EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MixDNN: Scheduling Optimization of Distributed DNN Training in Heterogeneous GPU Environments

Zhang Xiao, Ye Pengwei, Luo Xuwen, Wang Yumeng

Electronics · 11.11.2026 · 15 (22) · 5236

DOI: 10.3390/electronics15225236 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 11.11.2026. Verfasst von Zhang Xiao und 3 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundamental Research Funds for the Central Universities, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid advancement of deep neural networks (DNN), the scale of workloads continues to expand, exhibiting diverse training characteristics. Traditional scheduling frameworks for PyTorch and TensorFlow, such as Kubernetes, typically assume uniform training workloads across samples within a given task. Consequently, existing frameworks adopt static parallelism strategies throughout the entire training process, leading to suboptimal performance in heterogeneous distributed training environments. As model complexity and GPU heterogeneity grow, this rigidity results in inefficient resource utilization and degraded scalability. To address these challenges, we propose MixDNN, an event-driven rolling-horizon scheduler for distributed DNN models in heterogeneous GPU environments. First, MixDNN divides submitted jobs into two categories, hard-model and soft-model. Then, MixDNN associates utility with different resource requests to represent GPU computing performance. At each job-arrival or job-completion event, MixDNN selects execution configurations and planned start times for waiting jobs by maximizing overall utility subject to GPU-capacity constraints. Running jobs are not preempted, migrated, or reconfigured. In the evaluation, MixDNN reduces parallel training time by up to 11.45% compared to Kubernetes and up to 70% compared to static scheduling, especially for distributed DNN models in heterogeneous GPU environments.

Förderung

  • Fundamental Research Funds for the Central Universities
  • National Natural Science Foundation of China

Zitieren

APA

Xiao, Z., Pengwei, Y., Xuwen, L., & Yumeng, W. (2026). MixDNN: Scheduling Optimization of Distributed DNN Training in Heterogeneous GPU Environments. Electronics, 15(22), 5236. https://doi.org/10.3390/electronics15225236

Vancouver

Xiao Z, Pengwei Y, Xuwen L, Yumeng W. MixDNN: Scheduling Optimization of Distributed DNN Training in Heterogeneous GPU Environments. Electronics. 2026;15(22):5236. doi:10.3390/electronics15225236

BibTeX
@article{Xiao2026,
  author = {Xiao, Zhang and Pengwei, Ye and Xuwen, Luo and Yumeng, Wang},
  title = {MixDNN: Scheduling Optimization of Distributed DNN Training in Heterogeneous GPU Environments},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {5236},
  doi = {10.3390/electronics15225236}
}
Kostenlos starten →