Quantization Error and Accuracy-Performance Tradeoffs for Embedded Data Mining Workloads
作者： Ramanathan Narayanan , Berkin Özıṣıkyılmaz , Gokhan Memik , Alok Choudhary , Joseph Zambreno
DOI: 10.1007/978-3-540-72588-6_121
关键词: Data stream mining 、 Domain (software engineering) 、 Computer science 、 Process (computing) 、 Data mining 、 Software portability 、 Association rule learning
摘要: Data mining is the process of automatically finding implicit, previously unknown and potentially useful information from large volumes data. Embedded systems are increasingly used for sophisticated data algorithms to make intelligent decisions while storing analyzing Since applications designed implemented considering resources available on a conventional computing platform, their performance degrades when executed an embedded system. In this paper, we analyze bottlenecks faced in implementing these environment explore portability domain. Particularly, floating point computation convert them into fixed operations. Our results reveal that execution time five representative can be reduced by as much 11.5× 5.2× average, without significant impact accuracy.
springer.com
sci-hub.st 参考文章(7)
Gokhan Memik, Jayaprakash Pisharath, Alok Choudhary, Joseph Zambreno, Berkin Özıs.ıkyılmaz, Performance Characterization of Data Mining Applications using MineBench ,(2006)
Daniel Menard, Daniel Chillet, François Charot, Olivier Sentieys, Automatic floating-point to fixed-point conversion for DSP code generation compilers, architecture, and synthesis for embedded systems. pp. 270- 276 ,(2002) , 10.1145/581630.581674
Mike Estlick, Miriam Leeser, James Theiler, John J. Szymanski, Algorithmic transformations in the implementation of K- means clustering on reconfigurable hardware field programmable gate arrays. pp. 103- 110 ,(2001) , 10.1145/360276.360311
Rafael Canetti, Julio Pérez Acle, Ramiro González, Roberto Fleitas, Daniel Ferrer, NeuroFPGA-implementing artificial neural networks on programmable logic devices design, automation, and test in europe. ,vol. 3, pp. 30218- ,(2004) , 10.5555/968880.969246
Z.K. Baker, V.K. Prasanna, Efficient hardware data mining with the Apriori algorithm on FPGAs field-programmable custom computing machines. pp. 3- 12 ,(2005) , 10.1109/FCCM.2005.31
S. Roy, P. Banerjee, An algorithm for trading off quantization error with hardware resources for MATLAB-based FPGA design IEEE Transactions on Computers. ,vol. 54, pp. 886- 896 ,(2005) , 10.1109/TC.2005.106
Ramanathan Narayanan, Berkin Ozisikyilmaz, Joseph Zambreno, Gokhan Memik, Alok Choudhary, MineBench: A Benchmark Suite for Data Mining Workloads ieee international symposium on workload characterization. pp. 182- 188 ,(2006) , 10.1109/IISWC.2006.302743