Ampace Showcases AI Continuity Solutions at DCW London 2026
Ampace Advancing AI Continuity Infrastructure at DCW London 2026
On March 5, 2026, Ampace made waves at the Data Centre World (DCW) in London, showcasing its cutting-edge AI Continuity Infrastructure designed to address the unprecedented volatility posed by AI workloads in data centers. As artificial intelligence continues to drive higher demands on power systems, especially with rapid GPU load variations and increased rack densities, ensuring a stable energy supply has become crucial for the operational success of AI-driven facilities.
Addressing the Challenge: AI Workloads
AI data centers are experiencing significant transformation. The rise of high-density racks has concentrated power demands into increasingly smaller areas, leading to intense pressure on existing power systems. These high-stakes environments require not just compact solutions, but systems capable of maintaining consistent power output even amid the unpredictability of GPU load fluctuations. To this end, Ampace introduced its innovative PU Series, specifically engineered to handle the substantial load variations characteristic of AI workloads.
The PU Series not only boasts a high-rate capability but also ensures stable cabinet-level performance, all while supporting operational continuity. This is particularly vital as downtime becomes virtually unacceptable in the world of AI, where continuous uptime is essential.
Thermal Resilience in the AI Space
Thermal challenges add an additional layer of complexity to data center operations. Elevated ambient temperatures coupled with sustained workloads escalate the demands on power systems, leading to potential risks. In response, Ampace has designed a solution that promotes module natural cooling, ensuring that performance remains consistent under high-stress conditions. As a result, operations can continue smoothly, reducing the risk of power interruptions.
The Benefits of Semi-Solid LFP Technology
A standout feature of Ampace's offerings is its semi-solid Lithium Iron Phosphate (LFP) cell technology. This innovative design is revolutionizing battery safety in the context of AI-driven data centers. By eliminating the free electrolyte within the cell structure, Ampace minimizes the risk of leakage across the product's lifespan. Testing has shown a dramatic decrease in thermal runaway gas release—an impressive 58% less compared to conventional designs in similar conditions. This denotes not just an engineering triumph but a strategic advancement towards ensuring safety in densely populated rack systems.
Significantly, these semi-solid cells have obtained UL 9540A certification, validating their resistance to fire propagation through rigorous large-scale testing protocols. Such a certification adds immense value in environments where operational reliability and safety are non-negotiable. Beyond enhancing safety, this architecture also promises longer service life and robust performance, crucial for the continuous high-load operations expected of AI data centers.
Ampace's Global Vision
At DCW London 2026, Ampace positioned its AI Continuity Infrastructure as a progressive framework designed for long-lasting application rather than a one-off product release. With stable dynamic response and thermal resilience, these systems form a solid foundation for powering advanced AI data centers. Ampace’s solutions are already aiding in the deployment of large-scale data centers across various regions, including projects in India, China, Korea, Japan, and Indonesia. Each of these installations underscores the adaptability and reliability of Ampace technology under real-world conditions characterized by high computational density and demanding thermal requirements.
As Ampace looks to the future, it remains committed to expanding its AI Continuity Infrastructure. By continually enhancing their power solutions, the company aims to meet the evolving demands of AI data centers worldwide—ensuring that businesses are equipped to handle the complexities of tomorrow's technological landscape.