Lenovo Tackles Cooling Conundrum in High-Density Data Centers

Lenovo’s Research Triangle Park (RTP) labs are working to address the cooling challenges of data centers, where energy demands have surged with the growth of artificial intelligence (AI). The company faces a complex problem: with increasing computing power comes more heat, and traditional air-cooling methods are becoming less efficient. To combat this, Lenovo is exploring alternative cooling solutions.

The challenge lies in managing heat generated by servers. A typical server uses 750 watts, but AI-compatible servers can consume up to 10 times more, producing significantly more heat. The traditional industry answer has been massive air-cooling units, which are energy-intensive and noisy.

However, alternatives have gained interest since the release of ChatGPT in November 2022, sparking a race among companies to develop more efficient cooling solutions. Companies like Lenovo and Iceotope are experimenting with liquid cooling methods, such as direct-to-chip and immersion cooling. These approaches can remove heat without relying on fans, making them more efficient.

The growth of AI has led to increased demand for denser servers, which require more effective cooling systems. The market is shifting towards liquid cooling solutions, with companies like Phononic and Iceotope leading the way. While some experts predict limited adoption in certain applications, others believe widespread adoption is likely in late 2026 and early 2027.

Lenovo’s flagship liquid cooling system, Neptune, pumps treated water through copper tubes at 113°F (45°C), a temperature that can carry significantly more heat away per volume than air. This method allows for more efficient heat dissipation, making it an attractive option for data centers.

As the AI industry continues to evolve, companies like Lenovo will play a crucial role in developing innovative cooling solutions. The company’s thermal engineers are working tirelessly to innovate and invent new tools to address this complex problem.

With the growth of AI expected to continue, the need for efficient cooling systems will become increasingly important. Companies like Lenovo are at the forefront of this innovation, pushing the boundaries of what is possible in data center design and operations.

Source: https://www.newsobserver.com/news/business/article300673529.html