At the Open Compute Project (OCP) Global Summit this week, Microsoft made several new contributions to optimize infrastructure at all levels for breakthroughs in AI.
One focus is datacenter cooling to support higher power loads from AI systems. Microsoft presented designs for an advanced liquid cooling heat exchanger to be more easily deployed across the globe. The system is designed in a modular fashion for high efficiency in dissipating heat from ever-increasingly dense racks that process AI workloads.
Microsoft says power delivery is another area in need of innovation. It demonstrated how Microsoft and Meta are collaborating on a disaggregated 400 High Voltage Direct Current (VDC) rack design. The modular power architecture scales the racks up from hundreds of kilowatts to one megawatt for increased AI accelerators per rack, thereby assuring your infrastructure scale adapts to changing workload needs.
This is a new disaggregated rack design to address critical space and power constraints. The solution features a disaggregated 400 High Voltage Direct Current (VDC) unit that scales from hundreds of kW up to 1MW, enabling 15% to 35% more AI accelerators in each server rack. This modular approach allows for power adjustments in the disaggregated power rack to meet the changing demands of different inferencing and training SKUs.
Microsoft also previewed measures for shoring up AI against a new generation of quantum-attack machines. It is making the Adams Bridge quantum-resistant accelerator open source via the Caliptra silicon root of trust project.
The growing capabilities of quantum computers present challenges to hardware security, as classical asymmetric cryptographic algorithms used pervasively throughout hardware security can be easily defeated by a powerful enough quantum computer. In recognizing this risk, the National Institute of Standards and Technology (NIST) has published standards for the new quantum resilient algorithms.
Further, Microsoft co-founded the OCP-SAFE initiative to have consistent security auditing of hardware and firmware. This, in addition to projects like Caliptra, furthers supply chain transparency and integrity that are crucial for a trusted path toward hardware trust.