Highly-Efficient Parallel Convolution Acceleration by Using Multiple GPUs

被引：0

作者：

Sun, Kuangyuan ^{[1
]}

Li, Shuai ^{[1
]}

Luo, Yukui ^{[1
]}

Renteria, Raul ^{[1
]}

Choi, Ken ^{[1
]}

机构：

[1] IIT, Dept Elect & Comp Engn, VLSI Design & Automat Lab, 3301 S Dearborn St, Chicago, IL 60616 USA

来源：

PROCEEDINGS INTERNATIONAL SOC DESIGN CONFERENCE 2017 (ISOCC 2017) | 2017年

关键词：

Convolutional neural network; parallel acceleration; multiple GPUs;

D O I：

暂无

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Convolutional Neural Network (CNN) is a powerful tool in machine learning area. However, the convolution computation is time-consuming, which limited the application on embedded system. In this paper, we introduce a parallel convolution acceleration implementation by using multiple GPUs Mali-T628 MP6 on embedded system Odroid XU4 and have tested its time reduction and GPU utilization. The result show that the execution time is reduced 25.8% on average.

引用

页码：300 / 301

页数：2

共 50 条

[1] Efficient Parallel UPGMA algorithm Based on Multiple GPUs
Hung, Che-Lun
Wu, Fu-Che
Lin, Chun-Yuan
Chan, Yu-Wei
2016 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM), 2016, : 870 - 873
[2] Efficient parallel CKY parsing using GPUs
Yi, Youngmin
Lai, Chao-Yue
Petrov, Slav
JOURNAL OF LOGIC AND COMPUTATION, 2014, 24 (02) : 375 - 393
[3] A Highly-efficient Approach of Parallel Access to Routing Table on TBGP
Gao Lei
Lai Mingche
Gong Zhenghu
CHINESE JOURNAL OF ELECTRONICS, 2010, 19 (01): : 13 - 17
[4] Data Parallel Acceleration of Decision Support Queries Using COME and GPUs
Trancoso, Pedro
Othonos, Despo
Artemiou, Artemakis
CF'09: CONFERENCE ON COMPUTING FRONTIERS & WORKSHOPS, 2009, : 117 - 126
[5] Highly-Efficient Battery Chargers with Parallel-Loaded Resonant Converters
Chuang, Ying-Chun
Ke, Yu-Lung
Chang, Shun-Yi
2009 IEEE INDUSTRY APPLICATIONS SOCIETY ANNUAL MEETING, 2009, : 698 - 707
[6] SELF-CONVOLUTION: A HIGHLY-EFFICIENT OPERATOR FOR NON-LOCAL IMAGE RESTORATION
Guo, Lanqing
Zha, Zhiyuan
Ravishankar, Saiprasad
Wen, Bihan
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 1860 - 1864
[7] HIGHLY-EFFICIENT MEANS FOR TESTING
BYCHKOV, OD
MEASUREMENT TECHNIQUES-USSR, 1969, (08): : 1056 - &
[8] HIGHLY-EFFICIENT NEUTRON DETECTOR
ISHKHANOV, BS
KAPITONOV, IM
LAZUTIN, EV
PISKAREV, IM
SHEVCHENKO, OP
INSTRUMENTS AND EXPERIMENTAL TECHNIQUES-USSR, 1969, (06): : 1427 - +
[9] ACCELERATION OF RECURSIVE CROSS-CORRELATION PIV USING MULTIPLE GPUS
Tarashima, Shuhei
Someya, Satoshi
Okamoto, Koji
PROCEEDINGS OF THE ASME/JSME 8TH THERMAL ENGINEERING JOINT CONFERENCE 2011, VOL 1 PTS A AND B, 2011, : 1221 - 1227
[10] ACCELERATION of GENERALIZED ADAPTIVE PULSE COMPRESSION with PARALLEL GPUs
Cai, Jingxiao
Zhang, Yan
RADAR SENSOR TECHNOLOGY XIX; AND ACTIVE AND PASSIVE SIGNATURES VI, 2015, 9461

← 1 2 3 4 5 →