摘要
伴随信息技术不断革新互联网应用迅速发展,由于对数据传输速度及质量要求提高,光纤网络逐渐被广泛应用。为满足数据高效率传输目的,在云计算环境下,提出基于动态数据摆放的差异性调度特征匹配方法,对光纤数据实现有效调度。在保证数据任务请求满足期望执行效果条件下,采用数据摆放方法,对数据进行预处理操作,通过数据共享规则在云资源池中对数据块实行动态摆放,依据节点资源利用机率确定数据摆放位置,达到覆盖最多数据块的目的,为调度策略提供资源基础。以决策树最优解路径为基础,提出差异性数据调度匹配策略,在云环境下从全局角度考虑数据任务优先级,基于云资源公用库,将决策树节点分别布置到每个存储器上,并对数据资源分类进行控制,实现光纤数据差异化调度。实验证明,通过运用文中数据调度方法能够提升光纤数据传输质量。
With the continuous innovation of information technology and the rapid development of Internet applications,due to the increase in data transmission speed and quality requirements,fiber optic networks are gradually being widely used. In order to meet the purpose of high-efficiency data transmission,in the cloud computing environment,a differential scheduling feature matching method based on dynamic data placement is proposed to realize effective scheduling of fiber data. Under the condition that the data task request is guaranteed to meet the expected execution effect,the data placement method is used to preprocess the data,and the data block is dynamically placed in the cloud resource pool through the data sharing rule,and the data position is determined according to the probability of using the node resource. The placement location reaches the goal of covering the most data blocks to provide a resource basis for the scheduling strategy. Based on the optimal solution path of the decision tree,a differential data scheduling matching strategy is proposed. In the cloud environment,the data task priority is considered from a global perspective. Based on the cloud resource common library,the decision tree nodes are respectively arranged on each storage,and control data resource classification to achieve differential scheduling of fiber data. Experiments show that the data transmission quality of optical fiber can be improved by using the data scheduling method in the paper.
作者
吴华芹
WU Huaqin(Henan Vocational College of Applied Technology,Zhengzhou Henan 450042,China)
出处
《激光杂志》
北大核心
2019年第1期155-158,共4页
Laser Journal
关键词
云计算
海量光纤数据
差异化调度
cloud computing
mass optical fiber data
differential scheduling
