DocumentCode
589956
Title
Use of CUDA for the Continuous Space Language Model
Author
Thompson, Elizabeth A. ; Anderson, T.
Author_Institution
Dept. of Eng., Purdue Univ. Fort Wayne, Fort Wayne, IN, USA
fYear
2012
fDate
10-12 Sept. 2012
Firstpage
1
Lastpage
5
Abstract
The training phase of the Continuous Space Language Model (CSLM) was implemented in the NVIDIA hardware/software architecture Compute Unified Device Architecture (CUDA). Implementation was accomplished using a combination of CUBLAS library routines and CUDA kernel calls on three different CUDA enabled devices of varying compute capability and a time savings over the traditional CPU approach demonstrated.
Keywords
hardware-software codesign; language translation; parallel architectures; software architecture; statistical analysis; CPU approach; CSLM; CUBLAS library routines; CUDA; CUDA kernel calls; NVIDIA hardware-software architecture; compute unified device architecture; continuous space language model; statistical machine translation; training phase; Algorithm design and analysis; Computer architecture; Graphics processing units; Instruction sets; Kernel; Libraries; Training; CSLM; CUBLAS; CUDA; GPU; statistical signal processing;
fLanguage
English
Publisher
ieee
Conference_Titel
High Performance Extreme Computing (HPEC), 2012 IEEE Conference on
Conference_Location
Waltham, MA
Print_ISBN
978-1-4673-1577-7
Type
conf
DOI
10.1109/HPEC.2012.6408661
Filename
6408661
Link To Document