short-paper

Open access

Porting AI/ML Models to Intelligence Processing Units (IPUs)

Authors:

Abhinand Nasari,

Dhruva Chakravorty,

Honggao LiuAuthors Info & Claims

PEARC '23: Practice and Experience in Advanced Research Computing 2023: Computing for the Common Good

Pages 231 - 236

https://doi.org/10.1145/3569951.3603632

Published: 10 September 2023 Publication History

All formats PDF

Abstract

Intelligence processing units (IPUs) are specifically designed accelerators that are dedicated to support artificial intelligence (AI) and machine learning (ML) workflows. Here, we report on the performance characteristics and code-porting experiences on Graphcore IPUs offered on the new National Science Foundation (NSF)-funded Accelerating Computing for Emerging Sciences (ACES) testbed. Our benchmarks compared performance of AI/ML frameworks on ACES IPUS to similar runs on the Graphcloud environment, a commercial IPU cloud service offered by Graphcore. We also ported two PyTorch neural network models from Graphics Processing Units (GPUs) to IPUs to ensure the efficacy of the software environment. The ported models include the TransCycleGAN model that is used in reconstructing high-resolution images from low-resolution images, and the Hierarchical Autoencoder that is for large-scale high-resolution scientific data compression in climate models. These models were successfully ported on mulitple IPUs using utilities in the Graphcore Poplar software development kit. Increasing the number of IPUs resulted in a considerable enhancement in the model's throughput.

References

[1]

NSF ACES | Texas A&M High Performance Research Computing. Retrieved June 2, 2023 from https://hprc.tamu.edu/aces/

[2]

Abhinand Nasari, Hieu Le, Richard Lawrence, Zhenhua He, Xin Yang, Mario Krell, Alex Tsyplikhin, 2022. Benchmarking the Performance of Accelerators on National Cyberinfrastructure Resources for Artificial Intelli- gence/Machine Learning Workloads. Practice and Experience in Advanced Research Computing (2022), 1–9.https://doi.org/10.1145/3491418.3530772

Digital Library

[3]

Graphcore IPU hardware overview. Retrieved June 2, 2023 from https://docs.graphcore.ai/projects/ipu-overview/en/latest/about_ipu.html

[4]

Compression Vector-quantized Variational Autoencoder. Retrieved June 2, 2023 from https://github.com/abhinand5ai/tccs_torch

[5]

Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. 2015. Deep Learning Face Attributes in the Wild. In Proceedings of International Conference on Computer Vision (ICCV).

Digital Library

[6]

Graphcore IPU example GitHub repository. Retrieved June 2, 2023 from https://github.com/graphcore/examples.

[7]

Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. 2015. ImageNet Large Scale Visual Recognition Challenge. arXiv:1409.0575 [cs.CV]

[8]

Shuo Yang, Ping Luo, Chen Change Loy, and Xiaoou Tang. 2016. WIDER FACE: A Face Detection Benchmark. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]

Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, and Ming-Hsuan Yang. 2022. Restormer: Efficient Transformer for High-Resolution Image Restoration. arXiv:2111.09881 [cs.CV] International conference on curves and surfaces. Springer, 711–730.

[10]

Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A. Efros. 2020. Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. arXiv:1703.10593 [cs.CV]

[11]

SDRBench. Retrieved June 2, 2023 from https://sdrbench.github.io

[12]

K. Zhao, S. Di, X. Liang, S. Li, D. Tao, J. Bessac, Z. Chen, and F. Cappello, “SDRBench: Scientific Data Reduction Benchmark for Lossy Compressors”, International Workshop on Big Data Reduction (IWBDR2020), in conjunction with IEEE Bigdata20.

[13]

iHESP Archives. Retrieve June 2, 2023 from https://ihesp.github.io/archive

[14]

Ping Chang, Shaoqing Zhang, Gokhan Danabasoglu, Stephen G. Yeager, Haohuan Fu, Hong Wang, Frederic S. Castruccio, Yuhu Chen, James Edwards, Dan Fu, Yinglai Jia, Lucas C. Laurindo, Xue Liu, Nan Rosenbloom, R. Justin Small, Gaopeng Xu, Yunhui Zeng, Qiuying Zhang, Julio Bacmeister, David A. Bailey, Xiaohui Duan, Alice K. DuVivier, Dapeng Li, Yuxuan Li, Richard Neale, Achim Stössel, Li Wang, Yuan Zhuang, Allison Baker, Susan Bates, John Dennis, Xiliang Diao, Bolan Gan, Abishek Gopal, Dongning Jia, Zhao Jing, Xiaohui Ma, R. Saravanan, Warren G. Strand, Jian Tao, Haiyuan Yang, Xiaoqi Wang, Zhiqiang Wei, and Lixin Wu. 2020. An Unprecedented Set of High-Resolution Earth System Simulations for Understanding Multiscale Interactions in Climate Variability and Change. Journal of Advances in Modeling Earth Systems 12, 12 (2020), e2020MS002298. https://doi.org/10.1029/2020MS002298 arXiv:https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2020MS002298 e2020MS002298 2020MS002298

[15]

David A Huffman. 1952. A method for the construction of minimum-redundancy codes. Proceedings of the IRE 40, 9 (1952), 1098–1101.

