高性能GIS研究进展及评述

发布时间：2018-04-07 18:03

本文选题：高性能GIS　切入点：高性能GIS算法　出处：《地球信息科学学报》2017年04期

【摘要】：互联网技术的发展使地理信息技术得到了前所未有的发展和应用,地理信息计算呈现出计算速度快、运行效率高、应用多样化的发展特征。而随着计算机硬件性能飞速提升,传统的GIS数据处理方式并不能与之匹配,各种缺陷与弊端逐渐显现,亟待更高效的数据处理方式。目前,以并行集群计算技术和分布式网络技术为代表的高性能计算的出现,为这些问题的解决带来了新思路,并逐渐发展形成了新一代的多核并行高性能计算系统。当前,如何利用新型硬件体系结构带来的计算能力,研究新一代高性能GIS计算系统,解决现在所面临的时空数据密集和计算密集问题成为重要挑战。高性能计算是基于一组或几组计算机系统组成的集群,通过网络连接组成超级计算系统以加强数据处理、分析计算性能的一种技术。在实际应用中,逐渐形成Hadoop,Spark和Storm 3大主流分布式高性能计算系统,它们三者各具优缺点。本文从高性能GIS算法、并行GIS计算、内存计算和众核计算4个方面梳理、归纳总结了高性能GIS的技术体系,分析了每类高性能GIS技术特征,综合分析、评述了近年来高性能GIS的研究进展,并对高性能GIS未来发展进行展望,为更完备、高效的高性能GIS体系的建立、发展和应用提供参考。今后,并行GIS计算、高性能计算模式和分布式存储仍然是GIS技术领域发展的重要方向,通过高性能GIS系统可有效地解决时空数据密集、计算密集和网络通讯密集等问题,大大提升GIS地理分析效率。
[Abstract]:With the development of Internet technology, geographic information technology has been developed and applied in an unprecedented way. Geographical information computing has the characteristics of fast computing speed, high running efficiency and diversified application.With the rapid improvement of the performance of computer hardware, the traditional GIS data processing method can not match it, various defects and drawbacks gradually appear, it is urgent to more efficient data processing methods.At present, the emergence of high performance computing, represented by parallel cluster computing technology and distributed network technology, has brought new ideas for solving these problems, and a new generation of multi-core parallel high performance computing systems has been gradually developed.At present, how to make use of the computing power brought by the new hardware architecture to study the new generation of high-performance GIS computing system and solve the problem of space-time data intensive and computation-intensive is an important challenge.High performance computing is a kind of technology which is based on a group or groups of computer systems to form a super computing system through network connection to enhance the data processing and analyze the computing performance.In practical application, there are three main distributed high performance computing systems, Hadoop Spark and Storm, which have their own advantages and disadvantages.This paper summarizes the technical architecture of high performance GIS algorithm, parallel GIS computing, memory computing and multikernel computing, summarizes the technical system of high performance GIS, analyzes the technical characteristics of each kind of high performance GIS, and analyzes the comprehensive analysis of each kind of high performance GIS.The research progress of high performance GIS in recent years is reviewed, and the future development of high performance GIS is prospected, which provides a reference for the establishment, development and application of a more complete and efficient high performance GIS system.In the future, parallel GIS computing, high-performance computing mode and distributed storage are still important directions in the field of GIS technology. High-performance GIS systems can effectively solve the problems of space-time data density, computation-intensive and network communication, etc.The efficiency of GIS geographic analysis is greatly improved.
【作者单位】：北京超图软件股份有限公司;超图地理信息技术研究所;中国科学院地理科学与资源研究所;
【基金】：国家测绘公益项目(201512015) 北京市科技专项(Z151100003615012、Z141101004414011) 中国科学院国防科技创新基金项目(CXJJ-14-M13);中国科学院重点部署项目(KZZD-EW-07-01-001) 国家科技支撑计划项目(2013BAC03B00) 资源与环境信息系统国家重点实验室自主研究项目(088RAC00YA) 北京市优秀人才项目(201500002685XG242) 全国博士后国际交流计划(20150081) 朝阳区博士后基金项目
【分类号】：P208

【相似文献】