7. Known limitations
There currently exist the following limitations:
IPUs can be only accessed from within the IPU-POD network by default.
IPUJobPods must be run on a Kubernetes node that can access the IPUs, which means that at least one of the IPU-POD head nodes has to be a Kubernetes worker node.
IPUJobs(jobs with more than one
WorkerPod), you must specify the network interface which will be used for MPI communication using the mpirun
IPU partitions larger than 64 IPUs are currently not supported.