GPU Virtualization & Partitioning

GPU pooling

GPU pooling is the aggregation of physical GPUs across servers, racks, or sites into a single logical resource pool that workloads can draw from on demand.

Pooling is what separates a modern GPU cloud platform from a rack of static, siloed servers. It's the foundation that GPU orchestration and autoscaling are built on top of.

How hosted·ai approaches this

hosted·ai supports pooling GPUs across distributed sites, covered in How to monetize distributed GPU infrastructure across hundreds of sites.

Related terms