[16]

D.E. Rumelhart, G.E. Hinton, and R.J. Williams, "Learning internal representations by error propagation.", Parallel Distributed Processing. Vol 1: Foundations. MIT Press, Cambridge, MA, 1986.

[17]

Ballard, "Modular learning in neural networks," Proceedings AAAI (1987)

[18]

LeCun, Y. (1987). Modèles connexionistes de l'apprentissage. Ph.D. thesis, Université de Paris VI. 17, 499, 511

[19]

Doyub Kim, Minjae Lee, and Ken Museth. 2022. NeuralVDB: High-resolutionSparse Volume Representation using Hierarchical Neural Networks. arXiv preprint arXiv:2208.04448 (2022).

[20]

Ali Razavi, Aaron Van den Oord, and Oriol Vinyals. 2019. Generating diverse high-fidelity images with vq-vae-2. Advances in neural information processing systems 32 (2019).

[21]

Hieu Le, Hernan Santos, and Jian Tao. 2023. Hierarchical Autoencoder-based Lossy Compression for Large-scale High-resolution Scientific Data. Manuscript submitted for publication.

[22]

Texas AM High Performance Research Computing. Accessed on April 21, 2023. High Performance Research Computing Training. https://hprc.tamu.edu/files/training/2022/Fall/IPU_Training_Labs_Fall2022.pdf.

[23]

Zhenhua He, Sandra Nite, Abhinand, Hieu Le, Jian Tao, Dhruva Chakravorty, Lisa Perez, Honggao Liu. 2023. Development of a Training Framework for Novel Accelerators. Manuscript submitted for publication.

[24]

Christian Ledig, Lucas Theis, Ferenc Husźar, Jose Caballero, Andrew Cunningham, Alejandro Acosta, Andrew Aitken, Alykhan Tejani, Johannes Totz, Zehan Wang, Photo-realistic single image super-resolution using a generative adversarial network. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4681–4690, 2017

[25]

Bee Lim, Sanghyun Son, Heewon Kim, Seungjun Nah, and Kyoung Mu Lee. Enhanced deep residual networks for single image super-resolution, 2017.

[26]

Yulun Zhang, Kunpeng Li, Kai Li, Lichen Wang, Bineng Zhong, and Yun Fu.Image super-resolution using very deep residual channel attention networks. In Proceedings of the European conference on computer vision (ECCV), pages 286–301, 2018

Digital Library

[27]

Xiangyu Chen, Xintao Wang, Jiantao Zhou, and Chao Dong. Activating more pixels in image super-resolution transformer.arXiv preprint arXiv:2205.04437, 2022

[28]

NSF - Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support. Retrieved June 2, 2023 from https://access-ci.org/

[29]

Zhenhua He, Aditi Saluja, Richard Lawrence, Dhruva K. Chakravorty, Francis Dang, Lisa M. Perez, and Honggao Liu. 2023 (Accepted). Performance of Distributed Deep Learning Workloads on a Composable Cyberinfrastructure. In Practice and Experience in Advanced Research Computing (PEARC ‘23), Portland, OR, USA. ACM, New York, NY, USA, 12 pages.

[30]

Edmundo Medina-Gurrola, Dhruva K. Chakravorty, Diana V. Dugas, Tim Cockerill, Lisa M. Perez, Emily Hunt. (2022). Regional Collaborations Supporting Cyberinfrastructure-Enabled Research During a Pandemic: The Structure and Support Plan of the SWEETER CyberTeam. In Practice and Experience in Advanced Research Computing (PEARC ‘22), 4 pages. https://doi.org/10.1145/3491418.3535186

Digital Library

[31]

Richard Lawrence, Dhruva K. Chakravorty, Francis Dang, Lisa M. Perez, Wesley Brashear, Zhenhua He, and Honggao Liu. 2023 (Accepted). Developing Synthetic Applications Benchmarks on Composable Cyberinfrastructure: A Study of Scaling Molecular Dynamics Applications on GPUs. In Practice and Experience in Advanced Research Computing (PEARC ‘23), Portland, OR, USA. ACM, New York, NY, USA, 6 pages.

