Cooperative Distributed GPU Power Capping for Deep Learning Clusters
| Autor: | Limei Peng, Dong-Ki Kang, Yun-Gi Ha, Chan-Hyun Youn |
|---|---|
| Rok vydání: | 2022 |
| Předmět: |
Profiling (computer programming)
Online model Computational complexity theory Artificial neural network business.industry Computer science Deep learning Parallel computing Power budget Task (computing) Control and Systems Engineering Artificial intelligence Electrical and Electronic Engineering Frequency scaling business |
| Zdroj: | IEEE Transactions on Industrial Electronics. 69:7244-7254 |
| ISSN: | 1557-9948 0278-0046 |
| DOI: | 10.1109/tie.2021.3095790 |
| Popis: | The recent GPU-based clusters that handle deep learning (DL) tasks have the features of GPU device heterogeneity, a variety of deep neural network (DNN) models, and high computational complexity. Thus, the traditional power capping methods for CPU-based clusters or small-scale GPU devices do not apply to the GPU-based clusters handling DL tasks. This paper develops a cooperative distributed GPU power capping (CD-GPC) system for GPU-based clusters, aiming to minimize the training completion time of invoked DL tasks without exceeding the limited power budget. Specifically, we first design the frequency scaling (FS) approach using the online model estimation based on the recursive least square (RLS) method. This approach achieves the accurate tuning for DL task training time and power usage of GPU devices without needing offline profiling. Then, we formulate the proposed FS problem as a Lagrangian dual decomposition-based economic model predictive control (EMPC) problem for large-scale heterogeneous GPU clusters. We conduct both the NVIDIA GPU-based lab-scale real experiments and real job trace-based simulation experiments for performance evaluation. Experimental results validate that the proposed system improves the power capping accuracy to have a mean absolute error |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|