期刊文献+

集群下Cholesky分解的数据重用算法

Data Reusing Algorithm of Cholesky Decomposition on Clusters
下载PDF
导出
摘要 核外计算中,由于磁盘I/O操作特点是启动开销大,所以对文件的访问时间占的比例较大。如果能减少读取文件操作的次数则可以大幅度地提高运行效率。数据重用是一种有效的减少I/O操作次数的技术。本文将数据分成几个文件,然后将本次Cholesky分解完毕的文件继续的留在内存缓冲区中。当对下一个文件进行分解时,可用上一个刚分解完的文件进行数据的更新。这样就减少了读取数据的I/O操作次数,从而提高了分解效率。 Access to files is a larger proportion of time because disk I/O operations are characterized by a large overhead starts in the Out-of-core computation.We can greatly improve the operating efficiency by reducing the number of operations to read the file.Data reusing is a technology which can reduce effectively the number of I/O operations.In this article the data was first divided into several files,then the file of Cholesky decomposition remain in the memory buffer.We used the file that has just been decomposed to update the current data file when decomposing the next file.Which reduces the number of I/O operations of reading data and improve the efficiency of Cholesky decomposition.
作者 刘凤 刘青昆
出处 《微计算机应用》 2011年第2期15-20,共6页 Microcomputer Applications
关键词 数据重用 核外 集群 CHOLESKY分解 Data Reusing Out-of-core Clusters Cholesky decomposition
  • 相关文献

参考文献5

二级参考文献13

  • 1朱传琪,臧斌宇,陈彤.程序自动并行化系统[J].软件学报,1996,7(3):180-186. 被引量:34
  • 2丁永华,陈彤,臧斌宇,朱传琪.过程繁衍及其实现方法[J].软件学报,1996,7(11):662-668. 被引量:3
  • 3J. Ramanujam, M. Kandemir, A. Choudhary, et al.Compilation techniques for out-of-core parallel computations.Parallel Computing, 1998, 23(3-4): 597~628. 被引量:1
  • 4H. Simitci, D. Reed. A. comparison of logical and physical parallel I/O patterns. The Int'l Journal of High Performance Computing Applications, 1998, 12(3): 364~380. 被引量:1
  • 5M. Kandemir, A. Choudhary, J. Ramanujam. Compiler optimizations for I/O-intensive computations. In: Proc. 1999Int'l Conf. Parallel Processing. Wakamatsu, Japan: IEEE Computer Press, 1999. 164 ~171. 被引量:1
  • 6D. Callahan, K. Kennedy, A. Porterfield. Software prefetching.In: Proc. 4th Int'l Conf. Architectural Support for Programming Languages and Operation Systems. New York: ACM Press,1991. 40~52. 被引量:1
  • 7W.Y. Chen. Data preload for superscalar and VLIW processors:[Ph. D. dissertation]. Illinois: University of Illinois, 1993. 被引量:1
  • 8A.D. Brown, T. C. Mowry, O. Krieger. Compiler-based I/O prefetching for out-of-core applications. ACM Trans. Computer Systems, 2001, 19(2): 111~170. 被引量:1
  • 9S. Carr, K. S. McKinley, C. W. Tseng. Compiler optimizations for improving data locality. In: Proc. 6th Int'l Conf.Architectural Support for Programming Languages and Operating Systems. New York: ACM Press, 1994. 252~262. 被引量:1
  • 10M.S. Lam, M. E. Wolf. A. data locality optimizing algorithm.ACM SIGPLAN Notices, 2004, 39(4): 442~459. 被引量:1

共引文献7

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部