EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments

Sichao Chen, Desheng Wang, Weizhe Zhang, Meng Hao, Yu-Chu Tian, Hui He

ACM Transactions on Architecture and Code Optimization · 30.06.2026 · 23 (2) · 1-25

DOI: 10.1145/3807454 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 25.06.2026. Verfasst von Sichao Chen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

GPU sharing is commonly employed in GPU clusters to improve utilization, with spatial sharing being one of the most widely adopted techniques. However, spatial sharing can lead to resource interference, making task execution times difficult to predict. Predictable execution times for each task are crucial in GPU cluster management and task scheduling. In this article, we propose a performance predictor for multi-DNN training tasks in GPU spatial sharing environments. We first conduct experiments on spatial sharing for multiple DNN workloads on a single GPU, demonstrating that concurrent execution of multiple tasks improves overall performance and GPU resource utilization compared to serial execution. By analyzing warp stall reasons collected during task execution, we investigate the interference for computation and memory resources under MPS on GPUs. Finally, we design a performance predictor that predicts the execution time of a target DNN training task when it runs concurrently with other tasks under GPU spatial sharing via MPS. The predictor is capable of predicting the execution time of each task for previously unseen combinations of DNN training tasks. Extensive evaluations on modern GPUs show that compared to other baseline methods, our approach exhibits higher prediction accuracy, as well as improved stability and robustness. Experiments on multiple GPU architectures, as well as at higher concurrency levels, further demonstrate that our method possesses strong generalization and scalability. We also conducted a performance analysis under diverse workload pattern and a case study to validate the practical applicability of our predictor in real scheduling environments.

Zitieren

APA

Chen, S., Wang, D., Zhang, W., Hao, M., Tian, Y., & He, H. (2026). Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments. ACM Transactions on Architecture and Code Optimization, 23(2), 1-25. https://doi.org/10.1145/3807454

Vancouver

Chen S, Wang D, Zhang W, Hao M, Tian Y, He H. Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments. ACM Transactions on Architecture and Code Optimization. 2026;23(2):1-25. doi:10.1145/3807454

BibTeX
@article{Chen2026,
  author = {Chen, Sichao and Wang, Desheng and Zhang, Weizhe and Hao, Meng and Tian, Yu-Chu and He, Hui},
  title = {Performance Prediction of Concurrent DNN Training Tasks in GPU Spatial Sharing Environments},
  journal = {ACM Transactions on Architecture and Code Optimization},
  year = {2026},
  volume = {23},
  number = {2},
  pages = {1--25},
  doi = {10.1145/3807454}
}
Kostenlos starten →