LANL Moves Seven Production HPC Clusters to OpenCHAMI

Los Alamos National Laboratory (LANL) has transitioned seven production clusters—comprising more than 11,000 compute nodes—to the open-source OpenCHAMI management stack. This is a large operational deployment rather than a disclosed contract award or solicitation. LANL reports two practical outcomes: less downtime during system upgrades and support for mixed-vendor environments. Those results make the transition relevant to federal technology contractors that design, integrate, operate, or secure high-performance computing (HPC) infrastructure, while providing no basis to infer a current procurement or award value.

The shift reflects a management approach suited to complex computing environments where hardware may come from multiple vendors and systems must remain available through maintenance and upgrades. Open-source tooling can offer agencies flexibility, but production use also places weight on integration, configuration management, operational support, security controls, and dependable service across heterogeneous equipment. LANL’s reported experience offers a concrete operational reference for contractors assessing how their capabilities fit federal HPC environments. It does not, by itself, establish that LANL is seeking vendors or that OpenCHAMI adoption will lead to a specific acquisition.

The news also sits within a broader federal investment landscape for large-scale computing. The Doudna system at the National Energy Research Scientific Computing Center (NERSC) is a related example: Dell was selected in May 2025. At LANL, the planned Mission and Vision supercomputers are expected to be deployed in 2027 and 2028, respectively, with HPE and NVIDIA identified in connection with those systems. These projects indicate continuing demand for advanced computing infrastructure, but the source does not provide contract values, detailed scopes, or procurement timelines for the planned systems.

For industry, the most actionable implication is capability positioning rather than near-term bid activity. Contractors can evaluate whether their HPC offerings work with open-source management stacks, support mixed-vendor clusters, and reduce service disruption during upgrades. They can also connect those capabilities to operational outcomes—such as availability, maintainability, and secure administration—when preparing for future federal requirements. The reporting identifies LANL, the U.S. Department of Energy (DOE), the National Nuclear Security Administration (NNSA), the DOE Office of Science, and NERSC as relevant organizations, but it does not specify which office would sponsor or acquire any future OpenCHAMI-related work.

• LANL moved seven production clusters to OpenCHAMI, covering more than 11,000 compute nodes.
• LANL reports less downtime during system upgrades after the transition.
• OpenCHAMI supports mixed-vendor environments, a relevant capability for heterogeneous federal HPC estates.
• No active solicitation, identified procurement, or contract value is disclosed; treat the story as an operational demand signal, not a bid notice.
• Contractors can assess integration, operations, and security offerings for compatibility with open-source HPC management stacks.
• Dell was selected in May 2025 for the Doudna system at NERSC.
• LANL’s planned Mission and Vision supercomputers have deployments expected in 2027 and 2028, respectively.
• HPE and NVIDIA are identified in connection with the planned LANL systems; the source provides no additional contract details.