Hosted by Tzag-Elita in collaboration with SchedMD, this webinar brought together HPC and AI professionals from Israel, Italy, Spain, the United States, Russia, and South Africa.
Led by Josep Oriol Vilarrubi, Senior Software Engineer at SchedMD, and Yuval Peer, Head of DevOps & HPC Infrastructure at Tzag-Elita, the session explored practical approaches to managing HPC and AI workloads using Slurm.
Key Topics Covered:
- Slurm fundamentals and workload scheduling
- Job queues, priorities, and fair-share policies
- Improving GPU and cluster resource utilization
- Real-world applications in HPC and AI environments
The session provided practical insights into optimizing compute resources and improving cluster efficiency across research and enterprise environments.