NVIDIA has snapped up the company behind Slurm, an open-source cluster resource management platform for high-performance computing (HPC.)
It announced the buyout of Utah-headquartered SchedMD early on Wednesday – and is promising to keep the GPL-licensed Slurm open-source.
Among SchedMD's recent projects has been Slinky, a set of projects to run and manage Slurm clusters inside Kubernetes environments.
NVIDIA pledged to keep “supporting a diverse hardware and software ecosystem, so customers can run heterogeneous clusters with the latest Slurm innovations” – and to continue OSS support for SchedMD customers.
Slurm is used in more than half of the top 10 and top 100 systems in the TOP500 list of supercomputers, NVIDIA said, adding that it is “also part of the critical infrastructure needed for generative AI, used by foundation model developers and AI builders to manage model training and inference.”
(When you are running Big Computation on Big Clusters, efficient resource allocation, scheduling et al are important; that’s where Slurp comes in.)
Slurp rug pull fears
The acquisition sparked immediate concern among some Slurp users about NVIDIA exerting too much dominance over the HPC and AI software
Hamza Mian, an HPC expert and founder of the UK’s HMx Labs, commented on LinkedIn: “Slurp has a large and fairly technically adept userbase. I don’t doubt that a number of large users have their fingers hovering over the fork button already. I know we’ve seen a few OSS license rug pulls in recent history but I’m quietly confident that we won’t see that in this space…”
He added: “Slurm, and many others, lack basic functionality that I find hard to believe is still missing… such as the ability to schedule work based on both memory and processor utilisation, [or] checkpoint and consolidate workloads. I’m hoping that Nvidia sees this too and will provide that and I won’t be in the least bit surprised if that support is limited to Nvidia GPUs.
Mian added today: “My next guess would be that Nvidia acquires one of the many startups that are selling GPU workload checkpointing and migration solutions next to integrate with both runAI and Slurm…”
Andrei Huang, Head of HPC at US-based Biogen, commented: "NVIDIA’s move underscores that workload orchestration is now a strategic layer of the AI stack. This acquisition reinforces that hardware innovation alone isn’t enough! The software that governs how compute is allocated, shared, and optimized is equally critical.
He added: "From an HPC leadership perspective, this validates a direction many of us are already pursuing: treating HPC, AI, and hybrid cloud compute as integrated platforms, with common scheduling, policy, and governance models rather than siloed systems."
The terms of the acquisition were not disclosed.
The link has been copied!
