Exploiting Dependency and Parallelism: Real-Time Scheduling and Analysis for GPU Tasks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914347178000384 |
|---|---|
| author | Zhang, Yuanhai He, Songyang Gou, Ruizhe Cui, Mingyue Li, Boyang Zhao, Shuai Huang, Kai |
| author_facet | Zhang, Yuanhai He, Songyang Gou, Ruizhe Cui, Mingyue Li, Boyang Zhao, Shuai Huang, Kai |
| contents | With the rapid advancement of Artificial Intelligence, the Graphics Processing Unit (GPU) has become increasingly essential across a growing number of safety-critical application domains. Applying a GPU is indispensable for parallel computing; however, the complex data dependencies and resource contention across kernels within a GPU task may unpredictably delay its execution time. To address these problems, this paper presents a scheduling and analysis method for Directed Acyclic Graph (DAG)-structured GPU tasks. Given a DAG representation, the proposed scheduling scales the kernel-level parallelism and establishes inter-kernel dependencies to provide a reduced and predictable DAG response time. The corresponding timing analysis yields a safe yet nonpessimistic makespan bound without any assumption on kernel priorities. The proposed method is implemented using the standard CUDA API, requiring no additional software or hardware support. Experimental results under synthetic and real-world benchmarks demonstrate that the proposed approach effectively reduces the worst-case makespan and measured task execution time compared to the existing methods up to 32.8% and 21.3%, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20826 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Exploiting Dependency and Parallelism: Real-Time Scheduling and Analysis for GPU Tasks Zhang, Yuanhai He, Songyang Gou, Ruizhe Cui, Mingyue Li, Boyang Zhao, Shuai Huang, Kai Operating Systems With the rapid advancement of Artificial Intelligence, the Graphics Processing Unit (GPU) has become increasingly essential across a growing number of safety-critical application domains. Applying a GPU is indispensable for parallel computing; however, the complex data dependencies and resource contention across kernels within a GPU task may unpredictably delay its execution time. To address these problems, this paper presents a scheduling and analysis method for Directed Acyclic Graph (DAG)-structured GPU tasks. Given a DAG representation, the proposed scheduling scales the kernel-level parallelism and establishes inter-kernel dependencies to provide a reduced and predictable DAG response time. The corresponding timing analysis yields a safe yet nonpessimistic makespan bound without any assumption on kernel priorities. The proposed method is implemented using the standard CUDA API, requiring no additional software or hardware support. Experimental results under synthetic and real-world benchmarks demonstrate that the proposed approach effectively reduces the worst-case makespan and measured task execution time compared to the existing methods up to 32.8% and 21.3%, respectively. |
| title | Exploiting Dependency and Parallelism: Real-Time Scheduling and Analysis for GPU Tasks |
| topic | Operating Systems |
| url | https://arxiv.org/abs/2602.20826 |