# COMET: A Comprehensive Cluster Design Methodology for Distributed Deep Learning Training

arXiv: https://arxiv.org/abs/2211.16648 · HTML: https://mlsanity.com/paper/2211.16648/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 86 references that are in the map.
- First posted on arXiv 30 Nov 2022.

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [LaMDA: Language Models for Dialog Applications](https://mlsanity.com/paper/2201.08239/) (2022)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model](https://mlsanity.com/paper/2201.11990/) (2022)
- [Deep Speech 2: End-to-End Speech Recognition in English and Mandarin](https://mlsanity.com/paper/1512.02595/) (2015)
- [ZeRO-Offload: Democratizing Billion-Scale Model Training](https://mlsanity.com/paper/2101.06840/) (2021)
- [Deep Interest Evolution Network for Click-Through Rate Prediction](https://mlsanity.com/paper/1809.03672/) (2018)
- [Deep Learning Recommendation Model for Personalization and Recommendation Systems](https://mlsanity.com/paper/1906.00091/) (2019)
- [An Empirical Model of Large-Batch Training](https://mlsanity.com/paper/1812.06162/) (2018)
- [ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning](https://mlsanity.com/paper/2104.07857/) (2021)
- [Beyond Data and Model Parallelism for Deep Neural Networks](https://mlsanity.com/paper/1807.05358/) (2018)
- [Mesh-TensorFlow: Deep Learning for Supercomputers](https://mlsanity.com/paper/1811.02084/) (2018)
- [Compressing Large-Scale Transformer-Based Models: A Case Study on BERT](https://mlsanity.com/paper/2002.11985/) (2020)
- [Checkmate: Breaking the Memory Wall with Optimal Tensor Rematerialization](https://mlsanity.com/paper/1910.02653/) (2019)
- [The Architectural Implications of Facebook's DNN-based Personalized Recommendation](https://mlsanity.com/paper/1906.03109/) (2019)
- [Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads](https://mlsanity.com/paper/1901.05758/) (2019)
- [Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models](https://mlsanity.com/paper/2104.05158/) (2021)
- [Understanding Training Efficiency of Deep Learning Recommendation Models at Scale](https://mlsanity.com/paper/2011.05497/) (2020)
- [Deep Learning Training in Facebook Data Centers: Design of Scale-up and Scale-out Systems](https://mlsanity.com/paper/2003.09518/) (2020)
- [M6-10T: A Sharing-Delinking Paradigm for Efficient Multi-Trillion Parameter Pretraining](https://mlsanity.com/paper/2110.03888/) (2021)
- [Enabling Compute-Communication Overlap in Distributed Deep Learning Training Platforms](https://mlsanity.com/paper/2007.00156/) (2020)
- [Themis: A Network Bandwidth-Aware Collective Scheduling Policy for Distributed Training of DL Models](https://mlsanity.com/paper/2110.04478/) (2021)
- [Characterizing Deep Learning Training Workloads on Alibaba-PAI](https://mlsanity.com/paper/1910.05930/) (2019)
- [Hierarchical Roofline Performance Analysis for Deep Learning Applications](https://mlsanity.com/paper/2009.05257/) (2020)
- [TPP: Transparent Page Placement for CXL-Enabled Tiered-Memory](https://mlsanity.com/paper/2206.02878/) (2022)
- [Time-Based Roofline for Deep Learning Performance Analysis](https://mlsanity.com/paper/2009.04598/) (2020)
- [Impact of RoCE Congestion Control Policies on Distributed Training of DNNs](https://mlsanity.com/paper/2207.10898/) (2022)

## Built on by (most cited first, 2 of 2)

- [ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale](https://mlsanity.com/paper/2303.14006/) (2023)
- [MAD Max Beyond Single-Node: Enabling Large Machine Learning Model Acceleration on Distributed Systems](https://mlsanity.com/paper/2310.02784/) (2023)

## Abstract (arXiv)

Modern Deep Learning (DL) models have grown to sizes requiring massive clusters of specialized, high-end nodes to train. Designing such clusters to maximize both performance and utilization--to amortize their steep cost--is a challenging task requiring careful balance of compute, memory, and network resources. Moreover, a plethora of each model's tuning knobs drastically affect the performance, with optimal values often depending on the underlying cluster's characteristics, which necessitates a complex cluster-workload co-design process. To facilitate the design space exploration of such massive DL training clusters, we introduce COMET, a holistic cluster design methodology and workflow to jointly study the impact of parallelization strategies and key cluster resource provisioning on the performance of distributed DL training. We develop a step-by-step process to establish a reusable and flexible methodology, and demonstrate its application with case studies of training large models on cluster configurations of variable compute, memory, and network resources. Our case studies demonstrate COMET's utility in identifying promising architectural optimization directions and guiding system designers in configuring key model and cluster parameters. To illustrate, cluster configuration comparisons identify performance differences of up to 7.7x and highlight performance optimization opportunities of up to 1.4x when employing memory expansion as an optimization technique.