Slurm

How I glued together the per-node scratch directories of our SLURM cluster into a single, unified mount point — using mergerfs — and an interactive bash script to manage mounts across all nodes from one place.

Two recipes for moving VSCode (and PyCharm) off the Slurm login node and into actual compute jobs: a quick code-server web setup, and a per-job sshd workflow for full IDE remote-development. Stops resource-hungry editor processes from piling up on shared login nodes and lets you debug on the GPU you actually allocated.

Designed and deployed a SLURM cluster for the TUM Visual Computing Group, including LDAP-backed authentication, NFS storage, monitoring, and Ansible-driven node provisioning. Built a Vue.js + Flask dashboard so researchers can visualize cluster state, jobs, and resource usage, and released companion open-source tools (LDAP Triggers, LDAP Emails) used in day-to-day administration.