Panduit Launches Direct-to-Chip Cooling System for AI and HPC Infrastructure
Panduit, one of the world’s leading manufacturers of electrical, network, and infrastructure connectivity solutions, has launched its new Direct-to-Chip (DTC) Cooling System, designed for environments with a high concentration of computing resources, as well as AI and High Performance Computing (HPC) applications. The advanced liquid-cooling solution enables direct heat removal from processors, while supporting faster deployment, reduced installation risk, and scalable thermal management within a standard server cabinet footprint.
As AI workloads continue to drive higher compute demands, data centers are facing increasing challenges in managing heat efficiently. Panduit’s DTC system addresses these challenges by delivering a future-ready solution that aligns with sustainability objectives while improving overall operational efficiency.
“Our DTC Cooling System reflects Panduit’s commitment to solving real-world infrastructure challenges through smart, scalable solutions. This launch represents an important step toward enabling AI-ready infrastructure that is efficient, reliable, and future-proof,” said David Weksel, Director of Strategic Growth at Panduit Ventures.
The system is engineered for high-density computing environments and incorporates FlexFusion cabinets and rack manifolds, which work together to deliver uniform and reliable liquid cooling directly to the chip. Its design supports fast and secure installation without the need for cabinet extensions or additional retrofitting, allowing data center operators to deploy high-density AI servers without compromising space, deployment speed, or reliability.
The DTC Cooling System forms part of Panduit’s broader liquid-cooling portfolio and complements its existing Rear Door Heat Exchanger (RDHx) solutions. Together, these offerings provide a comprehensive cooling strategy for modern data centers, supporting scalable infrastructure across AI, HPC, and edge computing environments.






















