invited-talk

Accelerated Deep-Learning inference on FPGAs in the Space Domain

Authors:

Martin WernerAuthors Info & Claims

CF '23: Proceedings of the 20th ACM International Conference on Computing Frontiers

Pages 222 - 228

https://doi.org/10.1145/3587135.3592763

Published: 04 August 2023 Publication History

Abstract

Artificial intelligence has found its way into space, and similar to the situation on ground demands powerful hardware to unfold its full potential. With the heterogeneous compute platform that is offered by the space-grade variant of the Versal, AMD Xilinx presents a system that is particularly targeted at accelerating AI inference in space. This paper investigates the design flow and the achievable performance of this novel device. We present benchmark results in terms of concrete figures and measurements, i.e., throughput, latency, and power consumption, achieved by a predesigned hardware accelerator realized on the system, and compare them to a previous generation platform.

References

[1]

3d Generation Partnership Project. 2023. Release 17. https://www.3gpp.org/specifications-technologies/releases/release-17. [Online; accessed 22-January-2023].

[2]

A. Gupta. 2020. Architecture apocalypse dream architecture for deep learning inference and compute - versal ai core. https://www.xilinx.com/content/dam/xilinx/support/documents/white_papers/EW2020-Deep-Learning-Inference-AICore.pdf. [Online; accessed 28-January-2023].

[3]

Advanced Micro Devices, Inc. 2019. ZCU102 Evaluation Board User Guide. https://docs.xilinx.com/v/u/en-US/ug1182-zcu102-eval-bd. [Online; accessed 25-January-2023].

[4]

Advanced Micro Devices, Inc. 2022. AI Engine Kernel and Graph Programming Guide. https://www.xilinx.com/content/dam/xilinx/support/documents/sw_manuals/xilinx2022_2/ug1079-ai-engine-kernel-coding.pdf. [Online; accessed 25-January-2023].

[5]

Advanced Micro Devices, Inc. 2022. AI Engines and Their Applications. https://docs.xilinx.com/v/u/en-US/wp506-ai-engine. [Online; accessed 25-January-2023].

[6]

Advanced Micro Devices, Inc. 2022. Versal Architecture and Product Data Sheet: Overview. https://docs.xilinx.com/v/u/en-US/ds950-versal-overview. [Online; accessed 25-January-2023].

[7]

Advanced Micro Devices, Inc. 2022. XQR Versal for Space 2.0 Applications. https://www.xilinx.com/content/dam/xilinx/publications/product-briefs/xilinx-xqr-versal-product-brief.pdf. [Online; accessed 22-January-2023].

[8]

Advanced Micro Devices, Inc. 2022. Zynq UltraScale+ MPSoC Product Brief. https://www.xilinx.com/content/dam/xilinx/support/documents/product-briefs/zynq-ultrascale-plus-product-brief.pdf. [Online; accessed 25-January-2023].

[9]

Advanced Micro Devices, Inc. 2023. Defense-Grade Zynq UltraScale+ MP-SoCs. https://www.xilinx.com/products/silicon-devices/soc/xq-zynq-ultrascale-mpsoc.html. [Online; accessed 25-January-2023].

[10]

Advanced Micro Devices, Inc. 2023. Dpuczdx8g for zynq ultrascale+ mpsocs. https://docs.xilinx.com/r/en-US/pg338-dpu. [Online; accessed 29-January-2023].

[11]

Advanced Micro Devices, Inc. 2023. Versal AI Core Series VCK190 Evaluation Kit. https://www.xilinx.com/products/boards-and-kits/vck190.html. [Online; accessed 25-January-2023].

[12]

Advanced Micro Devices, Inc. 2023. Vitis ai user guide. https://docs.xilinx.com/r/en-US/ug1414-vitis-ai. [Online; accessed 29-January-2023].

