a US Federal agency, is a practical application of this integrated framework and reveals the benefits of this integrated approach for an investment decision. The decision making process in the framework—as an integrated, organized, and adaptable set of management and control practices—offers a defensible recommendation and provides accountability to stakeholders.
文摘Cloud computing has created a paradigm shift that affects the way in which business applications are developed. Many business organizations use cloud infrastructures as platforms on which to deploy business applications. Increasing numbers of vendors are supplying the cloud marketplace with a wide range of cloud products. Different vendors offer cloud products in different formats. The cost structures for consuming cloud products can be complex. Finding a suitable set of cloud products that meets an application’s requirements and budget can be a challenging task. In this paper, an ontology-based resource mapping mechanism is proposed. Domain-specific ontologies are used to specify high-level application’s requirements. These are then translated into high-level infrastructure ontologies which then can be mapped onto low-level descriptions of cloud resources. Cost ontologies are proposed for cloud resources. An exemplar media transcoding and delivery service is studied in order to illustrate how high-level requirements can be modeled and mapped onto cloud resources within a budget constraint. The proposed ontologies provide an application-centric mechanism for specifying cloud requirements which can then be used for searching for suitable resources in a multi-provider cloud environment.
文摘为了解决IaaS(Infrastructure as a Service)云的工作流调度优化问题,提出基于预算约束的工作流调度算法。以最小化工作流调度时长为目标,算法分调度任务选择和虚拟机实例选择两阶段进行。第一阶段将工作流任务依据依赖关系作层次划分,同层次组成包任务,以Min-Max方法对层次任务估算时间作标准化处理,定义最迟完成时间与最早完成时间差值最大者为调度任务;第二阶段在期望预算下以最早完成时间最小为标准选择资源,实现任务与资源间的映射。利用算例阐述了算法实现过程,并通过仿真实验测试了算法性能。结果证实,改进算法执行效率与调度成功率优于同类算法。
