https://github.com/jamesetsmith/mtxio
Matrix Market IO
https://github.com/jamesetsmith/mtxio
matrix-market openmp parallel sparse
Last synced: 22 days ago
JSON representation
Matrix Market IO
- Host: GitHub
- URL: https://github.com/jamesetsmith/mtxio
- Owner: jamesETsmith
- License: apache-2.0
- Created: 2023-03-29T18:57:45.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-06-30T00:02:02.000Z (almost 2 years ago)
- Last Synced: 2025-02-12T08:55:25.042Z (2 months ago)
- Topics: matrix-market, openmp, parallel, sparse
- Language: C
- Homepage:
- Size: 116 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Faster Parsing for Matrix Market
[](https://github.com/jamesETsmith/mmio/actions/workflows/build.yaml)
There are several great C++ projects ([PIGO](https://github.com/GT-TDAlab/PIGO) and [fast_matrix_market](https://github.com/alugowski/fast_matrix_market)) for reading matrix market files faster in serial and parallel.
However, GraphBLAS is written in ANSI C and I wanted to try and write a matrix market reader (and eventually writer) in C so it could be used side-by-side with GraphBLAS.
This is based heavily on the great C++ projects mentioned above and several of the utility functions are lifted almost verbatim from those projects.
:warning: This project is pre-alpha and is currently unstable. Currently only reading matrix market coordinate files with doubles is supported. Please understand there will be breaking changes.
## TODOs
Features required to meet specification:
- [ ] Object
- [X] matrix
- [ ] vector
- [ ] Format
- [X] coordinate
- [ ] array
- [ ] Fields
- [X] real
- [ ] complex
- [ ] integer
- [ ] pattern (i.e. no values)
- [ ] Symmetry
- [X] general
- [ ] symmetric
- [ ] skew-symmetric
- [ ] hermitian
- [ ] Feature: Add sanitizer checks to CI
- [ ] Feature: handle inf, nan, etc
- [ ] Feature: gracefully handle non-conformant/non-mtx files
- [ ] Feature: add parallel write
- [ ] Feature: Add proper CMake aliases (like `mtxio::mtxio`) to make it easier to integrate with other CMake projects
- [ ] Feature: Clean up CMake default flags (maybe use CMake configurations?)
- [ ] Performance: Improve parallel read performance, still slower than PIGO by 30-50%.
## Prerequisites
- C/C++ compiler
- Python3 and SciPy (Optional, needed to make benchmarking data)
## Build
```
[email protected]:jamesETsmith/mtxio.git
cd mtxio
export CPM_SOURCE_CACHE=$HOME/.cache/CPM # (optional)
cmake -B build
cmake --build build
ctest --test-dir build
```
## Benchmarking
```
cd data && python generate_data.py # May take a while
cd ..
perf stat -- env OMP_NUM_THREADS=12 ./build/bench/pigo_read data/n_15_0.3.mtx
perf stat -- env OMP_NUM_THREADS=12 ./build/bench/mtxio_read data/n_15_0.3.mtx
```