Delta Electronics has unveiled an integrated power and thermal management infrastructure designed specifically for NVIDIA’s next-generation DSX AI factories.
These solutions are engineered to tackle the immense energy and heat challenges posed by large-scale AI model training and deployment. This collaboration between power systems specialist Delta and chipmaker NVIDIA aims to create a blueprint for rapidly deploying massive computing facilities, known as ‘AI factories’.
Delta reveals 800 new power and heat solutions
This move addresses a critical bottleneck in the expansion of artificial intelligence. Compute performance has grown exponentially, but the facility infrastructure required to power and cool these systems has struggled to keep pace. Power and infrastructure remain significant physical barriers to entry and scale for companies building AI capabilities.
At the heart of Delta’s solution is a philosophy of tight integration. This applies between the facility’s power systems and the IT hardware’s thermal management. This departs from traditional data centre construction, where these elements are often specified and installed separately, leading to inefficiencies.
Delta states that engineering power and cooling as a single, coordinated system is key to maximising AI output per megawatt. This unified stack includes everything from on-site energy storage systems and solid-state transformers for managing grid intake. It also encompasses medium-voltage equipment that distributes power throughout the facility.
The integrated stack extends into the data hall with row- and rack-level power distribution and liquid cooling. It finally delivers precisely conditioned power directly to the chips themselves. This holistic view is essential for handling the volatile power demands of GPU-heavy workloads.
Vladimir Troy, Vice President of AI Infrastructure at NVIDIA, underscored this necessity. He stated that “AI factories must be designed as complete systems, with power, cooling and compute working together.” Troy added, “Delta’s expertise in 800 VDC power delivery and liquid cooling will help customers deploy AI factories based on NVIDIA DSX faster and get more AI performance from every megawatt.”
The engineering behind the 800 VDC architecture
The standout technology in Delta’s announcement is its 800 VDC power architecture. Traditionally, data centres have used alternating current (AC) power. This power must undergo multiple conversions before reaching the DC components of a server.
Each conversion step results in energy loss, primarily as waste heat. This heat then must be removed by the cooling system, consuming even more power. By distributing power at a higher DC voltage, Delta’s architecture eliminates several wasteful conversion stages.
The company claims this approach achieves up to 98% power-conversion efficiency. This applies from the facility entrance to the computing hardware. At the board level, specialised DC-DC converters step the 800 volts down to the 50 VDC or 12 VDC required by the processors. These achieve a peak efficiency of up to 98.5%.
Fewer conversions mean less energy lost, less heat generated, and less copper cabling required. This efficiency is critical for managing operational costs and maximising the computational workload that can be supported within a given power envelope. The system is also designed for resilience. In-row systems provide up to 800 kW in a single power rack.
Integrated battery and capacitance backup technologies help stabilise the system against the rapid load changes characteristic of GPU clusters. This capability is a key part of turning raw data into industrial AI engineering intelligence.
Integrated liquid cooling for extreme density
Power delivery is only half the battle. The heat generated by tightly packed racks of high-performance GPUs can easily overwhelm conventional air-cooling systems. Delta’s solution pairs its power architecture with a comprehensive liquid cooling portfolio. This portfolio is designed to remove heat efficiently at every level of the AI factory.
Delta’s thermal-management portfolio includes 2.4 MW and 3 MW liquid-to-liquid cooling distribution units, in-rack cooling systems, and thermal solutions for AI racks. These systems are designed to coordinate cooling at the facility, row, rack, and component levels for AI infrastructure based on NVIDIA DSX.
In-rack cooling systems circulate liquid directly to cold plates mounted on the hottest components, like GPUs and CPUs. This removes heat at its source.
Prefabrication as a deployment accelerator
To address the urgent demand for new AI capacity, Delta is leveraging a modular construction strategy. Its Prefabricated AI Modular Data Center Solution integrates 800 VDC power systems and 3 MW of liquid-cooling capacity into self-contained, factory-built blocks. These modules are assembled and fully tested in a controlled environment before being shipped to the site.
This approach significantly reduces the complexity and duration of on-site construction. With much of the integration work completed off-site, commissioning the data centre is streamlined. This potentially cuts deployment time by up to 60%, according to Delta. Operators can then add capacity in predictable phases and begin generating value from their hardware sooner.
“AI factory economics begin with a simple question: how much intelligence can be produced from every available megawatt?” explained Austin Tseng, President of Delta Electronics Americas. Tseng added, “The answer depends on more than facility power alone. Delta and NVIDIA connect those layers — energy, power, cooling and components — allowing AI factory operators to reach time-to-first-token sooner and optimize token output at scale.”
Implications for industrial infrastructure
While designed for NVIDIA’s DSX platform, the principles behind Delta’s infrastructure have broader implications. The push toward high-voltage DC power distribution and integrated liquid cooling provides a potential roadmap for other energy-intensive industrial applications.
This includes high-performance computing clusters in research and engineering. It also extends to advanced manufacturing facilities, where energy efficiency and thermal management are increasingly critical.
The NVIDIA DSX Platform synergy
Delta’s infrastructure is specifically tailored for the NVIDIA DSX AI Factory Platform. This platform unifies reference designs, open software, accelerated computing systems, facilities infrastructure, and partner technologies. Its aim is to streamline AI factory design, deployment, and operation.
The DSX platform itself aims for the lowest token cost and accelerates time to first production across the full stack. This includes DSX MaxLPS software, designed for maximising token performance per megawatt.
Additionally, open-source, modular DSX OS software handles lifecycle management, runtime consistency, health automation, and resiliency. It also supports multi-tenant AI factory operations and platform services. This comprehensive software suite works in concert with Delta’s physical infrastructure.
