high performance computing on graphics processing units: hgpu.org

hgpu.org » nVidia Jetson AGX Xavier

FC_ACCEL: Enabling Efficient, Low-Latency and Flexible Inference in DNN Fully Connected Layers, using Optimized Checkerboard Block matrix decomposition, fast scheduling, and a resource efficient 1D PE array with a custom HBM2 memory subsystem

Nick Iliev, Amit R Trivedi

View

Tags: CNN, Computer science, Deep learning, FPGA, Matrix decomposition, Neural networks, nVidia, nVidia Jetson AGX Xavier, Performance

February 13, 2022 by hgpu

Performance Analysis and Optimization Opportunities for NVIDIA Automotive GPUs

Hamid Tabani, Fabio Mazzocchetti, Pedro Benedicte, Jaume Abella, Francisco J. Cazorla

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, nVidia, nVidia Jetson AGX Xavier, nVidia Jetson TX2, Performance

April 25, 2021 by hgpu

Using Parallel Programming Models for Automotive Workloads on Heterogeneous Systems – a Case Study

Lukas Sommer, Florian Stock, Leonardo Solis-Vasquez, Andreas Koch

View

Download (PDF)

Tags: Automotive, Benchmarking, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia Jetson AGX Xavier, nVidia Jetson TX2, OpenCL, OpenMP

January 26, 2020 by hgpu

* * *

GPJSON: GPU-based JSON data processing system accessible via all GraalVM languages

GPU-based JSON data processing using structural indexes

NEPTUNE: Network- and GPU-aware management of serverless functions at the edge

NEPTUNE: Network- and GPU-aware Management of Serverless Functions at the Edge

HeCBench: a collection of Heterogeneous Computing benchmarks written with CUDA, HIP, SYCL (DPC++), and OpenMP-4.5

Experience of Migrating a Parallel Graph Coloring Program from CUDA to SYCL

cuPSO: GPU Parallelization for Particle Swarm Optimization Algorithms

WARPSPEED: An analytic Data Volume and Performance Estimator for GPU Kernels

Analytical Performance Estimation during Code Generation on Modern GPUs

Celerity HLA: A high-level API on top of Celerity to support functional, C++20 range-adaptors-like patterns

The Celerity High-level API: C++20 for Accelerator Clusters

GPU-Sched

CASE: A Compiler-Assisted SchEduling Framework for Multi-GPU Systems

Efficient Execution of OpenMP on GPUs

Explicit Caching HYB based SpMV

Explicit caching HYB: a new high-performance SpMV framework on GPGPU

matrix_inversion algorithm

Performance study on GPU offloading techniques using the Gauss matrix inverse algorithm

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

FC_ACCEL: Enabling Efficient, Low-Latency and Flexible Inference in DNN Fully Connected Layers, using Optimized Checkerboard Block matrix decomposition, fast scheduling, and a resource efficient 1D PE array with a custom HBM2 memory subsystem

Performance Analysis and Optimization Opportunities for NVIDIA Automotive GPUs

Using Parallel Programming Models for Automotive Workloads on Heterogeneous Systems – a Case Study

Recent source codes

GPJSON: GPU-based JSON data processing system accessible via all GraalVM languages

NEPTUNE: Network- and GPU-aware management of serverless functions at the edge

HeCBench: a collection of Heterogeneous Computing benchmarks written with CUDA, HIP, SYCL (DPC++), and OpenMP-4.5

cuPSO: GPU Parallelization for Particle Swarm Optimization Algorithms

WARPSPEED: An analytic Data Volume and Performance Estimator for GPU Kernels

Celerity HLA: A high-level API on top of Celerity to support functional, C++20 range-adaptors-like patterns

GPU-Sched

Efficient Execution of OpenMP on GPUs

Explicit Caching HYB based SpMV

matrix_inversion algorithm

Most viewed papers (last 30 days)