Salute & MCIM: Cutting Liquid Cooling Risk In AI DCs

Salute and MCIM have announced a partnership focused on improving operational intelligence for data centres deploying direct-to-chip liquid cooling.

The agreement brings MCIM into Salute’s ecosystem of technology partners supporting its Direct-to-Chip Liquid Cooling Operations service that targets the operational challenges created by high-density AI and high-performance computing (HPC) deployments.

The partnership centres on providing data centre operators with stronger visibility and control as rack densities increase and liquid cooling becomes more common.

MCIM’s operational intelligence platform integrates with Salute’s service model to support facilities that require precise execution, continuous monitoring and reduced operational risk.

Both Salute and MCIM already support Applied Digital’s AI and HPC facilities, bringing shared experience of operating liquid-cooled environments at scale.

Navigating high-density AI operations

Salute positions its Direct-to-Chip Liquid Cooling Operations service to address the complexity of running AI and HPC data centres.

Direct-to-chip cooling involves delivering liquid coolant directly to processors such as graphics processing units (GPUs) to remove heat more efficiently than traditional air-based systems.

While this approach supports higher compute densities, it also introduces new operational risks linked to fluid management monitoring and maintenance.

Through its partner ecosystem, Salute combines operational processes, training and staffing with specialist technologies.

The addition of MCIM strengthens this model by embedding a unified operational platform that captures telemetry workflows and asset data across the facility. Salute and MCIM state that this integration helps operators accelerate readiness while reducing the risks associated with liquid-cooled infrastructure.

An integrated model for operational intelligence

Under the partnership, Salute provides on-the-ground execution, including operational procedures training and access to specialist staff experienced in liquid-cooled facilities. MCIM delivers portfolio-level governance and real-time operational intelligence through a single platform.

According to MCIM, its platform is already deployed across more than 6.5GW of IT capacity worldwide and manages over 1 million critical data centre assets. It is used by three of the top four global colocation providers.

The combined model is designed to remove data silos and provide a consistent view of performance from individual racks through to the wider portfolio.

According to Salute and MCIM, this approach allows operators to deploy AI and HPC infrastructure faster while maintaining tighter control over operational and financial risk. The partnership is positioned as purpose-built for facilities where densities are high and tolerance for downtime is low.

John Shultz, Chief Product, AI and Learning Officer at Salute, says: “AI/HPC computing requires an entirely new approach to data centre operations because of the requirements of GPUs and the risks of DTC liquid cooling. Together, Salute and MCIM are able to help data centre operators protect their investments in AI by mitigating operational and financial risks.

“MCIM provides powerful visibility that delivers insights for optimising operations and eliminating risks. That is even more powerful when you combine it with Salute’s industry-leading operational model for DTC operations, our proven training capabilities and our ability to provide staffing with difficult-to-find expertise. Together we can help operators achieve and maintain operational excellence.”

Delivering mission-critical operations at scale

MCIM frames the partnership to address the speed and scale of AI infrastructure buildouts. The platform aggregates data across assets and sites, enabling consistent workflows and real-time insight for operators managing complex environments.

Combining operational data with trained teams allows issues to be identified and addressed earlier. This includes monitoring coolant flow, thermal performance and asset health across high-density rooms where failures can escalate quickly.

Mike Parks, CEO of MCIM, says: “AI is redefining what ‘mission critical’ means. The speed and scale of this buildout demand more than hardware and headcount. They demand a new operational model. Together, Salute and MCIM are delivering that model where every person, process and asset is connected, visible and performing to plan.”

This integrated approach of connecting people, processes and assets on a single platform could provide the foundation needed for managing the next generation of mission-critical AI facilities.​​​​​