This repository contains customized sst-core and sst-elements forked from SST Github page maintained by [SPEAR Lab](The lab moved from Illinois Institute of Technology to University of Illinois Chicago in August 2023. https://spear.lab.uic.edu/)
Visit sst-simulator.org to learn more about SST.
The current master branch has the implementation of Q-adatpive routing in sst-elements/merlin Dragonfly topology.
Note: if you use this SST implementation, please cite the following paper:
Yao Kang, Xin Wang, and Zhiling Lan. "Q-adaptive: A Multi-Agent Reinforcement Learning Based Routing on Dragonfly Network". In Proceedings of the 30th International Symposium on High-Performance Parallel and Distributed Computing (HPDC ’21).
Y. Kang, X. Wang and Z. Lan, "Study of Workload Interference with Intelligent Routing on Dragonfly," SC22: International Conference for High Performance Computing, Networking, Storage and Analysis, Dallas, TX, USA, 2022,
- Download sst-core and sst-elements from this repository, use other SST versions may cause compatible issue.
git clone --recurse-submodules https://github.com/SPEAR-UIC/q-adaptive_sst.git- Follow installation guide to install SST.
mpirun -np 4 sst q-adaptive_sst/sst-elements/tests/dragonfly_q-adaptive.py