As a Reinforcement Learning (RL) researcher I enjoy using Ray for various projects. However, Ray had limited community support for the Slurm architecture (which my university uses). So I built my own container orchestration solution for Ray on Slurm.
I recently published a paper in IEEE describing a method to perform container orchestration of Ray on Slurm. Benchmarks of RLLib and Mujoco environments are provided here. The official website is found here.
I would like to get the website and paper linked to the Ray official documentation for Slurm. I think it will be helpful to other researchers who want to utilize Ray on Slurm.
Description
As a Reinforcement Learning (RL) researcher I enjoy using Ray for various projects. However, Ray had limited community support for the Slurm architecture (which my university uses). So I built my own container orchestration solution for Ray on Slurm.
I recently published a paper in IEEE describing a method to perform container orchestration of Ray on Slurm. Benchmarks of RLLib and Mujoco environments are provided here. The official website is found here.
I would like to get the website and paper linked to the Ray official documentation for Slurm. I think it will be helpful to other researchers who want to utilize Ray on Slurm.
Link
No response