Alibaba Cloud introduced Aegaeon, a new GPU pooling system built with Peking University to transform AI workload management. The system scales models dynamically at the token level, running up to seven models on a single GPU. In testing, the team cut GPU usage from 1,192 to 213—an 82% drop. By tackling idle capacity from rarely used models, Aegaeon aims to make large-scale AI deployment far more efficient.
Fresh takes, great finds and engaging stories on the cloud and web hosting industry. Send us a news tip.
Enterprise Dedicated Servers - Intel/AMD EPYC & RYZEN - 100% Uptime 24/7 Support
Official Plesk Partner, Instant License Delivery, No Contract Commitment. Grab Your Savings NOW!
Significant discounts on KVM VPS SSD. Worldwide Locations. Full Root Access. Instant Deployment.
Get a .CA domain, with domain privacy, full DNS record control, domain forwarding, excellent support.
Premium professional WordPress sites that will not break your wallet. Optimized for SEO to drive traffic.
