Mark Harris

hemi. Simple utilities to enable code reuse and portability between CUDA C/C++ and standard C/C++.

numba_examples. Examples using Numba.

mini-nbody. A simple gravitational N-body simulation in less than 100 lines of C code, with CUDA optimizations.

sublimetext-cuda-cpp. CUDA C++ package for Sublime Text 2 & 3

nsys_easy. Easier, quicker command-line CUDA profiling

ranger. Generate simple index ranges in C++ and CUDA C++

cpp11-range. Range-based for loops to iterate over a range of numbers or values

cuda_event_benchmark. Unit benchmarks of CUDA event APIs.

culeidoscope. Parallel extension of the Kaleidoscope toy language (from the LLVM project) on the CUDA platform.

stream-safety-first. Examples from Mark Harris's 2023 GTC presentation on Stream Safety with stream-ordered allocation.

numba. NumPy aware dynamic Python compiler using LLVM

raft. Rapids Analytics Framework Toolset to share building blocks between cuGraph and cuML

flaky-extensions-installer. Batch install vscode extensions with retry for flaky network connections

cuspatial. CUDA-accelerated GIS and spatiotemporal algorithms

cuml. cuML - RAPIDS Machine Learning Library

new-ubuntu-setup. My personal setup for new Ubuntu instances

rmm. RAPIDS Memory Manager

forage_maps. Source code for foragemaps.com website

tHogbomCleanHemi. Portable CUDA / OpenMP implementation of the Hogbom Clean Benchmark

libgdf. C GPU Dataframe Library

cudf. Python GPU DataFrame Library