EUVIMEDEUVIMED entdecken
Crossref

GOLDYLOC: Global Optimizations & Lightweight Dynamic Logic for Concurrency

Suchita Pati, Shaizeen Aga, Nuwan Jayasena, Matthew Sinclair

ACM Transactions on Architecture and Code Optimization · 30.06.2025 · 22 (2) · 1-28

DOI: 10.1145/3730584 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, 2025-07. Verfasst von Suchita Pati und 3 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 130 Literaturangaben. Gefördert von: University of Wisconsin-Madison by the Vilas Life Cycle Professorship, Fall Research Competition, National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern accelerators like GPUs increasingly execute independent operations concurrently to improve the device’s compute utilization. However, effectively harnessing it on GPUs for important primitives such as general matrix multiplications (GEMMs) remains challenging. Although modern GPUs have significant hardware and software GEMM support, their kernel implementations and optimizations typically assume each kernel executes in isolation and can utilize all GPU resources. This approach is highly efficient when kernels execute in isolation, but causes significant resource contention and slowdowns when kernels execute concurrently. Moreover, current approaches often only statically expose and control parallelism within an application, without considering runtime information such as varying input size and concurrent applications—often exacerbating contention. These issues limit performance benefits from concurrently executing independent operations. Accordingly, we propose GOLDYLOC , which considers the global resources across all concurrent operations to identify performant GEMM kernels, which we call globally optimized (GO)-Kernels. GOLDYLOC also introduces a lightweight dynamic logic which considers the dynamic execution environment for available parallelism and input sizes to execute performant combinations of concurrent GEMMs on the GPU. Overall, GOLDYLOC improves the performance of concurrent GEMMs on a real GPU by up to 2× (18% geomean per workload) versus the default concurrency approach and provides up to 2.5× (43% geomean per workload) speedup over sequential execution.

Förderung

  • University of Wisconsin-Madison by the Vilas Life Cycle Professorship
  • Fall Research Competition
  • National Science Foundation

Zitieren

APA

Pati, S., Aga, S., Jayasena, N., & Sinclair, M. (2025). GOLDYLOC: Global Optimizations & Lightweight Dynamic Logic for Concurrency. ACM Transactions on Architecture and Code Optimization, 22(2), 1-28. https://doi.org/10.1145/3730584

Vancouver

Pati S, Aga S, Jayasena N, Sinclair M. GOLDYLOC: Global Optimizations & Lightweight Dynamic Logic for Concurrency. ACM Transactions on Architecture and Code Optimization. 2025;22(2):1-28. doi:10.1145/3730584

BibTeX
@article{Pati2025,
  author = {Pati, Suchita and Aga, Shaizeen and Jayasena, Nuwan and Sinclair, Matthew},
  title = {GOLDYLOC: Global Optimizations & Lightweight Dynamic Logic for Concurrency},
  journal = {ACM Transactions on Architecture and Code Optimization},
  year = {2025},
  volume = {22},
  number = {2},
  pages = {1--28},
  doi = {10.1145/3730584}
}
Kostenlos starten →