high performance computing on graphics processing units: hgpu.org

hgpu.org » Performance

Performance Modelling of Deep Learning on Intel Many Integrated Core Architectures

Andre Viebke, Sabri Pllana, Suejb Memeti, Joanna Kolodziej

View

Download (PDF)

Tags: Computer science, Deep learning, Intel Xeon Phi, Neural networks, OpenMP, Performance

June 12, 2019 by hgpu

Diagnosing Performance Bottlenecks in HPC Applications

Kenneth Czechowski

View

Download (PDF)

Source codes

Tags: Computer science, HPC, Package, Performance, Thesis

June 9, 2019 by hgpu

LAMDA: Learning-Assisted Multi-Stage Autotuning for FPGA Design Closure

Ecenur Ustun, Shaojie Xiang, Jinny Gui, Cunxi Yu, Zhiru Zhang

View

Download (PDF)

Tags: Computer science, FPGA, OpenCL, Performance

June 9, 2019 by hgpu

The Impact of GPU DVFS on the Energy and Performance of Deep Learning: an Empirical Study

Zhenheng Tang, Yuxin Wang, Qiang Wang, Xiaowen Chu

View

Download (PDF)

Tags: Computer science, Computer vision, CUDA, Deep learning, Neural networks, nVidia, nVidia GeForce GTX 2080 Ti, Performance, Tesla P100, Tesla V100

May 30, 2019 by hgpu

Massively Parallel GPU Memory Compaction

Matthias Springer, Hidehiko Masuhara

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, nVidia, nVidia GeForce GTX Titan XP, Performance

May 30, 2019 by hgpu

Automatic generation of warp-level primitives and atomic instructions for fast and portable parallel reduction on GPUs

Simon Garcia De Gonzalo, Sitao Huang, Juan Gomez-Luna, Simon Hammond, Onur Mutlu, Wen-mei Hwu

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, nVidia, nVidia GeForce GTX 980, Performance, performance portability, Tesla K40, Tesla P100

May 26, 2019 by hgpu

A Hybrid Framework for Fast and Accurate GPU Performance Estimation through Source-Level Analysis and Trace-Based Simulation

Xiebing Wang, Kai Huang, Alois Knoll, Xuehai Qian

View

Download (PDF)

Tags: Computer science, nVidia, nVidia GeForce GTX 645, nVidia GeForce GTX 940 M, nVidia Quadro K600, nVidia Quadro K620, OpenCL, Performance

May 26, 2019 by hgpu

Performance Analysis of Deep Learning Workloads on Leading-edge Systems

Yihui Ren, Shinjae Yoo, Adolfy Hoisie

View

Download (PDF)

Tags: Cloud, Computer science, Computer vision, CUDA, Deep learning, Machine learning, NLP, nVidia, nVidia DGX-2, nVidia GeForce RTX 2080 Ti, Performance, Tesla V100

May 23, 2019 by hgpu

Instructions’ Latencies Characterization for NVIDIA GPGPUs

Yehia Arafa, Abdel-Hameed Badawy, Gopinath Chennupati, Nandakishore Santhi, Stephan Eidenbenz

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, nVidia, nVidia GeForce GTX Titan X, nVidia Titan RTX, Performance, PTX, Tesla K40, Tesla P100, Tesla V100

May 23, 2019 by hgpu

Neural Query Language: A Knowledge Base Query Language for Tensorflow

William W. Cohen, Matthew Siegler, Alex Hofer

View

Download (PDF)

Source codes

Tags: Computer science, Databases, Deep learning, Package, Performance

May 19, 2019 by hgpu

OpenDNN: An Open-source, cuDNN-like Deep Learning Primitive Library

Daeyeon Kim

View

Download (PDF)

Tags: Computer science, Deep learning, FPGA, Neural networks, OpenCL, Performance, Thesis

May 19, 2019 by hgpu

Performance Engineering for a Tall & Skinny Matrix Multiplication Kernel on GPUs

Dominik Ernst, Georg Hager, Jonas Thies, Gerhard Wellein

View

Download (PDF)

Source codes

Tags: Benchmarking, Code generation, Computer science, CUBLAS, CUDA, Matrix multiplication, nVidia, Package, Performance, Tesla V100

May 12, 2019 by hgpu

MSCCL++: A GPU-driven communication stack for scalable AI applications

MSCCL++: Rethinking GPU Communication Abstractions for Cutting-edge AI Applications

Benchmark compute shader of Unity against InteropUnityCUDA

InteropUnityCUDA: A Tool for Interoperability Between Unity and CUDA

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

* * *

high performance computing on graphics processing units: hgpu.org

Performance Modelling of Deep Learning on Intel Many Integrated Core Architectures

Diagnosing Performance Bottlenecks in HPC Applications

LAMDA: Learning-Assisted Multi-Stage Autotuning for FPGA Design Closure

The Impact of GPU DVFS on the Energy and Performance of Deep Learning: an Empirical Study

Massively Parallel GPU Memory Compaction

Automatic generation of warp-level primitives and atomic instructions for fast and portable parallel reduction on GPUs

A Hybrid Framework for Fast and Accurate GPU Performance Estimation through Source-Level Analysis and Trace-Based Simulation

Instructions’ Latencies Characterization for NVIDIA GPGPUs

Neural Query Language: A Knowledge Base Query Language for Tensorflow

OpenDNN: An Open-source, cuDNN-like Deep Learning Primitive Library

Performance Engineering for a Tall & Skinny Matrix Multiplication Kernel on GPUs

Recent source codes

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

Shamrock: Multi-GPU hydrodynamics for astrophysics

LLMPerf: GPU Performance Modeling meets Large Language Models

Most viewed papers (last 30 days)