In common deployment
models, Service Functions (SFs) are inserted into the data-forwarding path of
peers communicating with each other. However, with the introduction of service
chaining functionality, SFs are not required to be located on the direct data
path, rather the network traffic is routed through required SFs, wherever they
NSH Service Chaining
allows traffic flows to be classified so that only the desired flows are passed
to the service. Moreover, classification enables network traffic to be
dynamically moved to different service functions and service function paths
without the need for major configuration changes or topology rewiring.
Network Service Header
NSH is added to
network traffic, in the packet header, to create a dedicated service plane that
is independent of the underlying transport control protocol. In general, NSH
contains path identification information, which is needed to realize a service
path. In addition, NSH adds the metadata information about the packet, service
chain or both to an IP packet, depending on the header type configured.
Chaining feature can be configured either by using the Command Line Interface
(CLI), or by using Enterprise Policy Application (EPA). EPA is an application
that is hosted on controllers such as Application Policy Infrastructure
Controller Enterprise Module (APIC-EM). You can use EPA GUI to configure a
service chain based on services available in the network and apply a classifier
to that chain. This information is then pushed to the controller (APIC-EM) to
be applied to the network.