Direct-to-Chip & ImmersionNarrow moat

Vertiv's liquid-cooling leadership rests on offering the full range of approaches rather than betting on one. Direct-to-chip cooling brings coolant through cold plates mounted directly on the hottest components — the GPUs and CPUs — carrying their heat away at the source while the rest of the system is cooled conventionally; it is the approach gaining the fastest adoption in mainstream AI deployments. Immersion cooling goes further, submerging entire servers in a non-conductive fluid so that every component is bathed in liquid — a more radical, higher-density approach used in the most demanding cases. Underpinning both are the coolant-distribution units (CDUs) that pump, filter, and regulate the liquid loops, and the heat-rejection systems that ultimately move the heat outside.

Offering the whole toolkit matters because customers are still deciding which approaches suit their designs, and a partner who can supply, integrate, and service any of them — and pair the liquid loop with the rest of the power and cooling stack — is more valuable than a single-technique specialist. Vertiv's engineering depth across the range, and its ability to manufacture and deploy at scale, is the substance behind its market leadership. The caveat, taken up in the threat, is that these are engineered systems facing capable specialist competitors and the standardizing influence of the chipmakers' reference designs, so breadth and quality must be continually proven rather than assumed — proven lately in a backlog that more than doubled to ~$15B¹.