Details

  • NVIDIA has acquired SchedMD, the company behind Slurm, an open-source workload management system foundational to high performance computing (HPC) and AI. The announcement was made on December 15, 2025.
  • Slurm is a dominant scheduler in its category, powering more than half of the TOP500 supercomputers globally, including all of the top 10 and top 100 systems. It orchestrates resource allocation, job scheduling, and computational workflow for massive parallel workloads.
  • NVIDIA has pledged to continue advancing Slurm as open-source, vendor-neutral software, reinforcing its position across varied hardware and software landscapes and addressing community concerns around open collaboration.
  • Slurm's robust features include GPU management, distributed training support, and fault tolerance—making it critical infrastructure for training foundation models and running large-scale AI inference jobs.
  • With over a decade of collaboration between NVIDIA and SchedMD, NVIDIA plans to accelerate Slurm development by integrating early access to its latest hardware and systems. The scheduler is relied upon by industry, government labs, and cloud providers for a range of advanced technology applications.

Impact

NVIDIA's acquisition of SchedMD places it at the heart of AI and HPC infrastructure, enhancing its ability to shape how the largest models are trained and deployed. By promising to keep Slurm open and vendor-neutral, NVIDIA allays some competitive concerns while strengthening its ecosystem leadership. Competitors like Kubernetes may serve adjacent roles, but this move reinforces NVIDIA's dominance in specialized HPC and AI workload orchestration.