paper-with-me

Papers

GPU Performance Portability needs Autotuning

2025-04-30 · Burkhard Ringlein, Thomas Parnell, Radu Stoica

As LLMs grow in complexity, achieving state-of-the-art performance requires tight co-design across algorithms, software, and hardware. Today's reliance on a single dominant platform limits portability, creates vendor lock-in, and raises barriers for new AI hardware. In this work, we make the case for combining just-in-time (JIT) compilation with kernel parameter autotuning to enable portable LLM inference with state-of-the-art performance without code changes. Focusing on flash attention -- a widespread performance critical LLM kernel -- we demonstrate that this approach explores up to 15x more kernel parameter configurations, produces significantly more diverse code across multiple dimensions, and even outperforms vendor-optimized implementations by up to 230%, all while reducing kernel code size by 70x and eliminating manual code optimizations. Our results highlight autotuning as a promising path to unlocking model portability across GPU vendors.

📄 PDF Abstract BibTeX arXiv:2505.03780

Code (1)

IBM/triton-dejavu 공식 구현 pytorch

Tasks

GPU

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Using hardware performance counters to speed up autotuning convergence on GPUs

2021-02-10 · Jiří Filipovič, Jana Hozzová, Amin Nezarat, Jaroslav Oľha 외

Nowadays, GPU accelerators are commonly used to speed up general-purpose computing tasks on a variety of hardware. However, due to the diversity of GPU architectures and processed data, optimization of codes for a partic…

GPUNavigate

A Metaprogramming and Autotuning Framework for Deploying Deep Learning Applications

2016-11-21 · Matthew W. Moskewicz, Ali Jannesari, Kurt Keutzer

In recent years, deep neural networks (DNNs), have yielded strong results on a wide range of applications. Graphics Processing Units (GPUs) have been one key enabling factor leading to the current popularity of DNNs. How…

Deep LearningGPU

Surrogate-based Autotuning for Randomized Sketching Algorithms in Regression Problems

2023-08-30 · Younghyun Cho, James W. Demmel, Michał Dereziński, Haoyun Li 외

Algorithms from Randomized Numerical Linear Algebra (RandNLA) are known to be effective in handling high-dimensional computational problems, providing high-quality empirical performance as well as strong probabilistic gu…

regression

Autotuning PolyBench Benchmarks with LLVM Clang/Polly Loop Optimization Pragmas Using Bayesian Optimization (extended version)

2021-04-27 · Xingfu Wu, Michael Kruse, Prasanna Balaprakash, Hal Finkel 외

In this paper, we develop a ytopt autotuning framework that leverages Bayesian optimization to explore the parameter space search and compare four different supervised learning methods within Bayesian optimization and ev…

Bayesian Optimization

Transfer-Learning-Based Autotuning Using Gaussian Copula

2024-01-09 · Thomas Randall, Jaehoon Koo, Brice Videau, Michael Kruse 외

As diverse high-performance computing (HPC) systems are built, many opportunities arise for applications to solve larger problems than ever before. Given the significantly increased complexity of these HPC systems and ap…

Transfer Learning