high performance computing on graphics processing units: hgpu.org

hgpu.org » Optimization

Analyzing Use of OpenCL on the Cell Broadband Engine and a Proposal for OpenCL Extensions

Jens Breitbart, Claudia Fohry

View

Tags: Cell processor, Computer science, Distributed computing, Heterogeneous systems, Matrix multiplication, nVidia, nVidia GeForce GTX 280, OpenCL, Optimization, Playstation

September 24, 2011 by hgpu

Automatic Translation of CUDA to OpenCL and Comparison of Performance Optimizations on GPUs

Deepthi Nandakumar

View

Download (PDF)

Tags: ATI, ATI Radeon HD 5870, Benchmarking, Code generation, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia GeForce GTX 280, OpenCL, Optimization, Thesis

September 24, 2011 by hgpu

OpenCL: a viable solution for high-performance medical image reconstruction?

Christian Siegl, H. G. Hofmann, B. Keck, M. Prummer, J. Hornegger

View

Download (PDF)

Tags: ATI, ATI Radeon HD 5870, Image processing, Image reconstruction, nVidia, OpenCL, Optimization, Tesla C1060

September 24, 2011 by hgpu

Integrated Framework for Heterogeneous Embedded Platforms Using OpenCL

Kulin V. Seth

View

Download (PDF)

Tags: Benchmarking, Computer science, DSP, Embedded high-performance computing, Heterogeneous systems, OpenCL, Optimization, Thesis

September 23, 2011 by hgpu

Simple optimizations for an applicative array language for graphics processors

Bradford Larsen

View

Download (PDF)

Tags: Code generation, Compilers, Computer science, CUDA, nVidia, nVidia GeForce 8800 GT, Optimization, Programming Languages, Programming techniques

September 23, 2011 by hgpu

Compact data structure and scalable algorithms for the sparse grid technique

Alin Murarasu, Josef Weidendorfer, Gerrit Buse, Daniel Butnaru, Dirk Pfluger

View

Download (PDF)

Tags: Algorithms, Compression, Computer science, CUDA, Numerical Analysis, nVidia, Optimization, Programming techniques, Tesla C1060

September 23, 2011 by hgpu

Orchestration by approximation: mapping stream programs onto multicore architectures

Sardar M. Farhad, Yousun Ko, Bernd Burgstaller, Bernhard Scholz

View

Download (PDF)

Tags: Algorithms, Benchmarking, Compilers, Computer science, Optimization, Performance, Programming Languages

September 23, 2011 by hgpu

On-the-fly elimination of dynamic irregularities for GPU computing

Eddy Z. Zhang, Yunlian Jiang, Ziyu Guo, Kai Tian, Xipeng Shen

View

Download (PDF)

Tags: Algorithms, Compilers, Computer science, CUDA, nVidia, Optimization, Performance, Programming Languages, Tesla C1060

September 22, 2011 by hgpu

Reducing branch divergence in GPU programs

Tianyi David Han, Tarek S. Abdelrahman

View

Download (PDF)

Tags: Algorithms, Benchmarking, Computer science, CUDA, nVidia, nVidia GeForce GTX 480, Optimization, Programming Languages

September 22, 2011 by hgpu

A case for neuromorphic ISAs

Atif Hashmi, Andrew Nere, James Jamal Thomas, Mikko Lipasti

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, Java, MPI, nVidia, nVidia GeForce 9800 GX2, nVidia GeForce GTX 280, Optimization, Performance

September 22, 2011 by hgpu

Multicore performance optimization using partner cores

Eric Lau, Jason E. Miller, Inseok Choi, Donald Yeung, Saman Amarasinghe, Anant Agarwal

View

Download (PDF)

Tags: Computer science, Optimization, Programming techniques

September 21, 2011 by hgpu

Mint: realizing CUDA performance in 3D stencil methods with annotated C

Didem Unat, Xing Cai, Scott B. Baden

View

Download (PDF)

Source codes

Tags: Algorithms, Code generation, Compilers, Computer science, CUDA, nVidia, OpenMP, Optimization, Package, Performance, Programming Languages, Tesla C1060

September 21, 2011 by hgpu

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

DeepCompile: A Compiler-Driven Approach to Optimizing Distributed Deep Learning Training

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

GigaAPI for GPU Parallelization

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

Advances in Semantic Patching for HPC-oriented Refactorings with Coccinelle

DuoReduce: MLIR's benchmark

Hardware-Assisted Software Testing and Debugging for Heterogeneous Computing

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Analyzing Use of OpenCL on the Cell Broadband Engine and a Proposal for OpenCL Extensions

Automatic Translation of CUDA to OpenCL and Comparison of Performance Optimizations on GPUs

OpenCL: a viable solution for high-performance medical image reconstruction?

Integrated Framework for Heterogeneous Embedded Platforms Using OpenCL

Simple optimizations for an applicative array language for graphics processors

Compact data structure and scalable algorithms for the sparse grid technique

Orchestration by approximation: mapping stream programs onto multicore architectures

On-the-fly elimination of dynamic irregularities for GPU computing

Reducing branch divergence in GPU programs

A case for neuromorphic ISAs

Multicore performance optimization using partner cores

Mint: realizing CUDA performance in 3D stencil methods with annotated C

Recent source codes

PELSI: Power-Efficient Layer-Switched Inference

Ouroboros: Virtualized Queues for dynamic memory management

MSCCL++: A GPU-driven communication stack for scalable AI applications

Benchmark compute shader of Unity against InteropUnityCUDA

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

DuoReduce: MLIR's benchmark

Most viewed papers (last 30 days)