Are shared resources keeping reserves for each Edge to ensure some are always available?

[KalleBuilds]

Hello, I'm currently investigating if a misused NSX Edge and its shared resources, can in any way negatively impact other Edges using those same resources. However ive had a hard time finding answers for these particular 3 resources, and I hope someone here maybe can enlighten me.

Do shared

1. Datastores
2. Clusters/Resource pools
3. Port Groups

keep reserved capacity to ensure that all Edges always are able to operate?

[Sreec]

Reserving the resources for edges are always good as they are first class candidates and we can avoid edge transport node getting impacted by any resource constraint . We can either go with 100% or 50% reservations for edge transport node.

1. Datastores - Do check for underlying lun and HCI storage policy protection types and IOPS limits. 
2. Clusters/Resource pools - Don't use resource pools unless there is a strong reason which can justify that design. Parent cluster resources should be used and do size the platform accordingly. 
3. Port Groups - Number of uplinks are the key factor. Based on the server model and overall network connectivity for different traffic types we should be good to know how to segregate the traffic types.