Towards Universal Performance Modeling for Machine Learning Training on Multi-GPU Platforms

doi:10.1109/TPDS.2024.3507814

Towards Universal Performance Modeling for Machine Learning Training on Multi-GPU Platforms

2025-02-01

0

PRE

AI

DOI：10.1109/TPDS.2024.3507814

Original

Original request for help

Save

Abstract

En 中文

Characterizing and predicting the training performance of modern machine learning (ML) workloads on compute systems with compute and communication spread between CPUs, GPUs, and network devices is not only the key to optimization and planning but also a complex goal to achieve. The primary challenges include the complexity of synchronization and load balancing between CPUs and GPUs, the variance in input data distribution, and the use of different communication devices and topologies (e.g., NVLink, PCIe, network cards) that connect multiple compute devices, coupled with the desire for flexible training configurations. Built on top of our prior work for single-GPU platforms, we address these challenges and enable multi-GPU performance modeling(1) by incorporating (1) data-distribution-aware performance models for embedding table lookup, and (2) data movement prediction of communication collectives, into our upgraded performance modeling pipeline equipped with inter-and intra-rank synchronization for ML workloads trained on multi-GPU platforms. Beyond accurately predicting the per-iteration training time of deep learning recommendation models (DLRM) models with random configurations with a geomean error of 5.21% on two multi-GPU platforms, our prediction pipeline generalizes well to other types of ML workloads, such as Transformer-based natural language processing (NLP) models with a geomean error of 3.00%. Moreover, even without actually running ML workloads like DLRMs on the hardware, it is capable of generating insights such as quickly selecting the fastest embedding table sharding configuration (with a success rate of 85%).

Keywords:

Training

Computational modeling

Predictive models

Performance evaluation

Data models

Pipelines

Graphics processing units

Sharding

Synchronization

Parallel processing

DLRM

machine learning

model training

multi -GPU

NLP

performance modeling

Towards Universal Performance Modeling for Machine Learning Training on Multi-GPU Platforms

Abstract

Journal

Organization