云计算环境下格陵兰海盐度数据提取算法研究

RESEARCH ON SALINITY DATA EXTRACTION ALGORITHM FOR GREENLAND SEA IN CLOUD COMPUTING ENVIRONMENT

海洋盐度与海洋浮游植物和海洋温度具有紧密联系,是研究海洋环流和海洋对气候影响的重要参量。以美国NOAA全球海洋信息数据库为数据来源,以经度65°N-85°N、维度20°W-10°E之间的北极格陵兰海地区为研究区域,详细讨论盐度数据的提取,提出时间复杂度为O(n)的盐度数据分步归并算法。利用微软Azure公有云按需付费、动态扩展的特点,获取廉价、便捷的计算资源,大大提升了计算效率。该算法具有很强的可扩展性,可以根据实际计算需求动态调整所需的计算资源,从而能满足不同计算规模的需求。实验结果表明,该算法可以对海量原始盐度数据进行快速分析和归并,生成经纬度、时间、盐度三个维度上的数据。

Ocean salinity has close relationship with ocean temperature and marine phytoplankton,and is one of the important parameters in the study of the influence of ocean circulation and ocean on climate. This paper discusses in detail the extraction of salinity data,of which the world ocean information database of American National Oceanographic Data Centre( NOAA) is used as the data resource,and the area of the Greenland Sea in Arctic within 20° W- 10° E,65° N- 85° N is taken as the studying region. We propose a multi-steps merging algorithm for salinity data with a time complexity of O( n),it utilises the characteristics of Microsoft Azure public cloud that pay-on-demand and dynamic expansion to capture the cheap and convenient computation resources,and greatly improves computation efficiency. The algorithm has strong scalability and is able to dynamically adjust the required computation resources according to actual computation demand,therefore can meet the demands of different computation sizes. Experimental result shows that this algorithm can fast analyse and merge massive original salinity data and creates the formatted salinity data in 3 dimensions: longitude and latitude,time and salinity.