high performance computing on graphics processing units: hgpu.org

hgpu.org » Applications

LazyTensor: combining eager execution with domain-specific compilers

Alex Suhan, Davide Libenzi, Ailing Zhang, Parker Schuh, Brennan Saeta, Jie Young Sohn, Denys Shabalin

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Package, Programming Languages, TPU

March 7, 2021 by hgpu

Benchmarking Modern Edge Devices for AI Applications

Pilsung Kang, Jongmin Jo

View

Download (PDF)

Tags: Artificial intelligence, Benchmarking, Computer science, Deep learning, TPU

March 7, 2021 by hgpu

Integrating Accelerators in Heterogeneous Systems

Ján Veselý

View

Download (PDF)

Tags: Computer science, CUDA, Heterogeneous systems, nVidia, Python, Thesis

March 7, 2021 by hgpu

Scalable communication for high-order stencil computations using CUDA-aware MPI

Johannes Pekkilä, Miikka S. Väisälä, Maarit J. Käpylä, Matthias Rheinhardt, Oskar Lappi

View

Download (PDF)

Tags: Computer science, CUDA, Finite difference, Magnetohydrodynamics, MPI, nVidia, Tesla V100

March 7, 2021 by hgpu

Neural Network Libraries: A Deep Learning Framework Designed from Engineers’ Perspectives

Akio Hayakawa, Masato Ishii, Yoshiyuki Kobayashi, Akira Nakamura, Takuya Narihira, Yukio Obuchi, Andrew Shin, Takuya Yashima, Kazuki Yoshiyama

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Neural networks, nVidia, nVidia DGX-1, Package, Tesla V100

March 7, 2021 by hgpu

Accelerating AutoDock4 with GPUs and Gradient-Based Local Search

Diogo Santos-Martins, Leonardo Solis-Vasquez, Andreas Koch, and Stefano Forli

View

Download (PDF)

Source codes

Tags: Algorithms, Chemistry, Macromolecule, molecular docking, OpenCL, Package, Physics

February 28, 2021 by hgpu

Acceleration of Intrusion Detection in Encrypted Network Traffic Using Heterogeneous Hardware

Eva Papadogiannaki, Sotiris Ioannidis

View

Download (PDF)

Tags: Computer science, Filtering, Heterogeneous systems, Intel UHD 630, nVidia, nVidia GeForce GTX 980, OpenCL, Security

February 28, 2021 by hgpu

Multi-GPU performance optimization of a computational fluid dynamics code using OpenACC

Weicheng Xue, Christoper J. Roy

View

Download (PDF)

Tags: cfd, Fluid dynamics, MPI, nVidia, OpenACC, Tesla C2050, Tesla M2015

February 28, 2021 by hgpu

BASEMENT v3: a modular freeware for river process modelling over multiple computational backends

Davide Vanzo, Samuel Peter, Lukas Vonwiller, Matthias Buergler, Manuel Weberndorfer, Annunziato Siviglia, Daniel Conde, David F. Vetsch

View

Download (PDF)

Source codes

Tags: cfd, Computational Physics, CUDA, Fluid dynamics, nVidia, nVidia GeForce GTX 1050 Ti, nVidia GeForce GTX 1080 Ti, Package, Tesla P100

February 28, 2021 by hgpu

GPU-aware Communication with UCX in Parallel Programming Models: Charm++, MPI, and Python

Jaemin Choi, Zane Fink, Sam White, Nitin Bhat, David F. Richards, Laxmikant V. Kale

View

Download (PDF)

Source codes

Tags: Benchmarking, Computer science, CUDA, Distributed computing, MPI, nVidia, Package, Python, Tesla V100

February 28, 2021 by hgpu

Using hardware performance counters to speed up autotuning convergence on GPUs

Jiří Filipovič, Jana Hozzová, Amin Nezarat, Jaroslav Oľha, Filip Petrovič

View

Download (PDF)

Source codes

Tags: Auto-Tuning, Computer science, CUDA, HPC, Machine learning, nVidia, nVidia GeForce GTX 1070, nVidia GeForce RTX 2080, OpenCL, Package, Performance, Vulkan

February 23, 2021 by hgpu

Optimal program variant generation for hybrid manycore systems

Cristian Urlea

View

Download (PDF)

Tags: Computer science, Design space exploration, FPGA, Heterogeneous systems, OpenCL, performance portability, Thesis

February 21, 2021 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

Applications

LazyTensor: combining eager execution with domain-specific compilers

Benchmarking Modern Edge Devices for AI Applications

Integrating Accelerators in Heterogeneous Systems

Scalable communication for high-order stencil computations using CUDA-aware MPI

Neural Network Libraries: A Deep Learning Framework Designed from Engineers’ Perspectives

Accelerating AutoDock4 with GPUs and Gradient-Based Local Search

Acceleration of Intrusion Detection in Encrypted Network Traffic Using Heterogeneous Hardware

Multi-GPU performance optimization of a computational fluid dynamics code using OpenACC

BASEMENT v3: a modular freeware for river process modelling over multiple computational backends

GPU-aware Communication with UCX in Parallel Programming Models: Charm++, MPI, and Python

Using hardware performance counters to speed up autotuning convergence on GPUs

Optimal program variant generation for hybrid manycore systems

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)