high performance computing on graphics processing units: hgpu.org

hgpu.org » Applications » Computer science » Automatic Termination Analysis for GPU Kernels

Automatic Termination Analysis for GPU Kernels

Jeroen Ketema, Alastair F. Donaldson

Imperial College London

14th International Workshop on Termination (WST 2014), 2014

BibTeX

Download (PDF)

View

Source

1643

views

We describe a method for proving termination of massively parallel GPU kernels. An implementation in KITTeL is able to show termination of 94% of the 598 kernels in our benchmark suite.

Tags: Computer science, CUDA, nVidia, OpenCL, Programming techniques

May 26, 2014 by hgpu

No votes yet.

Please wait...

HPCTransCompile: An AI Compiler Generated Dataset for High-Performance CUDA Transpilation and LLM Preliminary Exploration

Efficient Graph Embedding at Scale: Optimizing CPU-GPU-SSD Integration

Can Large Language Models Predict Parallel Code Performance?

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Automatic Termination Analysis for GPU Kernels

Recent source codes

HPCTransCompile: An AI Compiler Generated Dataset for High-Performance CUDA Transpilation and LLM Preliminary Exploration

chemtrain: Training Molecular Dynamics Potentials in JAX

microSYCL: SYCL micro-benchmarks repository

XaaS containers

SYCL Container

CASS: Cuda-Amd aSSembly

Cluser of smartphones for edge computing application using TensorFlow

CFAL-bench

Efficient Graph Embedding at Scale: Optimizing CPU-GPU-SSD Integration

Can Large Language Models Predict Parallel Code Performance?

Most viewed papers (last 30 days)

Automatic Termination Analysis for GPU Kernels

Share this:

Recent source codes

Most viewed papers (last 30 days)