Seer: Predictive Runtime Kernel Selection for Irregular Problems
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916176256303104 |
|---|---|
| author | Swann, Ryan Osama, Muhammad Sangaiah, Karthik Mahmud, Jalal |
| author_facet | Swann, Ryan Osama, Muhammad Sangaiah, Karthik Mahmud, Jalal |
| contents | Modern GPUs are designed for regular problems and suffer from load imbalance when processing irregular data. Prior to our work, a domain expert selects the best kernel to map fine-grained irregular parallelism to a GPU. We instead propose Seer, an abstraction for producing a simple, reproduceable, and understandable decision tree selector model which performs runtime kernel selection for irregular workloads. To showcase our framework, we conduct a case study in Sparse Matrix Vector Multiplication (SpMV), in which Seer predicts the best strategy for a given dataset with an improvement of 2$\times$ over the best single iteration kernel across the entire SuiteSparse Matrix Collection dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_17017 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Seer: Predictive Runtime Kernel Selection for Irregular Problems Swann, Ryan Osama, Muhammad Sangaiah, Karthik Mahmud, Jalal Distributed, Parallel, and Cluster Computing Modern GPUs are designed for regular problems and suffer from load imbalance when processing irregular data. Prior to our work, a domain expert selects the best kernel to map fine-grained irregular parallelism to a GPU. We instead propose Seer, an abstraction for producing a simple, reproduceable, and understandable decision tree selector model which performs runtime kernel selection for irregular workloads. To showcase our framework, we conduct a case study in Sparse Matrix Vector Multiplication (SpMV), in which Seer predicts the best strategy for a given dataset with an improvement of 2$\times$ over the best single iteration kernel across the entire SuiteSparse Matrix Collection dataset. |
| title | Seer: Predictive Runtime Kernel Selection for Irregular Problems |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2403.17017 |