[13]

Ahmad Al-Zoubi, Gianluca Martino, Fin H. Bahnsen, Jun Zhu, Holger Schlarb, and Goerschwin Fey. 2022. CNN Implementation and Analysis on Xilinx Versal ACAP at European XFEL. In 2022 IEEE 35th International System-on-Chip Conference (SOCC). 1--6. https://doi.org/10.1109/SOCC56010.2022.9908101

[14]

Daniel Chew and A. Brinton Cooper. 2020. Spectrum Sensing in Interference and Noise Using Deep Learning. In 2020 54th Annual Conference on Information Sciences and Systems (CISS). 1--6. https://doi.org/10.1109/CISS48834.2020.1570617443

[15]

J. Duarte, S. Han, P. Harris, S. Jindariani, E. Kreinar, B. Kreis, J. Ngadiuba, M. Pierini, R. Rivera, N. Tran, and Z. Wu. 2018. Fast inference of deep neural networks in FPGAs for particle physics. Journal of Instrumentation 13, 07 (Jul 2018), P07027--P07027. https://doi.org/10.1088/1748-0221/13/07/p07027

[16]

Clément Farabet, Berin Martini, Polina Akselrod, Selçuk Talay, Yann LeCun, and Eugenio Culurciello. 2010. Hardware accelerated convolutional neural networks for synthetic vision systems. In Proceedings of 2010 IEEE International Symposium on Circuits and Systems. 257--260. https://doi.org/10.1109/ISCAS.2010.5537908

[17]

Fraunhofer Institute for Integrated Circuits IIS. 2023. 5G Satellite Integration. https://www.iis.fraunhofer.de/en/ff/kom/satkom/sat-5g.html. [Online; accessed 22-January-2023].

[18]

Brian Gaide, Dinesh Gaitonde, Chirag Ravishankar, and Trevor Bauer. 2019. Xilinx Adaptive Compute Acceleration Platform: VersalTM Architecture. In Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (Seaside, CA, USA) (FPGA '19). Association for Computing Machinery, New York, NY, USA, 84--93. https://doi.org/10.1145/3289602.3293906

Digital Library

[19]

Max Ghiglione and Vittorio Serra. 2022. Opportunities and Challenges of AI on Satellite Processing Units. In Proceedings of the 19th ACM International Conference on Computing Frontiers (Turin, Italy) (CF '22). Association for Computing Machinery, New York, NY, USA, 221--224. https://doi.org/10.1145/3528416.3530985

Digital Library

[20]

S. Haykin. 2005. Cognitive radio: brain-empowered wireless communications. IEEE Journal on Selected Areas in Communications 23, 2 (2005), 201--220. https://doi.org/10.1109/JSAC.2004.839380

Digital Library

[21]

Markus Nagel, Marios Fournarakis, Rana Ali Amjad, Yelysei Bondarenko, Mart van Baalen, and Tijmen Blankevoort. 2021. A White Paper on Neural Network Quantization. arXiv:arXiv:2106.08295

[22]

Dominika Przewlocka-Rus, Syed Shakib Sarwar, H. Ekin Sumbul, Yuecheng Li, and Barbara De Salvo. 2022. Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks. arXiv:arXiv:2203.05025

[23]

Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. https://doi.org/10.48550/ARXIV.1505.04597

[24]

The MathWorks, Inc. 2022. Custom IP Core Generation. https://www.mathworks.com/help/hdlcoder/ug/custom-ip-core-generation.html. [Online; accessed 22-January-2023].

[25]

Yaman Umuroglu, Nicholas J. Fraser, Giulio Gambardella, Michaela Blott, Philip Leong, Magnus Jahre, and Kees Vissers. 2017. FINN: A Framework for Fast, Scalable Binarized Neural Network Inference. In Proceedings of the 2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (Monterey, California, USA) (FPGA '17). Association for Computing Machinery, New York, NY, USA, 65--74. https://doi.org/10.1145/3020078.3021744

Digital Library

[26]

Woodrow Wilson International Center for Scholars. 2021. Seizing Opportunities: Four National Security Questions to Ask About the Use of Satellites in 5G Networks. https://5g.wilsoncenter.org/sites/default/files/media/uploads/documents/STIP-SeizingOpportunities.pdf. [Online; accessed 22-January-2023].

[27]

Xilinx Inc. 2022. DPUCVDX8G for Versal ACAPs. https://www.xilinx.com/content/dam/xilinx/support/documents/ip_documentation/dpucvdx8g/v1_1/pg389-dpucvdx8g.pdf. [Online; accessed 24-January-2023].

