Key Takeaway
Co-designed by Dell, Intel, and the University of Cambridge, the Dawn system achieved a Power Usage Effectiveness (PUE) of 1.14 through advanced liquid cooling, demonstrating energy savings even under heavy workloads. This technology extends to Dell’s PowerEdge XE9680L servers, which use direct liquid cooling (DLC) to cut cooling energy consumption by 40-50%. Advanced hybrid cooling solutions, like the Dell PowerCool Enclosed Rear Door Heat Exchanger, recirculate IT heat, reducing cooling power by up to 60%. Future trends include AI-driven workload optimization and grid-aware technologies, which can significantly lower energy costs and emissions while ensuring operational efficiency.
Co-designed by Dell, Intel, and the University of Cambridge, Dawn achieved an impressive Power Usage Effectiveness (PUE) of 1.14.
This success stems from the integration of advanced liquid cooling directly into the system, demonstrating that substantial energy savings can be realized even under the most demanding workloads.
However, this innovation is not limited to large supercomputers; we have applied the same principles to our core product offerings as well.
For instance, our PowerEdge XE9680L servers utilize DLC technology in critical components, resulting in a 40-50% reduction in cooling energy consumption and an improved PUE.
How do advanced hybrid cooling solutions function to capture IT heat and minimize power usage? What are the tangible outcomes regarding sustainability and performance?
Advanced hybrid cooling solutions operate by intelligently merging direct-to-chip liquid cooling for high-priority components (CPUs/GPUs) with a novel air-cooling method that captures heat from other components.
The Dell PowerCool Enclosed Rear Door Heat Exchanger (eRDHx), for example, captures 100% of IT heat.
Rather than expelling hot air, the system employs fans to recirculate it through a heat exchanger within the enclosed rack, delivering cooler air back to the server inlets.
This enclosed loop enables the system to function with warmer water, reducing reliance on energy-intensive refrigeration chillers.
The practical outcome is a cooling power reduction of up to 60% compared to other liquid-cooled racks.
This represents a significant achievement, lowering energy consumption without sacrificing the high-density performance required for modern AI workloads.
What innovations or trends should organizations monitor that will enable data centres to become even more sustainable while addressing increasing computational demands?
AI-driven workload optimization will play a crucial role in data centre energy management, particularly when combined with grid-aware technologies that adapt to energy usage patterns.
Dell’s Concept Astro, for instance, illustrates how leveraging agentic AI, digital twins, and real-time grid data can help synchronize energy consumption with availability.
By predicting energy needs and pinpointing optimal times and locations for executing workloads, initiatives like Concept Astro demonstrate how businesses can significantly cut data centre energy costs and emissions while preserving operational efficiency.
80 Comments