Return
Task-to-accelerator mapping for predictable OpenMP applications
DOI:10.1016/j.sysarc.2026.103901.png)
Abstract
En 中文
• Propose an efficient task-to-accelerator mapping for heterogeneous platforms. • Offer a set of lightweight heuristic algorithms for this accelerator mapping. • Apply a suspension-aware mechanism to resolve the busy-waiting problem. • Evaluate the proposal on OpenMP-compatible synthetic graphs under different settings. • Assess the proposal using graphs representative of three real-world applications under different configurations.
Keywords:
OpenMP
Accelerator
Mapping process
Heuristics
Multi-criteria decision analysis
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.1
Papers:
3.0K
Citations:
4.2K


