Lori is actively engaging in addressing the escalating cooling requirements in the high-performance computing sector, particularly for chips and servers that power cutting-edge AI technologies. As energy consumption emerges as a major obstacle in enhancing computational capabilities, Lori has become one of many component manufacturers dedicated to developing efficient liquid cooling solutions tailored for AI data centers.

The focus in this domain lies on AI data centers relying on powerful graphics processors from companies like NVIDIA or similar vendors, capable of executing parallel computing tasks crucial for enabling applications like ChatGPT and similar generative AI. However, the power consumption demands of these chips are soaring, jumping from early 400 watts to 700 watts and potentially reaching 1000 watts in the future, far exceeding the energy consumption levels of ordinary consumer electronics.

Lori has invested heavily in liquid cooling systems aimed at resolving the immense heat generation issues in AI data centers. With the market's surging demand for graphics processing units (GPUs) used in AI model training, the power management and cooling solutions industry, previously seen as a peripheral segment of the supply chain, is now receiving unprecedented attention.

As chip power consumption increases drastically, cooling has become a critical factor limiting performance. Temperature directly impacts chip operating efficiency, necessitating more sophisticated cooling technologies than ever before. Traditional air-cooling methods based on fans and ducts are struggling to meet the cooling demands of AI servers.

Simultaneously, as the tech industry strives to achieve net-zero emission goals, cooling technologies are crucial for data center operators to maintain low power usage effectiveness (PUE). Emerging technologies include liquid cooling and immersion cooling, where the former involves a water system flowing around servers to reduce temperature, and the latter submerges the entire server rack in a non-conductive liquid.

Traditional cooling methods have failed to meet the challenges posed by AI. Comparative tests have shown that chips using air-cooling techniques can only reach 60% of their performance and face overheating risks, whereas liquid cooling solutions can continuously optimize computational performance.

Facing future AI computing demands, Lori believes that "new liquid cooling technologies will become crucial solutions, replacing the current mainstream but less efficient air-cooling methods." Market analysts also point out that the energy consumption of AI servers has surpassed the limits of air-cooling technologies, driving the demand for more complex and efficient cooling technologies, such as water or liquid cooling. With the continually growing market share of AI servers, this trend is expected to accelerate.

To address the cooling needs of AI servers, Lori has introduced a customized liquid cooling loop thermal management solution for servers.

Lori's liquid cooling loop employs advanced fluid dynamics principles, using precision piping systems to directly deliver coolant to the vicinity of each heat-generating component, achieving efficient and precise heat transfer. This loop not only enhances cooling efficiency but also extends hardware lifespan, featuring high reliability and ease of maintenance, providing robust technical support for AI data centers.