high performance computing on graphics processing units: hgpu.org

Applications

hgpu.org » Applications » Computer science

A Performance-Portable SYCL Implementation of CRK-HACC for Exascale

Esteban M. Rangel, S. John Pennycook, Adrian Pope, Nicholas Frontiere, Zhiqiang Ma, Varsha Madananth

View

Tags: AMD Radeon Instinct MI250X, ATI, Computer science, Cosmology, CUDA, HIP, HPC, nVidia, nVidia A100, Performance, performance portability, Physics, SYCL

October 29, 2023 by hgpu

GEVO-ML: Optimizing Machine Learning Code with Evolutionary Computation

Jhe-Yu Liou, Stephanie Forrest, Carole-Jean Wu

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Evolutionary Computations, Machine learning, Neural and Evolutionary Computing, nVidia, nVidia P100

October 29, 2023 by hgpu

Performance/power assessment of CNN packages on embedded automotive platforms

Paolo Burgio, Gianluca Brilli

View

Download (PDF)

Tags: CNN, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia Jetson AGX Xavier, nVidia Jetson Nano, nVidia Jetson TX2, Performance

October 22, 2023 by hgpu

Performance portability analysis of SYCL with a classical CG on CPU, GPU, and FPGA

Julian Franquinet

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, FPGA, nVidia, nVidia Quadro GP100, Optimization, Performance, performance portability, SYCL

October 22, 2023 by hgpu

Predicting the Execution Time of a kernel on a specific GPU using PTX code

Monika Dagar, Jorge Roldan

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Machine learning, nVidia, nVidia GeForce GTX 1650, nVidia GeForce GTX Titan XP, Performance, PTX, Tesla K20, Tesla K80, Tesla M60, Tesla P100, Tesla T4, Tesla V100

October 22, 2023 by hgpu

SYCL in the Edge: Performance Evaluation for Heterogeneous Acceleration

Youssef Faqir-Rhazoui, Carlos García

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, FPGA, Heterogeneous systems, nVidia, nVidia Jetson Nano, Optical flow, Package, SYCL

October 22, 2023 by hgpu

Reverse-Mode AD of Reduce-by-Index and Scan in Futhark

Lotte Maria Bruun, Ulrik Stuhr Larsen, Nikolaj Hinnerskov, Cosmin Oancea

View

Download (PDF)

Tags: Computer science, nVidia, nVidia A100, Programming Languages

October 15, 2023 by hgpu

Strega: An HTTP Server for FPGAs

Fabio Maschi, Gustavo Alonso

View

Download (PDF)

Source codes

Tags: Computer science, FPGA, OpenCL, Package

October 15, 2023 by hgpu

Comparison of different n-body algorithms on various hardware platforms using SYCL

Tim Thüring

View

Download (PDF)

Source codes

Tags: AMD Radeon Pro VII, Astrophysics, ATI, Computer science, N-body simulation, nVidia, nVidia A100, nVidia GeForce RTX 3090, nVidia Quadro GP100, Package, Physics, SYCL, Thesis

October 15, 2023 by hgpu

Open SYCL on heterogeneous GPU systems: A case of study

Rocío Carratalá-Sáez, Francisco J. andújar, Yuri Torres, Arturo Gonzalez-Escribano, Diego R. Llanos

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Heterogeneous systems, HIP, nVidia, Package, performance portability, SYCL

October 15, 2023 by hgpu

Exploring the Limits of Generic Code Execution on GPUs via Direct (OpenMP) Offload

S. Tian, B. Chapman, J. Doerfert

View

Download (PDF)

Tags: Compilers, Computer science, LLVM, OpenMP

October 8, 2023 by hgpu

Fortran performance optimisation and auto-parallelisation by leveraging MLIR-based domain specific abstractions in Flang

Nick Brown, Maurice Jamieson, Anton Lydike, Emilien Bauer, Tobias Grosser

View

Download (PDF)

Source codes

Tags: Computer science, Fortran, HPC, LLVM, MPI, nVidia, Package, Performance, Tesla V100

October 8, 2023 by hgpu

PTXBench: an open-source environment for kernel agents with architecture-specific CUDA/PTX

PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX

RealisticTritonBench: A Benchmark for Triton-Kernel Generation in Real-World AI Frameworks

TensorCorePTX: PTX Tensor Core GEMM Exploration

Hand-Written PTX Tensor-Core GEMM Kernels: A Multi-Precision Study on NVIDIA L4

FlashPDE: Triton-fused PDE stencil kernels for physics-informed neural networks

FlashPDE: A Drop-In Fused Triton Operator Library for Neural PDE Solvers

PortLBM: Portable GPU-accelerated Lattice Boltzmann Method Simulations

PortLBM: A Portable Lattice Boltzmann Tool Leveraging SYCL on AMD, NVIDIA, and Intel GPUs

Harness Engineering for LLM-Driven GPU Kernel Generation

CANNBot Skills

Compiler-Grounded Hierarchical Diagnosis for LLM-Based Triton Kernel Optimization

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Applications

A Performance-Portable SYCL Implementation of CRK-HACC for Exascale

GEVO-ML: Optimizing Machine Learning Code with Evolutionary Computation

Performance/power assessment of CNN packages on embedded automotive platforms

Performance portability analysis of SYCL with a classical CG on CPU, GPU, and FPGA

Predicting the Execution Time of a kernel on a specific GPU using PTX code

SYCL in the Edge: Performance Evaluation for Heterogeneous Acceleration

Reverse-Mode AD of Reduce-by-Index and Scan in Futhark

Strega: An HTTP Server for FPGAs

Comparison of different n-body algorithms on various hardware platforms using SYCL

Open SYCL on heterogeneous GPU systems: A case of study

Exploring the Limits of Generic Code Execution on GPUs via Direct (OpenMP) Offload

Fortran performance optimisation and auto-parallelisation by leveraging MLIR-based domain specific abstractions in Flang

Recent source codes

FP4 FlashAttention forward, backward, and TorchTitan reproduction code

EinsummableV3 - graph-based computation with tensor operations support

MaxKernel: Intelligent TPU Kernel Engineering Agents

PTXBench: an open-source environment for kernel agents with architecture-specific CUDA/PTX

RealisticTritonBench: A Benchmark for Triton-Kernel Generation in Real-World AI Frameworks

TensorCorePTX: PTX Tensor Core GEMM Exploration

FlashPDE: Triton-fused PDE stencil kernels for physics-informed neural networks

PortLBM: Portable GPU-accelerated Lattice Boltzmann Method Simulations

Harness Engineering for LLM-Driven GPU Kernel Generation

CANNBot Skills

Most viewed papers (last 30 days)