欢迎访问《中国农学通报》,

中国农学通报 ›› 2019, Vol. 35 ›› Issue (28): 118-122.doi: 10.11924/j.issn.1000-6850.casb20190400031

所属专题: 资源与环境 现代农业发展与乡村振兴

• 农业科技信息 • 上一篇    下一篇

农业资源遥感监测系统的计算环境设计与应用

王利民, 刘佳, 滕飞, 杨福刚, 姚保民   

  1. 中国农业科学院农业资源与农业区划研究所
  • 收稿日期:2019-04-23 修回日期:2019-06-06 接受日期:2019-06-17 出版日期:2019-10-14 发布日期:2019-10-14
  • 通讯作者: 刘佳
  • 基金资助:
    高分辨率对地观测系统重大专项(民用部分)(09-Y20A05-9001-17/18)。

Computing Environment of Agricultural Resource Remote Sensing Monitoring System: Design and Application

  • Received:2019-04-23 Revised:2019-06-06 Accepted:2019-06-17 Online:2019-10-14 Published:2019-10-14

摘要: 计算环境是包括农情遥感监测系统在内的农业资源遥感监测系统的重要支撑。该文在对农业资源遥感监测系统硬件环境组成、逻辑结构、核心功能分析等内容阐述基础上,以中国农业科学院农业资源与农业区划研究所运行的“国家农情遥感监测业务运行系统”硬件环境为例,进行了具体说明。研究结果表明,农业资源遥感监测系统是以基础硬件层为支撑,以计算层为核心,以存储层作为辅助,在网络层、软件层支持下,开展农情遥感监测应用。核心功能包括括登录节点、管理节点、计算节点、存储服务器(包括磁盘)等5个部分。“国家农情遥感监测业务运行系统”的计算环境的硬件是以42台基于Intel 64位至强处理器的IBM Flex X240刀片服务器为核心,4台S5700 10G以太网交换机为中转设备,连接华为1套N8500 NAS集群、1台基于Intel 64位至强处理器的IBM X3650M4管理服务器;56G计算IB网络是通过1台Mellanox 36口FDR交换机实现,并包括KVM交换机、机架键盘、液晶显示器,以及机柜、电源插座、连接线缆等附件。系统理论运行效率可以达到13.312万亿次,实际运行效率可以达到理论值的85%以上。

关键词: 温光型, 温光型, 两系杂交小麦 研究进展

Abstract: Computing environment is an important support for agricultural resource remote sensing monitoring system, including agricultural condition remote sensing monitoring system. By taking hardware environment of the “National agricultural condition remote sensing monitoring operation system” operated by Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences as an example, the paper elaborates the analysis of the compositions, logical structure, and core functions of agricultural resource remote sensing monitoring system. The study result shows that, agricultural resource remote sensing monitoring system is supported by a basic hardware layer with computing layer as its core and with the support of a storage layer, and under the support of network layer and software layer, the system carries out the agricultural condition remote sensing monitoring operation. The core functions of the system include five parts of log-in node, management node, computing node, and storage server (including magnetic disks). The cores of hardware of computing environment of the “National agricultural condition remote sensing monitoring operation system” are 42 IBM Flex X240 blade servers based on Intel 64-bitSIntel Xeon. The system also includes four S5700 10G Ethernet interchangers as transit equipments, and is connected with as a set of Huawei N8500 NAS and an IBM X3650M4 management server with Intel 64-bit XeonSProcessor; 56G computing IB network is realized by a set of Mellanox 36口FDR interchanger, including KVM interchanger, frame and keyboard, and LCD , as well as accessories of cabinets, powerSsockets, and cables. Theoretical operation efficiency of the system could reach 13.312 trillion times, while the actual operation efficiency can exceed 85% of theoretical value.