Use Segfault voluntary users resources to build a massive computing cluster.
The main idea is to create what is called a Boewulf cluster using the resources given by voluntary users from the Segfault platform.
It would need a main node to manage all clients that would connect to it and form a massive computing cluster where ones could run heaving computing tasks shared across voluntary users nodes.
I'll try to describe here more in details how the main node should be implemented and software will be required to install.
On the main node, the following will be have to be installed:
The admin node will be used to control the main node.
The client nodes will connect to the main node and form the massive computing cluster. The more clients connected, the bigger the cluster get.
- https://en.wikipedia.org/wiki/Beowulf_cluster
- https://en.wikipedia.org/wiki/Slurm_Workload_Manager
- https://en.wikipedia.org/wiki/MPICH
- https://en.wikipedia.org/wiki/Open_MPI
- https://slurm.schedmd.com/
- https://slurm.schedmd.com/quickstart.html
- https://slurm.schedmd.com/quickstart_admin.html
- https://slurm.schedmd.com/tutorials.html
- https://slurm.schedmd.com/configless_slurm.html
- https://slurm.schedmd.com/network.html
- https://slurm.schedmd.com/containers.html
- https://slurm.schedmd.com/mpi_guide.html
- https://slurm.schedmd.com/configurator.html
- https://slurm.schedmd.com/configurator.easy.html
- https://slurm.schedmd.com/rest.html
- https://slurm.schedmd.com/rest_api.html
- https://slurm.schedmd.com/slurmrestd.html
- https://dun.github.io/munge/
- https://github.com/dun/munge/wiki/Installation-Guide