[28]

Xilinx Inc. 2022. System-Level Benefits of the Versal Platform. https://www.xilinx.com/content/dam/xilinx/support/documents/white_papers/wp539-versal-system-level-benefits.pdf. [Online; accessed 25-January-2023].

[29]

Xilinx Inc. 2022. UltraScale Architecture and Product Data Sheet: Overview. https://www.xilinx.com/content/dam/xilinx/support/documents/data_sheets/ds890-ultrascale-overview.pdf. [Online; accessed 25-January-2023].

Cited By

Petry MWuwer GKoch AGest PGhiglione MWerner M(2023)Accelerated Deep-Learning Inference on the Versal adaptive SoC in the Space Domain2023 European Data Handling & Data Processing Conference (EDHPC)10.23919/EDHPC59100.2023.10396011(1-8)Online publication date: 2-Oct-2023
https://doi.org/10.23919/EDHPC59100.2023.10396011

Index Terms

Accelerated Deep-Learning inference on FPGAs in the Space Domain
1. Computing methodologies
  1. Artificial intelligence
2. Hardware
  1. Emerging technologies
    1. Analysis and design of emerging devices and systems
      1. Emerging architectures

Recommendations

Elliptic Curve Cryptography hardware accelerator for high-performance secure servers

Security threats affecting electronics communications in the current world make necessary the encryption and authentication of every transaction. The increasing levels of security required are leading to an overload of transaction servers due to ...
A configurable mapreduce accelerator for multi-core FPGAs (abstract only)
FPGA '14: Proceedings of the 2014 ACM/SIGDA international symposium on Field-programmable gate arrays

MapReduce is a widely used programming framework for the implementation of cloud computing application in data centers. This work presents a novel configurable hardware accelerator that is used to speed up the processing of multi-core and cloud ...
Toward Evaluating High-Level Synthesis Portability and Performance between Intel and Xilinx FPGAs
IWOCL '21: Proceedings of the 9th International Workshop on OpenCL

Offloading computation from a CPU to a hardware accelerator is becoming a more common solution for improving performance because traditional gains enabled by Moore’s law and Dennard scaling have slowed. GPUs are often used as hardware accelerators, but ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

CF '23: Proceedings of the 20th ACM International Conference on Computing Frontiers

May 2023

419 pages

ISBN:9798400701405

DOI:10.1145/3587135

General Chairs:
Andrea Bartolini
Università di Bologna, IT
,
Kristian Rietveld
Leiden University, NL
,
Program Chairs:
Catherine Schuman
University of Tennessee, US
,
Jose Moreira
IBM, US

Copyright © 2023 Owner/Author.

Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author.

Sponsors

SIGMICRO: ACM Special Interest Group on Microarchitectural Research and Processing

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 04 August 2023

Check for updates

Author Tags

Qualifiers

Invited-talk
Research
Refereed limited

Conference

CF '23

Sponsor:

SIGMICRO

CF '23: 20th ACM International Conference on Computing Frontiers

May 9 - 11, 2023

Bologna, Italy

Acceptance Rates

CF '23 Paper Acceptance Rate 24 of 66 submissions, 36%;

Overall Acceptance Rate 273 of 785 submissions, 35%

Upcoming Conference

CF '25

Sponsor:
sigmicro

22nd ACM International Conference on Computing Frontiers

May 28 - 30, 2025

Cagliari , Italy

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

1
Total Citations
View Citations
94
Total Downloads

Downloads (Last 12 months)65
Downloads (Last 6 weeks)8

Reflects downloads up to 23 Dec 2024

Other Metrics

View Author Metrics

Citations

Cited By

Petry MWuwer GKoch AGest PGhiglione MWerner M(2023)Accelerated Deep-Learning Inference on the Versal adaptive SoC in the Space Domain2023 European Data Handling & Data Processing Conference (EDHPC)10.23919/EDHPC59100.2023.10396011(1-8)Online publication date: 2-Oct-2023
https://doi.org/10.23919/EDHPC59100.2023.10396011

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents