# Fathom: Reference Workloads for Modern Deep Learning Methods

arXiv: https://arxiv.org/abs/1608.06581 · HTML: https://mlsanity.com/paper/1608.06581/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Aug 2016.

## Built on by (most cited first, 20 of 20)

- [A Runtime-Based Computational Performance Predictor for Deep Neural Network Training](https://mlsanity.com/paper/2102.00527/) (2021)
- [MLPerf HPC: A Holistic Benchmark Suite for Scientific Machine Learning on HPC Systems](https://mlsanity.com/paper/2110.11466/) (2021)
- [Benchmarking of DL Libraries and Models on Mobile Devices](https://mlsanity.com/paper/2202.06512/) (2022)
- [FastML Science Benchmarks: Accelerating Real-Time Scientific Edge Machine Learning](https://mlsanity.com/paper/2207.07958/) (2022)
- [Benchmarking Resource Usage for Efficient Distributed Deep Learning](https://mlsanity.com/paper/2201.12423/) (2022)
- [Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud](https://mlsanity.com/paper/2209.08938/) (2022)
- [Mystique: Enabling Accurate and Scalable Generation of Production AI Benchmarks](https://mlsanity.com/paper/2301.04122/) (2022)
- [Hierarchical Training of Deep Neural Networks Using Early Exiting](https://mlsanity.com/paper/2303.02384/) (2023)
- [Demystifying BERT: Implications for Accelerator Design](https://mlsanity.com/paper/2104.08335/) (2021)
- [Experimenting with Emerging RISC-V Systems for Decentralised Machine Learning](https://mlsanity.com/paper/2302.07946/) (2023)
- [GNNBENCH: Fair and Productive Benchmarking for Single-GPU GNN System](https://mlsanity.com/paper/2404.04118/) (2024)
- [I/O in Machine Learning Applications on HPC Systems: A 360-degree Survey](https://mlsanity.com/paper/2404.10386/) (2024)
- [TorchBench: Benchmarking PyTorch with High API Surface Coverage](https://mlsanity.com/paper/2304.14226/) (2023)
- [An MLCommons Scientific Benchmarks Ontology](https://mlsanity.com/paper/2511.05614/) (2025)
- [An Overview of the Data-Loader Landscape: Comparative Performance Analysis](https://mlsanity.com/paper/2209.13705/) (2022)
- [MIGPerf: A Comprehensive Benchmark for Deep Learning Training and Inference Workloads on Multi-Instance GPUs](https://mlsanity.com/paper/2301.00407/) (2023)
- [Beyond Inference: Performance Analysis of DNN Server Overheads for Computer Vision](https://mlsanity.com/paper/2403.12981/) (2024)
- [A Case for 3D Integrated System Design for Neuromorphic Computing & AI Applications](https://mlsanity.com/paper/2103.04852/) (2021)
- [Metadata Representations for Queryable ML Model Zoos](https://mlsanity.com/paper/2207.09315/) (2022)
- [Benchmarking Storage Systems for Machine Learning Workloads Using NIO Bench](https://mlsanity.com/paper/2609.05418/) (2026)

## Abstract (arXiv)

Deep learning has been popularized by its recent successes on challenging artificial intelligence problems. One of the reasons for its dominance is also an ongoing challenge: the need for immense amounts of computational power. Hardware architects have responded by proposing a wide array of promising ideas, but to date, the majority of the work has focused on specific algorithms in somewhat narrow application domains. While their specificity does not diminish these approaches, there is a clear need for more flexible solutions. We believe the first step is to examine the characteristics of cutting edge models from across the deep learning community. Consequently, we have assembled Fathom: a collection of eight archetypal deep learning workloads for study. Each of these models comes from a seminal work in the deep learning community, ranging from the familiar deep convolutional neural network of Krizhevsky et al., to the more exotic memory networks from Facebook's AI research group. Fathom has been released online, and this paper focuses on understanding the fundamental performance characteristics of each model. We use a set of application-level modeling tools built around the TensorFlow deep learning framework in order to analyze the behavior of the Fathom workloads. We present a breakdown of where time is spent, the similarities between the performance profiles of our models, an analysis of behavior in inference and training, and the effects of parallelism on scaling.