Illuminaire

Episode 32: Your GPUs are waiting for the network

Podcast

Share this article

AI performance is not decided by GPUs alone because the network now determines how effectively thousands of accelerators can operate as one machine. Clayton Wagar, Leader of AI and High Performance Networking at Nokia, explains why congestion, resilience, optics and fabric design could matter more than adding another rack of compute.

Related Posts

Why the biggest challenge is not cooling AI but predicting tomorrow’s heat

The AI era has pushed data centre cooling ...

The AI infrastructure race is measuring the wrong thing

The AI industry has become fixated on counting ...

Powering AI means rethinking the grid not replacing the generator

The rapid expansion of AI infrastructure has exposed ...

The generation that is not waiting to be managed

A generation entering the workforce under economic pressure ...
Others have also viewed
Into The madverse podcast

Episode 32: Your GPUs are waiting for the network

AI performance is not decided by GPUs alone because the network now determines how effectively ...
Podcast

Why the biggest challenge is not cooling AI but predicting tomorrow’s heat

The AI era has pushed data centre cooling into the spotlight, but the real challenge ...
AI In Depth

AI factories will succeed or fail on infrastructure not algorithms

The rapid expansion of artificial intelligence has shifted attention away from software alone towards the ...
The Week in AI

The AI infrastructure race is measuring the wrong thing

The AI industry has become fixated on counting GPUs, announcing ever larger clusters and measuring ...
AI In Depth