Dear authors of the I2NSF terminology draft: I2NSF framework has two different types of clients:
1. One type of client sending inquiry/request for Flow Security Policy to a controller (who controls a network). For example: an overlay network controller sending inquiry/request for Flow Security Policy to an underlay network controller. 2. Another type of client is the controller sending Flow Security Policy to individual NSFs. Simply put, the client described in 1) is making request to the (I2NSF) Controller (i.e. the controller’s North Bound Interface); and the “client” described in 2) is making request to the NSFs ( i.e. the controller’s South Bound Interface). It seems to me that the “I2NSF Client” described in the current I2NSF Terminology draft is performing the “controller” function, and the “I2NSF Agent” is residing in the NSF for receiving the flow security policies from the “Controller”. Is it better to have two distinct terminologies to differentiate the “Client to controller” from the “Client to NSFs”? Any suggestions? Linda
_______________________________________________ I2nsf mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2nsf
