面向遥感影像协同处理的并行框架服务系统

The Parallel Framework Service System for Cooperative Processing of Remote Sensing Image

遥感影像种类的不断增加和数据量的急剧膨胀，使得建立良好可扩展性的遥感影像协同处理平台显得非常急迫和必要。本文从海量遥感影像协同处理需求出发，设计了一套与资源松耦合的并行框架服务系统，并给出了服务系统的组成和各关键服务的实现途径。实验结果表明，该框架不仅能合理分配作业任务、适应不同优先级的任务、实现负载均衡，而且其调度策略灵活，呈现出部署灵活方便、使用友好、适应性强的特点。

It is necessary and urgent to build up a remote sensing image cooperative processing system which has good extensibility due to the increase of remote sensing image types and inflation of the data size. In this paper, a parallel framework service system for cooperative processing of remote sensing image is designed, which is loose coupled with the computing resources, the system composition and realization approach of the key service are also presented. The tested and verified result shows that this parallel framework can not only allocate computing tasks reasonably, adapt to different priority level task and realize load balancing, but also has the characteristics of flexibility for scheduling strategy and deployment as well as the good adaptability.