Multi-stream 2D frequency table computation on dataflow architecture
作者： P. Skoda , V. Sruk , B. Medved Rogina
DOI: 10.1109/MIPRO.2015.7160279
关键词: Parallel computing 、 Dataflow 、 Speedup 、 Field-programmable gate array 、 Kernel (statistics) 、 Data stream mining 、 Computer science 、 Computation 、 Dataflow architecture 、 Decision tree learning
摘要: Frequency table computation is a common procedure used in variety of machine learning algorithms. In this paper we present parallelized kernel for computing frequency tables. The targeted dataflow architecture implemented on field programmable gate array (FPGA). Its performance was evaluated against software implementation running 6-core CPU. with six concurrent input data streams 300 MHz achieved speedup up to 6.26×, compared 6 threaded 3.2 GHz
sci-hub.se 参考文章(11)
Steven L. Salzberg, Alberto Segre, Programs for Machine Learning ,(1994)
Oded Maimon, Lior Rokach, Data Mining with Decision Trees: Theory and Applications ,(2007)
B. Medved Rogina, V. Sruk, P. Skoda, FPGA implementations of data mining algorithms international convention on information and communication technology, electronics and microelectronics. pp. 362- 367 ,(2012)
Grigorios Chrysos, Panagiotis Dagritzikos, Ioannis Papaefstathiou, Apostolos Dollas, HC-CART: A parallel system implementation of data mining classification and regression tree (CART) algorithm on a multi-FPGA system high performance embedded architectures and compilers. ,vol. 9, pp. 47- ,(2013) , 10.1145/2400682.2400706
Jack B Dennis, None, Data Flow Supercomputers IEEE Computer. ,vol. 13, pp. 48- 56 ,(1980) , 10.1109/MC.1980.1653418
Oliver Pell, Vitali Averbukh, Maximum Performance Computing with Dataflow Engines computational science and engineering. ,vol. 14, pp. 98- 103 ,(2012) , 10.1109/MCSE.2012.78
S. Ruggieri, Efficient C4.5 [classification algorithm] IEEE Transactions on Knowledge and Data Engineering. ,vol. 14, pp. 438- 444 ,(2002) , 10.1109/69.991727
P. Skoda, V. Sruk, B. Medved Rogina, Frequency table computation on dataflow architecture international convention on information and communication technology electronics and microelectronics. pp. 342- 346 ,(2014) , 10.1109/MIPRO.2014.6859588
Alok N. Choudhary, Daniel Honbo, Prabhat Kumar, Berkin Ozisikyilmaz, Sanchit Misra, Gokhan Memik, Accelerating data mining workloads: current approaches and future challenges in system architecture design Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery. ,vol. 1, pp. 41- 54 ,(2011) , 10.1002/WIDM.9
Ian Kuon, FPGA Architecture ,(2008)