Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2512.09196 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914201480462336 |
|---|---|
| author | Li, Haonan Man, Keyu Kanuparthy, Partha Chen, Hanning Sun, Wei Tallam, Sreen Zhu, Chenguang Zhu, Kevin Qian, Zhiyun |
| author_facet | Li, Haonan Man, Keyu Kanuparthy, Partha Chen, Hanning Sun, Wei Tallam, Sreen Zhu, Chenguang Zhu, Kevin Qian, Zhiyun |
| contents | High-performance GPU kernel optimization remains a critical yet labor-intensive task in modern machine learning workloads. Although Triton, a domain-specific language for GPU programming, enables developers to write efficient kernels with concise code, achieving expert-level performance still requires deep understanding of GPU architectures and low-level performance trade-offs. We present TritonForge, a profiling-guided framework for automated Triton kernel optimization. TritonForge integrates kernel analysis, runtime profiling, and iterative code transformation to streamline the optimization process. By incorporating feedback from profiling results, the system identifies performance bottlenecks, proposes targeted code modifications, and evaluates their impact automatically. Across diverse kernel types, TritonForge achieves up to 5x performance improvement over baseline implementations and on average 1.76x of the cases are successful, providing a foundation for future research in automated GPU performance optimization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_09196 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization Li, Haonan Man, Keyu Kanuparthy, Partha Chen, Hanning Sun, Wei Tallam, Sreen Zhu, Chenguang Zhu, Kevin Qian, Zhiyun Software Engineering High-performance GPU kernel optimization remains a critical yet labor-intensive task in modern machine learning workloads. Although Triton, a domain-specific language for GPU programming, enables developers to write efficient kernels with concise code, achieving expert-level performance still requires deep understanding of GPU architectures and low-level performance trade-offs. We present TritonForge, a profiling-guided framework for automated Triton kernel optimization. TritonForge integrates kernel analysis, runtime profiling, and iterative code transformation to streamline the optimization process. By incorporating feedback from profiling results, the system identifies performance bottlenecks, proposes targeted code modifications, and evaluates their impact automatically. Across diverse kernel types, TritonForge achieves up to 5x performance improvement over baseline implementations and on average 1.76x of the cases are successful, providing a foundation for future research in automated GPU performance optimization. |
| title | TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2512.09196 |