A Portable GPU Kernel Performance Modeling Method Based on LLVM IR Dynamic Feature Prediction
Open Access
- 19 May 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 271-279
Abstract
No abstract availableFunding Information
- National Key Research and Development Program of China (2023YFB3001804)
This publication has 26 references indexed in Scilit:
- Many-BSP: an analytical performance model for CUDA kernelsComputing, 2024
- Evaluating execution time predictions on GPU kernels using an analytical model and machine learning techniquesJournal of Parallel and Distributed Computing, 2022
- GPU Static Modeling Using PTX and Deep Structured LearningIEEE Access, 2019
- A Hybrid Framework for Fast and Accurate GPU Performance Estimation through Source-Level Analysis and Trace-Based Simulation2019
- Predicting Execution Time of CUDA Kernel Using Static Analysis2018
- OpenCL Performance Prediction using Architecture-Independent Features2018
- Detection of Outliers Using Interquartile Range Technique from Intrusion Dataset2018
- GPGPU Power Modeling for Multi-domain Voltage-Frequency Scaling2018
- Dynamic GPGPU Power Management Using Adaptive Model Predictive Control2017
- GPGPU performance and power estimation using machine learning2015