high performance computing on graphics processing units: hgpu.org

hgpu.org » Performance

Exploring Thread Coarsening on FPGA

Mostafa Eghbali Zarch, Reece Neff, Michela Becchi

View

Download (PDF)

Tags: Compilers, Computer science, FPGA, OpenCL, Performance, performance portability

August 28, 2022 by hgpu

Performance analysis of matrix-free conjugate gradient kernels using SYCL

Igor Baratta, Chris Richardson, Garth Wells

View

Download (PDF)

Source codes

Tags: Benchmarking, Computer science, Latency, nVidia, nVidia A100, OpenCL, Package, Performance, SYCL

August 21, 2022 by hgpu

Reducing Synchronous GPU Memory Transfers: Design and implementation of a Futhark compiler optimisation

Philip Jon Børgesen

View

Download (PDF)

Tags: AMD Radeon Instinct MI100, ATI, Computer science, CUDA, nVidia, nVidia A100, OpenCL, Performance, Thesis

July 17, 2022 by hgpu

High Performance Simulation for Scalable Multi-Agent Reinforcement Learning

Jordan Langham-Lopez, Sebastian M. Schmon, Patrick Cannon

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, nVidia, nVidia A100, nVidia GeForce GTX 1650, Performance

July 17, 2022 by hgpu

Evaluation of Intel’s DPC++ Compatibility Tool in heterogeneous computing

Germán Castaño, Youssef Faqir-Rhazoui, Carlos García, Manuel Prieto-Matías

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, FPGA, Heterogeneous systems, nVidia, Package, Performance, SYCL

July 3, 2022 by hgpu

Deep Learning Models on CPUs: A Methodology for Efficient Training

Quchen Fu, Ramesh Chukka, Keith Achorn, Thomas Atta-fosu, Deepak R. Canchi, Zhongwei Teng, Jules White, Douglas C. Schmidt

View

Download (PDF)

Tags: Computer science, Deep learning, Performance, PyTorch, Visualization

June 26, 2022 by hgpu

CoDL: Efficient CPU-GPU Co-execution for Deep Learning Inference on Mobile Devices

Fucheng Jia, Deyu Zhang, Ting Cao, Shiqi Jiang, Yunxin Liu, Ju Ren, Yaoxue Zhang

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Heterogeneous systems, Package, Performance

June 19, 2022 by hgpu

On the Compilation Performance of Current SYCL Implementations

Peter Thoman, Facundo Molina Heredia, Thomas Fahringer

View

Download (PDF)

Source codes

Tags: Code generation, Compilers, Computer science, Heterogeneous systems, OpenCL, Package, Performance, SYCL

June 12, 2022 by hgpu

Modeling GPU Dynamic Parallelism for Self Similar Density Workloads

Felipe A. Quezada, Cristóbal A. Navarro, Miguel Romero, Cristhian Aguilera

View

Download (PDF)

Tags: Computer science, CUDA, nVidia, nVidia A100, nVidia GeForce Titan V, nVidia Jetson NX, nVidia Titan RTX, Performance

June 12, 2022 by hgpu

Dissecting Tensor Cores via Microbenchmarks: Latency, Throughput and Numerical Behaviors

Wei Sun, Ang Li, Tong Geng, Sander Stuijk, Henk Corporaal

View

Download (PDF)

Tags: Benchmarking, Computer science, Matrix multiplication, nVidia, nVidia A100, nVidia GeForce RTX 2080, nVidia GeForce RTX 2080 Ti, Performance, PTX, Sparse matrix

June 12, 2022 by hgpu

Onesweep: A Faster Least Significant Digit Radix Sort for GPUs

Andy Adinets, Duane Merrill

View

Download (PDF)

Source codes

Tags: Algorithms, Computer science, CUDA, nVidia, nVidia A100, Package, Performance, Sorting

June 12, 2022 by hgpu

SOL: Reducing the Maintenance Overhead for Integrating Hardware Support into AI Frameworks

Nicolas Weber

View

Download (PDF)

Tags: AI, Artificial intelligence, Computer science, CUDA, Machine learning, Neural networks, nVidia, nVidia GeForce RTX 2080, Performance

May 29, 2022 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

Exploring Thread Coarsening on FPGA

Performance analysis of matrix-free conjugate gradient kernels using SYCL

Reducing Synchronous GPU Memory Transfers: Design and implementation of a Futhark compiler optimisation

High Performance Simulation for Scalable Multi-Agent Reinforcement Learning

Evaluation of Intel’s DPC++ Compatibility Tool in heterogeneous computing

Deep Learning Models on CPUs: A Methodology for Efficient Training

CoDL: Efficient CPU-GPU Co-execution for Deep Learning Inference on Mobile Devices

On the Compilation Performance of Current SYCL Implementations

Modeling GPU Dynamic Parallelism for Self Similar Density Workloads

Dissecting Tensor Cores via Microbenchmarks: Latency, Throughput and Numerical Behaviors

Onesweep: A Faster Least Significant Digit Radix Sort for GPUs

SOL: Reducing the Maintenance Overhead for Integrating Hardware Support into AI Frameworks

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)