trey-ornl Goto Github PK

always

MPI all-to-all microbenchmarks with sub-communicators.

amrex-tutorials

Tutorials for the AMReX framework.

cholla

A GPU-based hydro code

clang

Mirror kept for legacy. Moved to https://github.com/llvm/llvm-project

codesign-kernels

Climate kernels shared with interested parties for co-design

comb

Comb is a communication performance benchmarking tool.

cuda-samples

Samples for CUDA Developers which demonstrates features in CUDA Toolkit

e3sm

Energy Exascale Earth System Model source code.

faces

Microbenchmarks for testing programming strategies for nearest-neighbor communication with GPU-aware MPI.

fishfry

A benchmark that solves a 3D Poisson problem using MPI communication and GPU computation.

hipfort

Fortran interfaces for ROCm libraries

lsms

LSMS is a code for scalable first principles calculations of materials using multiple scattering theory.

mst2

olcf-user-docs

Sources for the Oak Ridge Leadership Computing Facility User Documentation

pairs

An MPI microbenchmark that measures the performance pairwise parallel ping pongs.

quicksilver

A proxy app for the Monte Carlo Transport Code, Mercury. LLNL-CODE-684037

quip

Header files that translate between Cuda and Hip.

rccl-tests

RCCL Performance Benchmark Tests

schmccl-tests

[NR]CCL tests implemented in MPI

scream

Exascale global atmosphere model written in C++ as part of the E3SM project

slate

SLATE is a distributed, GPU-accelerated, dense linear algebra library targetting current and upcoming high-performance computing (HPC) systems. It is developed as part of the U.S. Department of Energy Exascale Computing Project (ECP).

small_dgemms

Mini-app investigating performance of many small dgemm operations.

sumtimes

Microbenchmark for MPI_Allreduce with MPI_SUM using GPU memory.

tensile

Stretching GPU performance for GEMMs and tensor contractions.

timers

Timers for C++ with MPI and OpenMP.