| Python 多核并行计算在海量星表交叉证认中 的应用 |
| |
| 引用本文： | 裴彤,张彦霞,彭南博,赵永恒.Python 多核并行计算在海量星表交叉证认中 的应用[J].中国科学:物理学 力学 天文学,2011,41(1):102-107. |
| |
| 作者姓名： | 裴彤 张彦霞 彭南博 赵永恒 |
| |
| 作者单位： | 中国科学院国家天文台, 北京 100012 |
| |
| 基金项目： | 国家自然科学基金(批准号: 10778724, 10778616, 11033001)和国家高技术研究发展计划(编号: 2006AA01A120)资助项目 |
| |
| 摘 要： | 天文学研究中经常需要通过交叉证认将来自多波段多项目天文数据联系起来统一考虑. 当前天文数据急剧增长, 必然导致交叉证认的速度过慢. 针对这一问题, 提出一种在多核环境下使用Python 语言进行高效并行计算的方法, 与以往的研究结果相比, 速度提高了若干倍. 这为下一步的多波段数据统计研究和数据挖掘打下了良好的基础.
|
| 关 键 词： | 天文星表 巡天 数据库 交叉证认 并行计算 Python |
| 收稿时间： | 2009-12-23 |
| 修稿时间： | 2010-02-01 |
The application of multi-core parallel computing using python language in cross-matching of massive catalogues |
| |
| PEI Tong,ZHANG YanXia,PENG NanBo,ZHAO YongHeng.The application of multi-core parallel computing using python language in cross-matching of massive catalogues[J].Scientia Sinica Pysica,Mechanica & Astronomica,2011,41(1):102-107. |
| |
| Authors: | PEI Tong ZHANG YanXia PENG NanBo ZHAO YongHeng |
| |
| Institution: | National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China |
| |
| Abstract: | As astronomical data grows rapidly, cross-matching between huge catalogues that contain millions or billions of celestial objects becomes a hotspot of research. In this paper, we present a parallel cross-match program, which is written in Python language and is able to make full use of multi-core processors. We explain why Python programming language is chosen and how to implement parallel computing with it. A sky splitting function HTM is selected to partition catalogues. The results of experiments prove that our program has a significant performance advantage comparing with previous functions and lays a good foundation for further statistical research and data mining. |
| |
| Keywords: | astronomical catalogs sky surveys databases cross-match parallel computing Python |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《中国科学:物理学 力学 天文学》浏览原始摘要信息 |
| 点击此处可从《中国科学:物理学 力学 天文学》下载免费的PDF全文 |
