Performance Characterization of Distributed Deep Learning Strategies: A Quantitative Evaluation of DDP, FSDP, and Parameter Server Architectures on GPU Clusters

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Ovi, Md Sultanul Islam
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917181763092480
author Ovi, Md Sultanul Islam
author_facet Ovi, Md Sultanul Islam
contents Efficiently scaling deep neural networks across GPU clusters requires navigating complex trade-offs between computational throughput, memory utilization, and synchronization overhead. This paper presents a unified empirical evaluation of three dominant distributed training paradigms: Distributed Data Parallel (DDP), Fully Sharded Data Parallel (FSDP), and the Parameter Server (PS) architecture. We conduct side-by-side benchmarking on both high-performance (NVIDIA A100) and commodity-class (NVIDIA A10G) clusters to isolate the impact of communication bandwidth and gang-scheduling dependencies. Our results indicate that while DDP achieves a 2-3x speedup in training throughput for standard architectures, FSDP demonstrates a 4-6x reduction in peak memory usage, validating its utility for memory-constrained environments despite higher communication latency. Furthermore, we evaluate the elasticity of the Parameter Server architecture; while Asynchronous PS reduced training time by up to 28% compared to synchronous approaches, it incurred significant accuracy penalties (ranging from 4% to 17%) due to gradient staleness. We also analyze a modified, staleness-mitigating asynchronous protocol, which we found introduced synchronization overheads that negated throughput gains. These findings provide a decision framework for system designers, highlighting that while DDP remains optimal for homogeneous, gang-scheduled clusters, FSDP and PS offer critical alternatives for memory-bound and heterogeneous environments respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12832
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Performance Characterization of Distributed Deep Learning Strategies: A Quantitative Evaluation of DDP, FSDP, and Parameter Server Architectures on GPU Clusters
Ovi, Md Sultanul Islam
Distributed, Parallel, and Cluster Computing
Efficiently scaling deep neural networks across GPU clusters requires navigating complex trade-offs between computational throughput, memory utilization, and synchronization overhead. This paper presents a unified empirical evaluation of three dominant distributed training paradigms: Distributed Data Parallel (DDP), Fully Sharded Data Parallel (FSDP), and the Parameter Server (PS) architecture. We conduct side-by-side benchmarking on both high-performance (NVIDIA A100) and commodity-class (NVIDIA A10G) clusters to isolate the impact of communication bandwidth and gang-scheduling dependencies. Our results indicate that while DDP achieves a 2-3x speedup in training throughput for standard architectures, FSDP demonstrates a 4-6x reduction in peak memory usage, validating its utility for memory-constrained environments despite higher communication latency. Furthermore, we evaluate the elasticity of the Parameter Server architecture; while Asynchronous PS reduced training time by up to 28% compared to synchronous approaches, it incurred significant accuracy penalties (ranging from 4% to 17%) due to gradient staleness. We also analyze a modified, staleness-mitigating asynchronous protocol, which we found introduced synchronization overheads that negated throughput gains. These findings provide a decision framework for system designers, highlighting that while DDP remains optimal for homogeneous, gang-scheduled clusters, FSDP and PS offer critical alternatives for memory-bound and heterogeneous environments respectively.
title Performance Characterization of Distributed Deep Learning Strategies: A Quantitative Evaluation of DDP, FSDP, and Parameter Server Architectures on GPU Clusters
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2505.12832