NVIDIA Corp. on Monday said it has agreed to acquire SchedMD, the software firm behind Slurm, a widely used open-source workload management system that has become essential infrastructure for high-performance computing and artificial intelligence (AI) operations.
The purchase, at an undisclosed price, represents a strategic move by the chip designer to reinforce its position in the AI ecosystem amid intensifying competition. While NVIDIA built its reputation on powerful processors, the company increasingly recognizes that software capabilities are crucial to maintaining market dominance, it said in a blog post.
SchedMD, founded in 2010 by software developers Morris Jette and Danny Auble, specializes in software that schedules large computing jobs capable of occupying substantial portions of a data center’s server capacity.
Slurm has emerged as a leader in workload management and job scheduling, powering more than half of the systems in both the top 10 and top 100 of the TOP500 supercomputer rankings. The software’s scalability, throughput, and sophisticated policy management have made it indispensable for organizations running complex computational tasks.
NVIDIA emphasized that it would continue distributing Slurm as open-source, vendor-neutral software, ensuring broad accessibility across various hardware and software environments. The technology plays a critical role in generative AI infrastructure, helping foundation model developers and AI builders manage model training and inference processes.
“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,” SchedMD CEO Danny Auble said in a statement. He noted that NVIDIA’s expertise in accelerated computing would drive Slurm’s development while maintaining its open-source nature.
The partnership between NVIDIA and SchedMD spans over a decade, and NVIDIA plans to build on this relationship by investing in Slurm’s ongoing development. The acquisition will accelerate SchedMD’s access to new systems, enabling users of NVIDIA’s accelerated computing platform to optimize workloads across their entire computational infrastructure.
SchedMD’s customer base includes cloud infrastructure provider CoreWeave, the Barcelona Supercomputing Center, and other organizations across sectors including autonomous vehicles, healthcare, energy, financial services, manufacturing, and government.
The announcement came hours after NVIDIA unveiled a new family of open-source AI models designed to be faster, cheaper, and more capable than previous offerings. This dual announcement underscores NVIDIA’s commitment to open-source technology as it confronts growing competition from rival AI labs, particularly those emerging from China.
NVIDIA’s proprietary CUDA software remains a standard among developers and a major selling point for its chips, making software strategy central to the company’s continued leadership in the AI industry.
