Current Status

The AWS EC2 instance types Trainium and Inferentia (inf, trn) can be added to a Red Hat OpenShift Service on AWS (ROSA) cluster machine-pool, and with the help of the AWS Neuron Operator for OpenShift, the accelerator drivers are installed on worker nodes.

Support Status

Given the tested and validated software suite for the Trainium and Inferentia instance types on ROSA, AWS will support the use of these instances in a ROSA cluster. Customers can then expect to use these instances types and their accelerated features while using the AWS Neuron Operator for OpenShift, typically in combination with Red Hat OpenShift AI

For more information, see
- this AWS article.
- this Red Hat article.
- the operator repository

Appendix

Instance types list

The specific instance types made available are these:

trn1.2xlarge
trn1.32xlarge
trn1n.32xlarge

trn2.48xlarge
trn2u.48xlarge

inf1.xlarge
inf1.2xlarge
inf1.6xlarge
inf1.24xlarge

inf2.xlarge
inf2.8xlarge
inf2.24xlarge
inf2.48xlarge