EUVIMEDEUVIMED entdecken
Crossref

cuPTW: Leveraging Idle Compute Units for Massively Parallel GPU Page Table Walks

Zihang Chen, Tianao Ge, Lieven Eeckhout, Hongyuan Liu, Jiayi Huang

Proceedings of the ACM on Measurement and Analysis of Computing Systems · 29.05.2026 · 10 (2) · 1-26

DOI: 10.1145/3805633 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the ACM on Measurement and Analysis of Computing Systems, online erschienen am 29.05.2026. Verfasst von Zihang Chen und 4 weiteren Autor:innen. Die Arbeit nennt 70 Literaturangaben. Gefördert von: National Natural Science Foundation of China, CCF-Phytium Fund, Guangdong Provincial Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Virtual memory has become a cornerstone of modern GPUs, enabling unified address spaces and advanced memory management techniques. However, the performance of address translation has emerged as a critical bottleneck, particularly under irregular workloads with massive memory footprints, where frequent TLB misses and costly page table walks dominate total memory access latency. Prior work has primarily focused on improving TLB effectiveness or optimizing page table walks through batching and coalescing, but these approaches remain limited by the lack of locality and memory bandwidth constraints. In this work, we propose Compute Unit Page Table Walk ( cuPTW ), a novel address translation architecture that repurposes idle GPU compute resources to accelerate page table walks. Specifically, we introduce (1) a single-threaded synchronous cuPTW , which offloads translation requests to idle functional units within compute units, and (2) two optimizations that further reduce latency and improve throughput by caching page table walks in local data store memory ( cuPTW-SW ) and parallelizing them across multiple SIMD lanes ( cuPTW-MT ). When combined, cuPTW-Full transforms low-parallelism page table walks into a massively parallel computation task. Our evaluation across 15 representative GPU workloads, including deep learning, graph analytics and scientific simulations, demonstrates that cuPTW-Full achieves a performance speedup of 4.43× on average (up to 76.09×) by improving page table walk throughput by 9.92× on average over our baseline GPU architecture. Compared to state-of-the-art GPU address translation proposals, cuPTW-Full achieves a 1.97× to 2.08× average speedup.

Förderung

  • National Natural Science Foundation of China
  • CCF-Phytium Fund
  • Guangdong Provincial Project

Zitieren

APA

Chen, Z., Ge, T., Eeckhout, L., Liu, H., & Huang, J. (2026). cuPTW: Leveraging Idle Compute Units for Massively Parallel GPU Page Table Walks. Proceedings of the ACM on Measurement and Analysis of Computing Systems, 10(2), 1-26. https://doi.org/10.1145/3805633

Vancouver

Chen Z, Ge T, Eeckhout L, Liu H, Huang J. cuPTW: Leveraging Idle Compute Units for Massively Parallel GPU Page Table Walks. Proceedings of the ACM on Measurement and Analysis of Computing Systems. 2026;10(2):1-26. doi:10.1145/3805633

BibTeX
@article{Chen2026,
  author = {Chen, Zihang and Ge, Tianao and Eeckhout, Lieven and Liu, Hongyuan and Huang, Jiayi},
  title = {cuPTW: Leveraging Idle Compute Units for Massively Parallel GPU Page Table Walks},
  journal = {Proceedings of the ACM on Measurement and Analysis of Computing Systems},
  year = {2026},
  volume = {10},
  number = {2},
  pages = {1--26},
  doi = {10.1145/3805633}
}
Kostenlos starten →