One very important application in the data mining domain is frequent pattern mining. Various authors have worked on improving the efficiency of this computation, mostly focusing on algorithm-level improvement. More recent work has explored architecture specific optimizations of this computation. Our goal in this paper is to provide a systematic approach to architecture-level software optimizations by identifying applicable tuning patterns. We show the generality and effectiveness of these patterns by tuning several frequent pattern mining algorithms and showing significant performance improvements.
Description
Welcome to IEEE Xplore 2.0: Programming Patterns for Architecture-Level Software Optimizations on Frequent Pattern Mining
%0 Conference Paper
%1 4221682
%A Wei, Mingliang
%A Jiang, Changhao
%A Snir, Marc
%D 2007
%J Data Engineering, 2007. ICDE 2007. IEEE 23rd International Conference on
%K frequent-patterns parallel simd
%P 336-345
%R 10.1109/ICDE.2007.367879
%T Programming Patterns for Architecture-Level Software Optimizations on Frequent Pattern Mining
%X One very important application in the data mining domain is frequent pattern mining. Various authors have worked on improving the efficiency of this computation, mostly focusing on algorithm-level improvement. More recent work has explored architecture specific optimizations of this computation. Our goal in this paper is to provide a systematic approach to architecture-level software optimizations by identifying applicable tuning patterns. We show the generality and effectiveness of these patterns by tuning several frequent pattern mining algorithms and showing significant performance improvements.
@inproceedings{4221682,
abstract = {One very important application in the data mining domain is frequent pattern mining. Various authors have worked on improving the efficiency of this computation, mostly focusing on algorithm-level improvement. More recent work has explored architecture specific optimizations of this computation. Our goal in this paper is to provide a systematic approach to architecture-level software optimizations by identifying applicable tuning patterns. We show the generality and effectiveness of these patterns by tuning several frequent pattern mining algorithms and showing significant performance improvements.},
added-at = {2009-05-29T12:03:28.000+0200},
author = {Wei, Mingliang and Jiang, Changhao and Snir, Marc},
biburl = {https://www.bibsonomy.org/bibtex/295bb532233ce7e859c8edc8b8d2d39a5/claudio.lucchese},
description = {Welcome to IEEE Xplore 2.0: Programming Patterns for Architecture-Level Software Optimizations on Frequent Pattern Mining},
doi = {10.1109/ICDE.2007.367879},
interhash = {6f4c409998808c5d0a1690b5df55b98b},
intrahash = {95bb532233ce7e859c8edc8b8d2d39a5},
journal = {Data Engineering, 2007. ICDE 2007. IEEE 23rd International Conference on},
keywords = {frequent-patterns parallel simd},
month = {April},
pages = {336-345},
timestamp = {2009-05-29T12:03:28.000+0200},
title = {Programming Patterns for Architecture-Level Software Optimizations on Frequent Pattern Mining},
year = 2007
}