Nvidia Acquires SchedMD to Strengthen AI Software Capabilities

Nvidia has made headlines by acquiring SchedMD, an AI software company, on Monday. This strategic acquisition is a clear move to enhance its commitment to open-source technology while boosting investment in the artificial intelligence ecosystem. The acquisition positions Nvidia to better maintain its leadership, especially amid rising competition globally.

About SchedMD and Slurm

SchedMD is the leading developer of Slurm, an open-source workload management system specifically designed for High-Performance Computing (HPC) and AI. Slurm plays a crucial role in scheduling the extensive computing tasks that can consume a large portion of a data centre’s server capabilities. While the fundamental technology is available for free to developers and companies, SchedMD had previously offered engineering and support services.

Importance of Software in Nvidia’s Strategy

This deal illustrates the significance of software in Nvidia’s comprehensive strategy, reinforcing its proprietary CUDA platform, which most developers depend on. While the financial terms of the transaction were not disclosed, Nvidia is dedicated to ensuring that SchedMD’s software will remain open-source.

Role of Slurm in AI Development

Nvidia highlighted the essential function of Slurm in today’s AI landscape, stating in a blog update that Slurm, compatible with the latest Nvidia hardware, is fundamental infrastructure for generative AI. It is actively used by foundational model developers and AI innovators to effectively manage model training and inference tasks.

Support from SchedMD Leadership

The acquisition was positively received by the leadership team at SchedMD. Danny Auble, the company’s CEO, remarked that Nvidia’s extensive knowledge and investment in accelerated computing would significantly enhance the development of Slurm. He confirmed that Slurm would continue to be open-source, catering to the needs of the upcoming generation of AI and supercomputing.

Dual Announcements from Nvidia

This announcement coincided with Nvidia’s launch of a new series of open-source AI models, known as the Nemotron 3 lineup. Nvidia claims these models will prove to be faster, more cost-effective, and more intelligent than their predecessors. This dual announcement underscores the chipmaker’s resolve to fortify its ecosystem and counter a burgeoning wave of competing open-source models, especially those originating from Chinese AI labs. SchedMD has been in operation since 2010 in California, USA, and currently employs a workforce of 40 individuals.