Characterizing the Efficiency of Distributed Training: A Power, Performance, and Thermal Perspective
Open Access
- 17 October 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 626-642
Abstract
No abstract availableAll Related Versions
Funding Information
- University Funds ()
This publication has 22 references indexed in Scilit:
- Characterizing Compute-Communication Overlap in GPU-Accelerated Distributed Deep Learning: Performance and Power Implications2025
- Uncovering Real GPU NoC Characteristics: Implications on Interconnect Architecture2024
- Heterogeneous Acceleration Pipeline for Recommendation System Training2024
- ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale2023
- PyTorch2021
- ASTRA-SIM: Enabling SW/HW Co-Design Exploration for Distributed DL Training Platforms2020
- Tartan: Evaluating Modern GPU Interconnect via a Multi-GPU Benchmark Suite2018
- From high-level deep neural models to FPGAs2016
- Dissecting GPU Memory Hierarchy Through MicrobenchmarkingIEEE Transactions on Parallel and Distributed Systems, 2016
- TABLA: A unified template-based framework for accelerating statistical machine learning2016