A customer requests that you determine how much of their remote branch traffic into a central data center is related to a call manager that resides in the data center. Which solution do you recommend?

NetFlow on branch routers would capture outbound branch traffic but would require aggregating data from every branch and cannot easily isolate call-manager-destined flows as efficiently as a single central collection point.

Enabling NetFlow on the central data center routers collects detailed flow records - including source/destination IP, protocol, and port - for all traffic entering the DC from branch sites. By filtering flows destined for the call manager's IP address and port, you can calculate exactly what percentage of branch traffic is call-manager related. This provides granular, application-level visibility without requiring changes at the branch.

SNMP polling provides interface-level byte and packet counters only; it cannot distinguish call-manager traffic from other application flows.

SNMP polling of branch routers shares the same limitation as option C - no per-application or per-flow granularity.

An ACL with logging on the call manager switch generates syslog entries per connection but does not provide byte-level traffic volume statistics needed to measure bandwidth consumption.

SPAN mirroring to a data analyzer requires dedicated hardware/software at the DC and captures all traffic, making it operationally heavier than NetFlow for this specific measurement task.