#gpu-clusters

[ follow ]
#ai-infrastructure #supercomputing #microsoft #ai-data-centers #power-and-cooling-bottleneck #series-b-funding
Tech industry
fromComputerworld
5 days ago

Microsoft loses two senior AI infrastructure leaders as data center pressures mount

Microsoft can continue investing in AI data centers despite leadership departures that strain GPU-to-power ratios, with power availability now the primary bottleneck.
#ai-infrastructure
more#ai-infrastructure
fromTechzine Global
2 months ago

Clockwork launches FleetIQ to dramatically improve AI training

In large-scale AI training, billions in computing power are currently lost due to communication problems between processors. Clockwork tackles this problem with FleetIQ, a Software-Driven Fabric that provides real-time insight into GPU clusters. The system detects bottlenecks within microseconds and automatically redirects traffic via alternative routes. In addition, stateful fault tolerance prevents entire AI jobs from having to be restarted after a failure.
Artificial intelligence
Miscellaneous
fromTheregister
2 months ago

Europe's Jupiter supercomputer hits exascale threshold

Jupiter's Booster GPU cluster at Jülich surpassed exascale, delivering GPU-based computing for large-scale simulation and AI with about 6000 nodes.
[ Load more ]