
Shame H100 not H200 - more VRAM. I guess you have to draw the line somewhere.
Italy's Leonardo supercomputer is to get an AI upgrade to beef up support for the development of Large Language Models (LLMs) and multi-modal generative AI, in addition to the 13 AI factories now being procured around the EU. Europe's supercomputing body the European High-Performance Computing Joint Undertaking (EuroHPC JU) …
Yeah, at $25,000 per unit, 166x8 H100 would run them $33M, and yield (theoretically) 44.5 PF/s at FP64 ... a tight fit (not to mention switches, racks, power aduction, labor, tariffs, currency fluctuations, ...).
It'll be cool if they make it though, raising Leonardo to an even 300 PF/s Rpeak (plus all that joyful IT4LIA AI factory hype nonsense ...).