high performance computing on graphics processing units: hgpu.org

hgpu.org » Performance

Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance

Jiarong Xing, Leyuan Wang, Shang Zhang, Jack Chen, Ang Chen, Yibo Zhu

View

Download (PDF)

Tags: Auto-Tuning, Computer science, CUDA, Deep learning, Neural networks, nVidia, Performance, Tesla T4

October 31, 2021 by hgpu

Monitoring Collective Communication Among GPUs

Muhammet Abdullah Soyturk, Palwisha Akhtar, Erhan Tezcan, Didem Unat

View

Download (PDF)

Source codes

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

October 24, 2021 by hgpu

Principal Kernel Analysis: A Tractable Methodology to Simulate Scaled GPU Workloads

Cesar A. Baddouh, Mahmoud Khairy, Roland Green, Mathias Payer, Timothy G. Rogers

View

Download (PDF)

Tags: Computer science, CUDA, nVidia, Performance, Tesla V100

October 24, 2021 by hgpu

CompilerGym: Robust, Performant Compiler Optimization Environments for AI Research

Chris Cummins, Bram Wasti, Jiadong Guo, Brandon Cui, Jason Ansel, Sahir Gomez, Somya Jain, Jia Liu, Olivier Teytaud, Benoit Steiner, Yuandong Tian, Hugh Leather

View

Download (PDF)

Source codes

Tags: Artificial intelligence, Benchmarking, Compilers, Computer science, CUDA, Machine learning, nVidia, Package, Performance

September 26, 2021 by hgpu

Measurement and Analysis of GPU-accelerated Applications with HPCToolkit

Keren Zhou, Laksono Adhianto, Jonathon Anderson, Aaron Cherian, Dejan Grubisic, Mark Krentel, Yumeng Liu, Xiaozhu Meng, John Mellor-Crummey

View

Download (PDF)

Source codes

Tags: AMD Radeon Instinct Mi50, ATI, Code generation, Computer science, HPC, nVidia, OpenCL, Performance, Tesla V100

September 19, 2021 by hgpu

perf4sight: A toolflow to model CNN training performance on Edge GPUs

Aditya Rajagopal, Christos-Savvas Bouganis

View

Download (PDF)

Source codes

Tags: AI, Computer science, CUDA, Neural networks, nVidia, nVidia Jetson TX2, Package, Performance, Tesla P40

August 22, 2021 by hgpu

Performance assessment of CUDA and OpenACC in large scale combustion simulations

Guillermo Oyarzun, Daniel Mira, Guillaume Houzeaux

View

Download (PDF)

Tags: CUDA, nVidia, OpenACC, Performance, Physics, Tesla V100

August 8, 2021 by hgpu

On Efficient GPGPU Computing for Integrated Heterogeneous CPU-GPU Microprocessors

Daniel Gerzhoy

View

Download (PDF)

Tags: Computer science, Heterogeneous systems, OpenCL, Performance, Thesis

August 8, 2021 by hgpu

StreamBlocks: A compiler for heterogeneous dataflow computing

Endri Bezati, Mahyar Emami, Jörn Janneck, James Larus

View

Download (PDF)

Source codes

Tags: Computer science, FPGA, Heterogeneous systems, HLS, OpenCL, Package, Performance

July 25, 2021 by hgpu

Optimisation and GPU code generation of Stencils for Futhark

Christian Charlie Virt, Jonathan Wraa-Hansen

View

Download (PDF)

Source codes

Tags: Code generation, Computer science, CUDA, nVidia, nVidia GeForce GTX 780, nVidia GeForce RTX 2080 Ti, OpenCL, Performance, Thesis

July 18, 2021 by hgpu

Opening the Black Box: Performance Estimation during Code Generation for GPUs

Dominik Ernst, Georg Hager, Markus Holzer, Matthias Knorr, Gerhard Wellein

View

Download (PDF)

Source codes

Tags: Code generation, Computer science, CUDA, Lattice Boltzmann model, Machine learning, nVidia, Performance, Stencil computation, Tesla V100

July 11, 2021 by hgpu

kEDM: A Performance-portable Implementation of Empirical Dynamic Modeling using Kokkos

Keichi Takahashi, Wassapon Watanakeesuntorn, Kohei Ichikawa, Joseph Park, Ryousei Takano, Jason Haga, George Sugihara, Gerald M. Pao

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, HPC, nVidia, OpenCL, Package, Performance, performance portability

May 30, 2021 by hgpu

CFAL-bench

Comparing Parallel Functional Array Languages: Programming and Performance

Efficient Graph Embedding at Scale: Optimizing CPU-GPU-SSD Integration

Can Large Language Models Predict Parallel Code Performance?

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

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance

Monitoring Collective Communication Among GPUs

Principal Kernel Analysis: A Tractable Methodology to Simulate Scaled GPU Workloads

CompilerGym: Robust, Performant Compiler Optimization Environments for AI Research

Measurement and Analysis of GPU-accelerated Applications with HPCToolkit

perf4sight: A toolflow to model CNN training performance on Edge GPUs

Performance assessment of CUDA and OpenACC in large scale combustion simulations

On Efficient GPGPU Computing for Integrated Heterogeneous CPU-GPU Microprocessors

StreamBlocks: A compiler for heterogeneous dataflow computing

Optimisation and GPU code generation of Stencils for Futhark

Opening the Black Box: Performance Estimation during Code Generation for GPUs

kEDM: A Performance-portable Implementation of Empirical Dynamic Modeling using Kokkos

Recent source codes

CFAL-bench

Efficient Graph Embedding at Scale: Optimizing CPU-GPU-SSD Integration

Can Large Language Models Predict Parallel Code Performance?

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

Most viewed papers (last 30 days)