基于SVM的遥感数据并行服务体系研究

A Study on Parallel Service System Architecture of RS Image Based on SVM

该文研究了利用共享虚拟存储 (SVM )的并行普通机群作为遥感数据分布式服务系统的后台服务器 ,分析了在这种服务体系下多源遥感数据的组织和存储 ,提出了“塔、块、层、相”的立体索引机制 ,设计了服务器端的并行处理程序 ,并对基于SVM并行机群环境对后台服务器性能改进方面进行了测试和分析。

This paper has studied service architecture of multi source remote sensing(RS) data's distributed service system based on SVM.First,this paper has set up a solid index mechanism of Pyramid,Block,Layer,Epoch according to the properties of RS data and analyzed the organization and storage of multi source RS data on background.Secondly,this paper has designed parallel processes aiming at the background server's service functions.Moreover,these processes have been tested and analyzed for the improvement of parallel computer cluster performance as background server based on SVM.Also,this perofrmance improvement of the parallel PCs background server has been demonstrated.