Every time someone uploads a photo, watches a streaming series, or runs an application in the cloud, there is a data center working behind it. But this digital heart of the modern world consumes enormous amounts of energy and water. Faced with this challenge, OVHcloud has presented its new Smart Datacenter architecture, a system that integrates artificial intelligence and a complete redesign of its infrastructure to achieve more efficient and sustainable operation.
An architecture that learns from the environment
After two years of development, the company has launched the fifth generation of its racks, designed to make the most of the capabilities of artificial intelligence. These new servers are grouped in series-connected clusters, optimizing heat flow and improving cooling performance. Furthermore, its type hydraulic configuration pull It allows you to precisely adjust the pressure and water flow required by each component.
The result is a significant reduction in energy consumption aimed at maintaining the optimal temperature of the servers. Direct-to-chip water blocks, designed in-house, eliminate the need for bulky systems and simplify maintenance operations while extending hardware life.
Smart Datacenter: a system that integrates artificial intelligence and a complete redesign of its infrastructure to achieve more efficient and sustainable operation
Smart sensors and real-time control
The new cooling module, known as Cooling Moduleis now 50% more compact and can manage multiple rows of racks at the same time. Its operation is based on more than thirty sensors that constantly monitor the pressure, speed and temperature of the water.
Thanks to this network of sensors and artificial intelligence algorithms, the system autonomously adapts to the load conditions and the data center environment. “We are paving the way for intelligent data centers that leverage our most advanced technologies, as well as artificial intelligence, to solve what was previously an impossible equation,” says Miroslaw Klaba, R&D Director at OVHcloud.
This approach allows cooling electrical consumption to be reduced by up to 50%, in addition to reducing the environmental noise generated by the equipment.
Reducing water consumption with predictive AI
Energy savings are not the only benefit. The new OVHcloud architecture also incorporates a predictive AI system that analyzes the data collected by the Smart Racks and refrigeration modules in real time. This information is stored in a datalake and is combined with local weather data to anticipate system needs.
In this way, artificial intelligence can precisely adjust the volume of water needed to cool the equipment, which has reduced water consumption by up to 30%. The new Smart Dry Coolerinstalled outdoors, takes up half the space of its predecessor and requires fewer fans, which not only decreases energy use but also the complexity of maintenance.
Adaptive intelligence for a sustainable cloud
The predictive system does not limit itself to observing, it also makes decisions. You can modify parameters such as the speed of the pumps or the opening of valves according to external conditions, prioritizing water saving, noise reduction or electricity consumption, depending on the situation.
This ability to adapt makes the Smart Datacenter a truly intelligent infrastructure. AI not only optimizes performance, but balances operational and environmental variables, setting a new standard in technological sustainability.
