This episode is available to stream on-demand.

Over the last few years we’ve seen significant advances in AI capabilities, but such technologies require infrastructure that can cope with burgeoning workloads and adapt to future demand. Traditional data centers are unable to meet these requirements, so new facilities are being meticulously planned and deployed in strategic markets to meet new capacity and offer flexibility. In this episode Conapto will share their learnings of how to build for HPC, and examine the essential criteria required to make an operation ‘AI-ready’. In particular:

• How their new Stockholm facility features adaptable infrastructure capable of accommodate processors from the H100 to Blackwell
• The limitations of traditional facilities, and the necessity of future proofing designs
• Sustainable site selection criteria and the importance of TCO
• Evolving connectivity demands and network requirements