Enabling Multiple Tenants on PyTorch
Enabling Multiple Tenants on PyTorch¶
Since it is common to have multiple users sharing a Intel® Gaudi® AI accelerator server or cluster, Intel Gaudi supports multi-tenancy in the container runtime ecosystem for PyTorch. A single process workload can be run simultaneously on one or more Gaudi processors by multiple users.
Note
Running only one process per device is supported.
Partitioning and sharing one device across multiple users is not supported.