Transformer-OPU: An FPGA-based Overlay Processor for Transformer Networks

被引：6

作者：

Bai, Yueyin ^{[1
]}

Zhou, Hao ^{[1
]}

Zhao, Keqing ^{[1
]}

Chen, Jianli ^{[1
]}

Yu, Jun ^{[1
]}

Wang, Kun ^{[1
]}

机构：

[1] Fudan Univ, State Key Lab ASIC & Syst, Shanghai, Peoples R China

来源：

2023 IEEE 31ST ANNUAL INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES, FCCM | 2023年

关键词：

D O I：

10.1109/FCCM57271.2023.00049

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Existing implementations of transformer networks by field-programmable gate array (FPGA) focus only on attention computation, or suffer from fixed model structure without flexibility. In this article, we propose an FPGA-based overlay processor, named Transformer-OPU for general accelerations of transformer networks. Experimental result shows that our Transformer-OPU achieves 5.19-15.06x and 1.14-2.89x speedup compared with CPU and GPU, respectively. We also observe 1.10-2.47x better latency compared with previously customized FPGA accelerators, and is 1.45x faster than NPE.

引用

页码：222 / 222

页数：1

共 50 条

[41] Direct FPGA-based Power Profiling for a RISC Processor
Cernazanu-Glavan, Cosmin
Marcu, Marius
Amaricai, Alexandru
Fedeac, Stefan
Ghenea, Madalin
Wang, Zheng
Chattopadhyay, Anupam
Weinstock, Jan
Leupers, Rainer
2015 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE (I2MTC), 2015, : 1578 - 1583
[42] An FPGA-based Doppler processor for a spaceborne precipitation radar
Durden, S.L.
Fischman, M.A.
Johnson, R.A.
Chu, A.J.
Jourdan, M.N.
Tanelli, S.
Journal of Atmospheric and Oceanic Technology, 2007, 24 (10): : 1811 - 1815
[43] Analytical and Measured Sustained Bandwidth for an FPGA-Based Processor
Morris, Gerald R.
Silas, Antoinette R.
Abed, Khalid H.
2012 PROCEEDINGS OF IEEE SOUTHEASTCON, 2012,
[44] FPGA-Based Two-Processor CPU for PLC
Chmiel, Miroslaw
Kloska, Wojciech
Polok, Dariusz
Mocha, Jan
2016 INTERNATIONAL CONFERENCE ON SIGNALS AND ELECTRONIC SYSTEMS (ICSES) PROCEEDINGS, 2016, : 247 - 252
[45] Distributed Processor for FPGA-based Superconducting Qubit Control
Fruitwala, Neelay
Xu, Yilun
Naik, Ravi
Nowrouzi, Kasra
Huang, Gang
2022 IEEE INTERNATIONAL CONFERENCE ON QUANTUM COMPUTING AND ENGINEERING (QCE 2022), 2022, : 763 - 764
[46] Fully FPGA-based implementation of a modified control strategy for power electronic transformer in railway traction applications
Akbariavaz, Khalil
Fazel, Seyed Saeed
Khosravi, Mahyar
IET POWER ELECTRONICS, 2021, 14 (11) : 1904 - 1919
[47] Energy Proportional Computing with OpenCL on a FPGA-Based Overlay Architecture
Sani, Awais Hussain
Nunez-Yanez, Jose Luis
2016 2ND IEEE NORDIC CIRCUITS AND SYSTEMS CONFERENCE (NORCAS), 2016,
[48] FPGA-Based Coherent Doppler Processor for Marine Radar Applications
Abdelbagi, Hamdi
Aljohani, Mansour
Mrebit, Abdulmajid
Wicks, Michael C.
PROCEEDINGS OF THE 2015 IEEE NATIONAL AEROSPACE AND ELECTRONICS CONFERENCE (NAECON), 2015, : 58 - 69
[49] On the implementation of an efficient FPGA-based CFAR processor for target detection
Cumplido, R
Torres, C
López, S
2004 1ST INTERNATIONAL CONFERENCE ON ELECTRICAL AND ELECTRONICS ENGINEERING (ICEEE), 2004, : 214 - 218
[50] FPGA-Based Annealing Processor with Time-Division Multiplexing
Yamamoto, Kasho
Ikebe, Masayuki
Asai, Tetsuya
Motomura, Masato
Takamaeda-Yamazaki, Shinya
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2019, E102D (12) : 2295 - 2305

← 1 2 3 4 5 →