English

Applying the Roofline model for Deep Learning performance optimizations

Distributed, Parallel, and Cluster Computing 2020-09-24 v1 Artificial Intelligence Performance
Authors: Jacek Czaja , Michal Gallus , Joanna Wozna , Adam Grygielski , Luo Tao
View on arXiv ↗ PDF ↗

Abstract

In this paper We present a methodology for creating Roofline models automatically for Non-Unified Memory Access (NUMA) using Intel Xeon as an example. Finally, we present an evaluation of highly efficient deep learning primitives as implemented in the Intel oneDNN Library.

Cite

@article{arxiv.2009.11224,
  title  = {Applying the Roofline model for Deep Learning performance optimizations},
  author = {Jacek Czaja and Michal Gallus and Joanna Wozna and Adam Grygielski and Luo Tao},
  journal= {arXiv preprint arXiv:2009.11224},
  year   = {2020}
}

Comments

oneDNN library analysis with roofline model

Related papers

View all related →
Distributed, Parallel, and Cluster Computing · Computer Science
Time-Based Roofline for Deep Learning Performance Analysis
Yunsong Wang, Charlene Yang, Steven Farrell, Yan Zhang +2
2020-09-24
Distributed, Parallel, and Cluster Computing · Computer Science
Hierarchical Roofline Performance Analysis for Deep Learning Applications
Charlene Yang, Yunsong Wang, Steven Farrell, Thorsten Kurth +1
2020-11-26
Distributed, Parallel, and Cluster Computing · Computer Science
Ridgeline: A 2D Roofline Model for Distributed Systems
Fabio Checconi, Jesmin Jahan Tithi, Fabrizio Petrini
2025-11-18
Machine Learning · Computer Science
Deep Learning Models on CPUs: A Methodology for Efficient Training
Quchen Fu, Ramesh Chukka, Keith Achorn, Thomas Atta-fosu +4
2023-06-21
Distributed, Parallel, and Cluster Computing · Computer Science
Hierarchical Roofline Analysis: How to Collect Data using Performance Tools on Intel CPUs and NVIDIA GPUs
Charlene Yang
2020-10-06
Distributed, Parallel, and Cluster Computing · Computer Science
Intel nGraph: An Intermediate Representation, Compiler, and Executor for Deep Learning
Scott Cyphers, Arjun K. Bansal, Anahita Bhiwandiwalla, Jayaram Bobba +17
2018-01-31
Hardware Architecture · Computer Science
How to keep pushing ML accelerator performance? Know your rooflines!
Marian Verhelst, Luca Benini, Naveen Verma
2025-05-26
Machine Learning · Computer Science
Towards a learning-based performance modeling for accelerating Deep Neural Networks
Damiano Perri, Paolo Sylos Labini, Osvaldo Gervasi, Sergio Tasso +1
2022-12-12
Machine Learning · Computer Science
A Deep Learning Inference Scheme Based on Pipelined Matrix Multiplication Acceleration Design and Non-uniform Quantization
Yuyang Zhang, Dik Hin Leung, Min Guo, Yijia Xiao +5
2021-10-12
Signal Processing · Electrical Eng. & Systems
GPU-Accelerated Machine Learning in Non-Orthogonal Multiple Access
Daniel Schäufele, Guillermo Marcus, Nikolaus Binder, Matthias Mehlhose +2
2022-06-14
Machine Learning · Computer Science
A novel Deep Neural Network architecture for non-linear system identification
Luca Zancato, Alessandro Chiuso
2021-06-08
Distributed, Parallel, and Cluster Computing · Computer Science
Pagoda: An Energy and Time Roofline Study for DNN Workloads on Edge Accelerators
Prashanthi S. K., Kunal Kumar Sahoo, Amartya Ranjan Saikia, Pranav Gupta +3
2025-09-25
Machine Learning · Computer Science
NullaNet: Training Deep Neural Networks for Reduced-Memory-Access Inference
Mahdi Nazemi, Ghasem Pasandi, Massoud Pedram
2018-08-28
Computer Vision and Pattern Recognition · Computer Science
The Sparsity Roofline: Understanding the Hardware Limits of Sparse Neural Networks
Cameron Shinn, Collin McCarthy, Saurav Muralidharan, Muhammad Osama +1
2023-11-08
Machine Learning · Computer Science
Addressing the Memory Bottleneck in AI Model Training
David Ojika, Bhavesh Patel, G. Anthony Reina, Trent Boyer +2
2020-03-20
Distributed, Parallel, and Cluster Computing · Computer Science
Metrics and Design of an Instruction Roofline Model for AMD GPUs
Matthew Leinhauser, René Widera, Sergei Bastrakov, Alexander Debus +2
2021-11-11
Machine Learning · Computer Science
Deep learning architectures for inference of AC-OPF solutions
Thomas Falconer, Letif Mones
2020-12-02
Hardware Architecture · Computer Science
Dynamic DNNs Meet Runtime Resource Management on Mobile and Embedded Platforms
Lei Xun, Bashir M. Al-Hashimi, Jonathon Hare, Geoff V. Merrett
2022-06-08
Distributed, Parallel, and Cluster Computing · Computer Science
CARM Tool: Cache-Aware Roofline Model Automatic Benchmarking and Application Analysis
José Morgado, Leonel Sousa, Aleksandar Ilic
2026-05-29
Distributed, Parallel, and Cluster Computing · Computer Science
Autotuning Benchmarking Techniques: A Roofline Model Case Study
Jacob Odgård Tørring, Jan Christian Meyer, Anne C. Elster
2021-03-19
Neural and Evolutionary Computing · Computer Science
A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers
Thomas E. Potok, Catherine Schuman, Steven R. Young, Robert M. Patton +5
2017-07-17
Machine Learning · Computer Science
Leveraging the HW/SW Optimizations and Ecosystems that Drive the AI Revolution
Humberto Carvalho, Pavel Zaykov, Asim Ukaye
2022-08-05
Machine Learning · Computer Science
Memory-Optimized Once-For-All Network
Maxime Girard, Victor Quétu, Samuel Tardieu, Van-Tam Nguyen +1
2024-09-11
Machine Learning · Computer Science
Towards Memory-Efficient Neural Networks via Multi-Level in situ Generation
Jiaqi Gu, Hanqing Zhu, Chenghao Feng, Mingjie Liu +3
2021-09-07
Distributed, Parallel, and Cluster Computing · Computer Science
Tuna: A Static Analysis Approach to Optimizing Deep Neural Networks
Yao Wang, Xingyu Zhou, Yanming Wang, Rui Li +2
2021-05-18
R2 v1 2026-06-23T18:44:51.813Z