Improved Parallel Processing Function for High-Performance Large-Scale Astronomical Cross-Matching  

作　　者：赵青 孙济州 于策 肖健 崔辰州 张啸 

机构地区：[1]School of Computer Science and Technology, Tianjin University [2]National Astronomical Observatories, Chinese Academy of Sciences

出　　处：《Transactions of Tianjin University》2011年第1期62-67,共6页天津大学学报（英文版）

基　　金：Supported by National Natural Science Foundation of China (No.10978016);Natural Science Foundation of Tianjin (No. 08JCZDJC19700);Key Technologies Research and Development Program of Tianjin (No.09ZCKFGX00400)

摘　　要：Astronomical cross-matching is a basic method for aggregating the observational data of different wavelengths. By data aggregation, the properties of astronomical objects can be understood comprehensively. Aiming at decreasing the time consumed on I/O operations, several improved methods are introduced, including a processing flow based on the boundary growing model, which can reduce the database query operations; a concept of the biggest growing block and its determination which can improve the performance of task partition and resolve data-sparse problem; and a fast bitwise algorithm to compute the index numbers of the neighboring blocks, which is a significant efficiency guarantee. Experiments show that the methods can effectively speed up cross-matching on both sparse datasets and high-density datasets.Astronomical cross-matching is a basic method for aggregating the observational data of different wavelengths. By data aggregation, the properties of astronomical objects can be understood comprehensively. Aiming at decreasing the time consumed on I/O operations, several improved methods are introduced, including a processing flow based on the boundary growing model, which can reduce the database query operations; a concept of the biggest growing block and its determination which can improve the performance of task partition and resolve data-sparse problem; and a fast bitwise algorithm to compute the index numbers of the neighboring blocks, which is a significant efficiency guarantee. Experiments show that the methods can effectively speed up cross-matching on both sparse datasets and high-density datasets.

关 键 词：astronomical cross-matching boundary growing model HEALPix task partition data-sparse problem 

分 类 号：P113[天文地球—天文学]