DifferentiableUniverseInitiative / DifferentiableUniverseInitiative/flowpm

Benchmarking full N-body simulation on GPU clusters

Open
#80 24 comments 0 reactions 1 assignee Claimed by @dlanzieri View on GitHub
Mesh TensorFlow Performance
Dominant language
Jupyter Notebook
Stars
93
Forks
20
PR merge metrics
No merged PRs in 30d

Description

This issue is to track the work on benchmarking the new Horovod backend for GPU clusters and getting profiling information for FlowPM.

We want to do the following things:
- [x] : Run profiler on https://github.com/DifferentiableUniverseInitiative/flowpm/blob/mesh_update/scripts/mesh_nbody_benchmark.py and https://github.com/DifferentiableUniverseInitiative/flowpm/blob/mesh_update/scripts/pyramid_nbody.py
- [ ] : Compare the performance of both implementations as a function of the mesh size
- [x] : Identify the most costly parts of the computation

To learn how to do this profiling, keep an eye on https://github.com/DifferentiableUniverseInitiative/IDRIS-hackathon/issues/2

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.