RXMesh: A GPU Mesh Data Structure - SIGGRAPH 2021
C++ 111 10
A warp-oriented dynamic hash table for GPUs
Cuda 43 13
Code for paper "Design Principles for Sparse Matrix Multiplication on the GPU" accepted to Euro-Par 2018
C++ 40 11
Code for paper "Engineering a High-Performance GPU B-Tree" accepted to PPoPP 2019
Cuda 28 12
Multisplit is a primitive algorithm that categorizes input elements into contiguous buckets.
Cuda 9
BGHT: High-performance static GPU hash tables.
Cuda 16 2
GPU B-Tree with support for versioning (snapshots).
Multi-GPU dynamic scheduler using PGAS style cross-GPU communication
ML performance model for GPU training of DLRM and more.
Fast Block Sparse Matrices for Pytorch
Single-Source Shortest Path (SSSP) implementation in modern C++ for 2022 IPDPS workshop on Graphs, Architectures, Programming, and Learning (GrAPL 2022) submission.
Statistics on GPUs
Loading…