Cited By

Lawrence RChakravorty DPerez LLiu HHe ZBrashear WWinchell JMao JLu C(2024)Performance of Molecular Dynamics Acceleration Strategies on Composable CyberinfrastructurePractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670631(1-5)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670631
Brashear WReddy VBaum SChakravorty DDang FPerez LLiu H(2024)Exploring the Viability of Composable Architectures to Overcome Memory Limitations in High Performance Computing WorkflowsPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670620(1-4)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670620
Lawrence RChakravorty DPerez LBrashear WHe ZWinchell JLiu H(2024)Container Adoption in Campus High Performance Computing at Texas A&M UniversityPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670550(1-7)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670550
Show More Cited By

Recommendations

Benchmarking the Performance of Accelerators on National Cyberinfrastructure Resources for Artificial Intelligence / Machine Learning Workloads
PEARC '22: Practice and Experience in Advanced Research Computing 2022: Revolutionary: Computing, Connections, You

Upcoming regional and National Science Foundation (NSF)-funded Cyberinfrastructure (CI) resources will give researchers opportunities to run their artificial intelligence / machine learning (AI/ML) workflows on accelerators. To effectively leverage this ...
Insight Gained from Migrating a Machine Learning Model to Intelligence Processing Units
PEARC '24: Practice and Experience in Advanced Research Computing 2024: Human Powered Computing

The discoveries in this paper show that Intelligence Processing Units (IPUs) offer a viable accelerator alternative to GPUs for machine learning (ML) applications within the fields of materials science and battery research. We investigate the process of ...
IPUS An Architecture for the Integrated Processing and Understanding of Signals

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

PEARC '23: Practice and Experience in Advanced Research Computing 2023: Computing for the Common Good

July 2023

519 pages

ISBN:9781450399852

DOI:10.1145/3569951

Editors:
Robert Sinkovits
San Diego Supercomputer Center
,
Alana Romanella
University of Colorado Boulder
,
Shelley Knuth
University of Colorado Boulder
,
Ken Hackworth
Pittsburgh Supercomputing Center
,
Jeff Pummill
University of Arkansas

Copyright © 2023 ACM.

Permission to make digital or hard copies of all or part 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 components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 10 September 2023

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Short-paper
Research
Refereed limited

Funding Sources

NSF (National Science Foundation)

Conference

PEARC '23

Sponsor:

PEARC '23: Practice and Experience in Advanced Research Computing

July 23 - 27, 2023

OR, Portland, USA

Acceptance Rates

Overall Acceptance Rate 133 of 202 submissions, 66%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

7
Total Citations
View Citations
250
Total Downloads

Downloads (Last 12 months)213
Downloads (Last 6 weeks)36

Reflects downloads up to 01 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Lawrence RChakravorty DPerez LLiu HHe ZBrashear WWinchell JMao JLu C(2024)Performance of Molecular Dynamics Acceleration Strategies on Composable CyberinfrastructurePractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670631(1-5)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670631
Brashear WReddy VBaum SChakravorty DDang FPerez LLiu H(2024)Exploring the Viability of Composable Architectures to Overcome Memory Limitations in High Performance Computing WorkflowsPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670620(1-4)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670620
Lawrence RChakravorty DPerez LBrashear WHe ZWinchell JLiu H(2024)Container Adoption in Campus High Performance Computing at Texas A&M UniversityPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670550(1-7)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670550
Brashear WPerez LLeake ENite SPennings MStebenne SLiu HChakravorty D(2024)Cultivating Cyberinfrastructure Careers through Student Engagement at Texas A&M University High Performance Research ComputingPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670544(1-6)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670544
Mishra SChakravorty DPerez LDang FLiu HWitherden F(2024)Impact of Memory Bandwidth on the Performance of AcceleratorsPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670540(1-9)Online publication date: 17-Jul-2024
https://dl.acm.org/doi/10.1145/3626203.3670540
Le HHe ZLe MChakravorty DPerez LChilumuru AYao YChen J(2024)Insight Gained from Migrating a Machine Learning Model to Intelligence Processing UnitsPractice and Experience in Advanced Research Computing 2024: Human Powered Computing10.1145/3626203.3670527(1-9)Online publication date: 17-Jul-2024
https://doi.org/10.1145/3626203.3670527
Lawrence RChakravorty DHe ZDang FPerez LBrashear WLiu H(2023)Developing Synthetic Applications Benchmarks on Composable Cyberinfrastructure: A Study of Scaling Molecular Dynamics Applications on GPUsPractice and Experience in Advanced Research Computing 2023: Computing for the Common Good10.1145/3569951.3597556(216-220)Online publication date: 23-Jul-2023
https://dl.acm.org/doi/10.1145/3569951.3597556

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

HTML Format

View this article in HTML Format.

Get Access

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

Media

Figures

Other

Tables

View Table of Contents