Explore projects

Memory manager with events that can handled delayed allocations and deallocations in multiple memory spaces.

Ongoing research training transformer language models at scale, including: BERT & GPT-2

Simple test of a HIP implementation's ability of kernels to accept an unused object reference.

Simple "Hello World" type program used to test the layout of resources on a Summit node using jsrun.

Simple "Hello World" type program used to test the layout of resources on a Summit node using jsrun.

Repository of GPU performance optimization

ELM Kernel Library

Strategies to distribute simplex-shaped workload across thousands of GPUs through mathematical mapping and dynamic scheduling

DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

Simple tester for multi-architecture domain decomposed particle transport

Mirror of https://github.com/celeritas-project/celeritas

Adaptive Sparse Grid Discretization solver.