I've just set up my AWS account to train a model using SageMaker Studio. However, I'm running into a snag because it's asking for a quota increase on the A10G GPU instance (ml.g5.2xlarge). I submitted a request over a day ago, but I haven't heard anything back. Is this a normal wait time? I'm on a tight schedule and would appreciate any advice on how to handle this situation.
1 Answer
AWS is currently facing GPU shortages, and as a new account, you aren’t likely to be prioritized. You might want to check for availability in other regions, as that could help. Also, experimenting with different instance sizes might yield better results.
Which regions do you think are best for getting quicker access?