NVIDIA Acquires SchedMD, Doubling Down on Open-Source AI Infrastructure

NVIDIA has acquired SchedMD, the company behind Slurm, the open-source workload management system widely used across high-performance computing (HPC) and AI environments. The move highlights how infrastructure software, often invisible to end users, is becoming central to the next phase of AI and supercomputing at scale.

Crucially, NVIDIA said it will continue to develop and distribute Slurm as open-source, vendor-neutral software, keeping it available across diverse hardware and software environments. The acquisition does not change Slurm’s licensing model, positioning the deal as an ecosystem play rather than a platform lock-in.

Why Slurm Matters to AI at Scale

As AI workloads grow in size and complexity, efficient scheduling and resource allocation have become a bottleneck. Training and inference jobs increasingly run across large clusters, requiring precise coordination of compute, memory, and networking resources.

Slurm has emerged as a de facto standard in this layer of the stack. It is used in more than half of the top 10 and top 100 systems on the TOP500 list of supercomputers, a signal of its role in managing large-scale parallel workloads. The scheduler is also deeply embedded in environments used by foundation model developers to orchestrate training and inference pipelines.

In practical terms, Slurm decides which job runs where, when, and for how long—making it critical infrastructure for both scientific computing and enterprise AI.

NVIDIA’s acquisition reflects a broader shift in strategy. While the company is best known for accelerated hardware, AI performance increasingly depends on software layers that sit above silicon. Scheduling, orchestration, and policy enforcement determine how efficiently expensive compute resources are used.

By bringing SchedMD in-house, NVIDIA gains closer alignment between its accelerated computing platform and the software that manages workloads across clusters. At the same time, the company has emphasised continued support for heterogeneous environments, allowing Slurm users to run mixed hardware clusters while benefiting from ongoing innovation.

NVIDIA said it will accelerate SchedMD’s access to new systems, enabling users to optimise workloads across their full compute infrastructure.

An Open-Source Commitment Under Scrutiny

Open-source communities tend to watch acquisitions closely, particularly when a large vendor acquires critical infrastructure software. NVIDIA has sought to address those concerns directly by committing to keep Slurm open-source and vendor-neutral.

Danny Auble, CEO, SchedMD, said: “We’re thrilled to join forces with NVIDIA, as this acquisition is the ultimate validation of Slurm’s critical role in the world’s most demanding HPC and AI environments. NVIDIA’s deep expertise and investment in accelerated computing will enhance the development of Slurm, which will continue to be open source to meet the demands of the next generation of AI and supercomputing.”

NVIDIA noted that it has collaborated with SchedMD for over a decade and will continue investing in Slurm’s development to ensure it remains the leading open-source scheduler for HPC and AI.

Impact On Enterprises and Researchers

SchedMD’s customer base spans cloud providers, AI companies, manufacturers, and research labs, supporting workloads across industries including healthcare, life sciences, autonomous driving, energy, financial services, manufacturing, and government.

For enterprises, the deal suggests greater integration between scheduling software and accelerated computing platforms—potentially improving utilisation, reliability, and scalability of AI infrastructure without forcing architectural changes.

NVIDIA said it will continue to provide open-source support, training, and development for Slurm, maintaining continuity for existing users while expanding capabilities to meet growing AI demands

Unlike splashy acquisitions focused on applications or models, NVIDIA’s move into workload management underscores where long-term competitive advantage may lie: in the control plane of AI infrastructure. As clusters grow larger and AI systems become more autonomous, orchestration software becomes as critical as the compute itself.

The acquisition positions NVIDIA not just as a supplier of AI hardware but as a steward of core infrastructure software that underpins modern supercomputing and enterprise AI.