what is AWS Service Utilization Quota (Limits) ? Based on the definition provided in docs it says " AWS implements quotas to provide highly available and reliable service to all customers, and protect you from unintentional spend. Each quota starts with an AWS default value. " .
Basically my question here is what are resources in that definition ? Are they computers , nodes , cluster or some sort of distributed computation nodes or it is something else ?
Secondly , if they are different nodes or computation resources, they should be distributed, right ?
I really appreciate if someone can simplify this def for me ?
For instance what happens in a Sage Maker service with quota of 20 , does it mean in the training batches can be distributed over 20 nodes for fwd and back prop training ?