high performance computing on graphics processing units: hgpu.org

hgpu.org » Performance

Why is FPGA-GPU Heterogeneity the Best Option for Embedded Deep Neural Networks?

Walther Carballo-Hernández, Maxime Pelcat, François Berry

View

Tags: Computer science, Deep learning, FPGA, Heterogeneous systems, Neural networks, nVidia, nVidia Jetson TX2, Performance

February 7, 2021 by hgpu

Efficient MPI-based Communication for GPU-Accelerated Dask Applications

Aamir Shafi, Jahanzeb Maqbool Hashmi, Hari Subramoni, Dhabaleswar K. (DK) Panda

View

Download (PDF)

Source codes

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

January 31, 2021 by hgpu

Interactive Parallelization of C Programs in SAPFOR

Nikita Kataev

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, LLVM, nVidia, nVidia GeForce GTX 1660 Ti, OpenCL, Package, Performance

January 3, 2021 by hgpu

Directive-Based Data Partitioning and Pipelining and Auto-Tuning for High-Performance GPU Computing

Xuewen Cui

View

Download (PDF)

Tags: Artificial intelligence, Auto-Tuning, Computer science, CUDA, HPC, Machine learning, nVidia, OpenMP, Performance, Tesla K40, Tesla P100, Tesla V100, Thesis

December 20, 2020 by hgpu

Exploring FPGA Optimizations to Compute Sparse Numerical Linear Algebra Kernels

Federico Favaro, Ernesto Dufrechou, Pablo Ezzatti, Juan P. Oliver

View

Download (PDF)

Tags: Computer science, FPGA, Linear Algebra, OpenCL, Optimization, Performance, Sparse

December 6, 2020 by hgpu

Toward Accurate Platform-Aware Performance Modeling for Deep Neural Networks

Chuan-Chi Wang, Ying-Chiao Liao, Ming-Chang Kao, Wen-Yew Liang, Shih-Hao Hung

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, nVidia GeForce GTX 1080 Ti, Performance

December 6, 2020 by hgpu

Evaluating the Performance and Portability of Contemporary SYCL Implementations

Beau Johnston, Jeffrey S. Vetter, Josh Milthorpe

View

Download (PDF)

Source codes

Tags: AMD Radeon VII, ATI, Benchmarking, Computer science, CUDA, Heterogeneous systems, HIP, Matrix multiplication, nVidia, OpenCL, Package, Performance, Tesla P100

November 29, 2020 by hgpu

AZP: Automatic Specialization for Zero Values in Gaming Applications

Mark W. Stephenson, Ram Rangan

View

Download (PDF)

Tags: Computer science, Games, HLSL, nVidia, nVidia GeForce RTX 2080, OpenGL, Performance, Vulkan

November 29, 2020 by hgpu

Ginkgo – A Math Library designed for Platform Portability

Terry Cojean, Yu-Hsiang "Mike" Tsai, Hartwig Anzt

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, HIP, nVidia, Package, Performance, performance portability

November 22, 2020 by hgpu

Design and Performance Evaluation of Optimizations for OpenCL FPGA Kernels

Anthony M. Cabrera, Roger D. Chamberlain

View

Download (PDF)

Tags: Computer science, FPGA, HLS, OpenCL, Optimization, Performance

November 8, 2020 by hgpu

Performance Assessment of OpenMP Compilers Targeting NVIDIA V100 GPUs

Joshua Hoke Davis, Christopher Daley, Swaroop Pophale, Thomas Huber, Sunita Chandrasekaran, Nicholas J. Wright

View

Download (PDF)

Tags: Computer science, CUDA, Heterogeneous systems, nVidia, OpenACC, OpenMP, Performance, Tesla V100

October 25, 2020 by hgpu

On the performance of a highly-scalable Computational Fluid Dynamics code on AMD, ARM and Intel processors

Pablo Ouro, Unai Lopez-Novoa, Martyn Guest

View

Download (PDF)

Tags: ARM, cfd, Fluid dynamics, HPC, Performance

October 18, 2020 by hgpu

torchPDLP: Primal-Dual Linear Programming in PyTorch. In collaboration with AMD and IPAM

Accelerating a Linear Programming Algorithm on AMD GPUs

Benchmarks for Dissecting CPU-GPU Unified Physical Memory on AMD MI300A APUs

Dissecting CPU-GPU Unified Physical Memory on AMD MI300A APUs

kvcached: Elastic KV cache for dynamic GPU sharing and efficient multi-LLM inference

Towards Efficient and Practical GPU Multitasking in the Era of LLM

Bandicoot: C++ library for GPU accelerated linear algebra

Bandicoot: A Templated C++ Library for GPU Linear Algebra

Luthier: Bridging Auto-Tuning and Vendor Libraries for Efficient Deep Learning Inference

Fused Kernel Library (FKL)

The Fused Kernel Library: A C++ API to Develop Highly-Efficient GPU Libraries

GPUHammer: Rowhammer Attacks on GPU Memories are Practical

Block: Balance Loader of LLM Serving with Context, Knowledge and Predictive Scheduling

Block: Balancing Load in LLM Serving with Context, Knowledge and Predictive Scheduling

SIGMo: Scalable Isomorphism Graph Matching on GPUs

SIGMo: High-Throughput Batched Subgraph Isomorphism on GPUs for Molecular Matching

DGEMM without FP64 Arithmetic - using FP64 Emulation and FP8 Tensor Cores with Ozaki Scheme

DGEMM without FP64 Arithmetic – using FP64 Emulation and FP8 Tensor Cores with Ozaki Scheme

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Why is FPGA-GPU Heterogeneity the Best Option for Embedded Deep Neural Networks?

Efficient MPI-based Communication for GPU-Accelerated Dask Applications

Interactive Parallelization of C Programs in SAPFOR

Directive-Based Data Partitioning and Pipelining and Auto-Tuning for High-Performance GPU Computing

Exploring FPGA Optimizations to Compute Sparse Numerical Linear Algebra Kernels

Toward Accurate Platform-Aware Performance Modeling for Deep Neural Networks

Evaluating the Performance and Portability of Contemporary SYCL Implementations

AZP: Automatic Specialization for Zero Values in Gaming Applications

Ginkgo – A Math Library designed for Platform Portability

Design and Performance Evaluation of Optimizations for OpenCL FPGA Kernels

Performance Assessment of OpenMP Compilers Targeting NVIDIA V100 GPUs

On the performance of a highly-scalable Computational Fluid Dynamics code on AMD, ARM and Intel processors

Recent source codes

torchPDLP: Primal-Dual Linear Programming in PyTorch. In collaboration with AMD and IPAM

Benchmarks for Dissecting CPU-GPU Unified Physical Memory on AMD MI300A APUs

kvcached: Elastic KV cache for dynamic GPU sharing and efficient multi-LLM inference

Bandicoot: C++ library for GPU accelerated linear algebra

Luthier: Bridging Auto-Tuning and Vendor Libraries for Efficient Deep Learning Inference

Fused Kernel Library (FKL)

GPUHammer: Rowhammer Attacks on GPU Memories are Practical

Block: Balance Loader of LLM Serving with Context, Knowledge and Predictive Scheduling

SIGMo: Scalable Isomorphism Graph Matching on GPUs

DGEMM without FP64 Arithmetic - using FP64 Emulation and FP8 Tensor Cores with Ozaki Scheme

Most viewed papers (last 30 days)