遥感可视化建模工具的工作流模型及分布式调度算法  被引量:6

Workflow-based remote sensing visualization modeling system and its distributed scheduling algorithm

在线阅读下载全文

作  者:魏建新[1] 魏东琦[1] 吴信才[2] 

机构地区:[1]中国地质大学研究生院,湖北武汉430074 [2]中国地质大学信息工程学院,湖北武汉430074

出  处:《干旱区地理》2009年第2期304-309,共6页Arid Land Geography

基  金:"面向网络海量空间信息的大型GIS(2003AA133010)"十五863项目资助

摘  要:分布式工作流引擎相对集中式引擎在处理事务效率上有明显的优势。针对现有遥感可视化建模工具对于大规模、大数据的工作流程在效率上存在的问题,分析集中式与分布式工作流系统的特点,参考了当前流行的产品架构,结合实际应用需求,根据遥感影像处理的特征,提出基于有向图的分布式BFS算法的引擎调度算法,算法考虑了分布式系统中时间一致性以及数据一致性问题,引入一种新的冲突解决策略,解决分布式系统下由于时间不一致导致的部分结点不能得到正确的层号的问题;采用等待机制和数据流图的概念,控制较快的分支的执行速率,容许较慢的分支在没有被较快分支赶上时完成对流程图的搜索,保证任务结点所接收数据的完整性。The distributed workflow engine has obvious advantages of efficiency dealing with affairs, which is the relative concentration engine. This paper, in connection with the existing problems on the efficiency for the remote sensing visual modeling tools to handle the large-scale and large-data workflow, analyzing the characteristics of centralized and distributed workflow system, referring to the current structure of popular products, combining with the practical demand, according to the characteristics of remote sensing image processing, gives out an engine scheduling algorithm which is based on digraph of distributed BFS algorithm. This algorithm takes the problems of the time consistency of distributed systems and data consistency into account, introduces a new conflict resolution strategy, resolves the problem that some nodes can not obtain the correct layer number because of inconsistent time under the condition of the distributed system, adopts the back-off mechanism and the concept of data flow diagram. This method controls the branch of faster implementation rate, allows slower branch to complete the search of flow chart which has not been catch up the faster branch, and ensures the integrity of received data of the task node.

关 键 词:遥感 可视化建模 工作流 分布式BFS 

分 类 号:TP751[自动化与计算机技术—检测技术与自动化装置]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象