Data Layout and Scheduling Tasks in a Meteorological Cloud Environment  

作　　者：Kunfu Wang Yongsheng Hao Jie Cao 

机构地区：[1]School of Mathematics and Statistics,Nanjing University of Information Science&Technology,Nanjing,210044,China

出　　处：《Intelligent Automation & Soft Computing》2023年第7期1033-1052,共20页智能自动化与软计算（英文）

基　　金：funded in part byMajor projects of the National Social Science Fund(16ZDA054)of China;the Postgraduate Research&Practice Innovation Program of Jiansu Province(NO.KYCX18_0999)of China;the Engineering Research Center for Software Testing and Evaluation of Fujian Province(ST2018004)of China.

摘　　要：Meteorological model tasks require considerable meteorological basis data to support their execution.However,if the task and the mete-orological datasets are located on different clouds,that enhances the cost,execution time,and energy consumption of execution meteorological tasks.Therefore,the data layout and task scheduling may work together in the meteorological cloud to avoid being in various locations.To the best of our knowledge,this is the first paper that tries to schedule meteorological tasks with the help of the meteorological data set layout.First,we use the FP-Growth-M(frequent-pattern growth for meteorological model datasets)method to mine the relationship between meteorological models and datasets.Second,based on the relation,we propose a heuristics algorithm for laying out the meteorological datasets and scheduling tasks.Finally,we use simulation results to compare our proposed method with other methods.The simulation results show that our method reduces the number of involved clouds,the sizes of files from outer clouds,and the time of transmitting files.

关 键 词：Meteorologicalmodels datalayout bigdata DATAMINING 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]