Articles and blogs

Meeting the growing demand for AI data centre capacity means overcoming the complex challenges of scaling compute, power, and cooling. Mattias Jansson, senior VP of Embedded & Critical Power at Flex, explores the power architectures needed to stay ahead.

Fueled by the relentless demand for cloud-based storage and computing, data centers are growing at an unprecedented pace.

Part two in a blog series taking a closer look at the real-world challenges that shape power module reliability and performance.

Scaling compute, power and cooling to expand data center capacity is a challenge

A blog about MTBF (mean time between failure)

Mikael Appelberg, CTO, and Andreas Larsson, Senior R&D engineer, from Flex Power Modules present a firsthand review of APEC 2025.

In this blog we explore strategies for optimizing power distribution in data centers, highlighting intermediate bus converters (IBCs) as a critical component for handling increasing power loads while maintaining system efficiency.

In power system designs, timing is critical. The need to deliver reliable and efficient solutions under tight deadlines has driven the demand for robust technical support services and tools.

Most high-performance DC/DC converters used in servers come with an element of digital monitoring and control, with functionality set in firmware. In this blog, we consider whether this firmware could be accessible to users for modification, and what the potential implications are.

Power conversion schemes for processors have often prioritized overall efficiency, but with AI pushing load dissipation higher in smaller spaces, other constraints come into play such as the relative height of different ratio intermediate bus converters.

With the enormous power levels seen in latest AI-targeted processors, liquid cooling has become a necessity.

Liquid cooling is increasingly vital in AI data centers as traditional methods struggle to keep manage intense heat.