#python #cluster #slurm #ray #llm #parallelization > [!question] **What is Slurmray ?** > **SlurmRay** is a module designed to effortlessly distribute tasks on a [Slurm](https://slurm.schedmd.com/) cluster using the [Ray](https://ray.io/) library. Initially developed for the [Curnagl](https://wiki.unil.ch/ci/books/high-performance-computing-hpc/page/curnagl) cluster at the University of Lausanne, it now features **seamless integration with the Desi cluster** and advanced **queue management** capabilities. > > The core philosophy of SlurmRay is to allow users to **work locally while running highly complex functions on remote clusters**. It handles the distribution of computations—including **parallelization across multiple GPUs**—and automatically retrieves the results to the local machine once the job is complete. While optimized for its original environment, SlurmRay can be configured to run on any [Slurm](https://slurm.schedmd.com/) cluster with minimal setup. <a href="https://pypi.org/project/slurmray/" target="_blank" style="text-decoration: none; color: inherit;"> <div class="card"> <div class="card-image" style="background-image: url('https://pypi.org/static/images/twitter.abaf4b19.webp');"></div> <div class="card-content"> <div class="card-title"> <img src="https://pypi.org/static/images/favicon.35549fe8.ico" alt="favicon" class="card-favicon"> slurmray </div> <div class="card-description"> SlurmRay is a module for effortlessly distributing tasks on a Slurm cluster using the Ray library. </div> </div> </div> </a> <a href="https://github.com/hjamet/SLURM_RAY" target="_blank" style="text-decoration: none; color: inherit;"> <div class="card"> <div class="card-image" style="background-image: url('https://opengraph.githubassets.com/1/hjamet/SLURM_RAY');"></div> <div class="card-content"> <div class="card-title"> <img src="https://github.githubassets.com/favicons/favicon.svg" alt="favicon" class="card-favicon"> hjamet / SLURM_RAY </div> <div class="card-description"> Effortlessly distribute tasks on Slurm clusters using Ray. Supports seamless Desi cluster integration, remote execution from local machines, and GPU parallelization. </div> </div> </div> </a> <iframe src="https://html-preview.github.io/?url=https://github.com/hjamet/SLURM_RAY/blob/main/documentation/index.html#concept" width="100%" height="600px" style="border: 1px solid #e1e4e8; border-radius: 8px;" frameborder="0"> </iframe>