About traffic types, network interfaces and traffic schemes

A traffic type is a particular classification of network traffic. It may include more than one protocol, but generally all traffic of a particular traffic type serves exactly one purpose, such as Diameter signaling or VM management.

A network interface is a virtual NIC (vNIC) on the VM. These are mapped to physical NICs on the host, normally one vNIC to one physical NIC, but sometimes many vNICs to one physical NIC.

A traffic scheme is an assignment of each of the traffic types that a VM uses to one of the VM’s network interfaces. For example:

  • First interface: Management

  • Second interface: Cluster

  • Third interface: Diameter signaling and Internal signaling

  • Fourth interface: SS7 signaling

Applicable traffic types

For custom VMs only the traffic types defined in the node parameters file are available. The following table lists the traffic types that can be present on custom VMs.

Traffic type Name in SDF Description

Management

management

Used by Administrators for managing the node.

Cluster

cluster

Used by Rhino and the OCSS7 SGC for inter-node communication.

Diameter signaling

diameter

Used for Diameter traffic.

SIP signaling

sip

Used for SIP traffic.

SS7 signaling

ss7

Used for SS7 traffic.

Internal signaling

internal

Used for signaling traffic between a site’s custom Rhino application nodes.

HTTP signaling

http

Used for all HTTP traffic except HTTP traffic between a site’s custom Rhino application nodes.

Primary signaling

custom_signaling

General-purpose signaling interface that can be used as required by the custom Rhino application. Includes Internal signaling traffic.

Secondary signaling

custom_signaling2

General-purpose signaling interface that can be used as required by the custom Rhino application.

Note

Please note the cluster traffic type is only used when Rhino is configured to be clustered on this VM.

Defining a traffic scheme

Traffic schemes are defined in the SDF. Specifically, within the vnfcs section of the SDF there is a VNFC entry for each node type, and each VNFC has a networks section. Within each network interface defined in the networks section of the VNFC, there is a list named traffic_types, where you list the traffic type(s) (use the Name in SDF from the table above) that are assigned to that network interface.

Note

Traffic type names use lowercase letters and underscores only.

Specify traffic types as a YAML list, not a comma-separated list. For example:

traffic_types:
  - diameter
  - sip
  - internal

When defining the traffic scheme in the SDF, for each node type (VNFC), be sure to include only the relevant traffic types for that VNFC. If an interface in your chosen traffic scheme has no traffic types applicable to a particular VNFC, then do not specify the corresponding network in that VNFC.

The possible traffic schemes for a given custom VM are constructed from the allowed traffic types (which are configured when building the VM images/CSARs), and the following rules for signaling traffic separation:

  • All signaling together

  • SS7 signaling separated

  • SIP and internal signaling on one interface, all others separated

  • Internal signaling separated

  • SIP signaling separated

  • HTTP signaling separated

  • All signaling separated

Note

All traffic schemes have a separate interface for the management traffic type, and a separate interface for the cluster traffic type if the custom VM supports clustering.

For example if the allowed traffic types are management, cluster, diameter, sip, internal and custom_signaling, then one possible traffic scheme (resulting from rule internal signaling separated) is:

First interface Second interface Third interface Fourth interface
management
cluster
diameter
sip
custom_signaling
internal

For the same allowed traffic types another possible traffic scheme (resulting from rule sip and internal signaling on one interface, all others separated) is:

First interface Second interface Third interface Fourth interface Fifth interface
management
cluster
sip
internal
diameter
custom_signaling
Important
  • Choose a single traffic scheme for the entire deployment. All VMs in a deployment must use the same traffic scheme (apart from differences caused by particular traffic types only being present on some VM types).

  • The various IP addresses for the network interfaces must each be on a separate subnet. In addition, each cluster of VMs must share a subnet for each applicable traffic type (e.g. all management addresses for the VMs must be on the same subnet).

    The recommended configuration is to use one subnet per network interface. If your deployment has multiple sites, use one subnet per network interface per site.

  • It is not possible to add or remove traffic types, or change the traffic scheme, once the VM has been created. To do so requires the VM to be destroyed and recreated.

Previous page Next page
VM Build Container Version 1.0.0