Compal and ZutaCore Unveil Waterless Cooling for AI Data Centers

Sep 10, 2025
Compal and ZutaCore have introduced a waterless, two-phase liquid cooling solution for AI data centers, showcased at Yotta 2025.

Compal Electronics and ZutaCore have introduced a groundbreaking waterless, two-phase liquid cooling solution for AI data centers, announced in a press release. This innovative technology, showcased at Yotta 2025, aims to enhance efficiency, reliability, and sustainability in handling next-generation AI workloads.

The collaboration features ZutaCore's HyperCool platform, which uses a non-conductive, non-corrosive dielectric fluid to instantly remove heat at the source, ensuring processors remain cool even in constrained environments. This approach eliminates the risks associated with traditional water-based systems, such as condensation and corrosion, while maximizing performance and reducing energy costs.

Compal's platforms, integrated with ZutaCore's cooling technology, include the SG720-2A-L2 and SD221-8A-L2 systems. These platforms are designed for high-density computing and AI acceleration, achieving impressive Power Usage Effectiveness (PUE) and partial PUE (pPUE) scores, which are industry benchmarks for energy efficiency. The systems support both current and future data center infrastructures, ensuring compatibility and sustainability.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.