Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19754908 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This is a self-contained artifact to build and run the benchmarks in the SC`26 submission `Andes: GPU-Native Tasking for Data-Dependent<br>Parallelism`.</p> <p>To run the tests, you will need cmake, CUDA > 12.9, and CCCL (https://github.com/nvidia/cccl).</p> <p>To build and run all tests, run ./trailhead/trailhead_run_tests.sh. This will download the datasets, run all tests 5 times, and report the median result. To change the number of times tests are executed, pass --`repeat N` to the .sh script.</p> <p>Results can be viewed and scripts can be rerun in an interactive dashboard using `./trailhead/trailhead_watch.sh`.</p>