About F5 Virtual Network Function Manager (VNFM)¶
Customer-facing documentation can be found here: F5 VNF Manager
F5 utilizes a third-party orchestration framework to bring you the F5 Virtual Network Function Manager (VNFM). This cloud orchestration tool uses OASIS TOSCA-compliant blueprints and plugins to manage the processing resources between your packet gateway and the Internet (Gi-LAN), in a private cloud environment (such as, OpenStack), auto-scaling your BIG-IP VE virtual machines, during high-volume periods. VNFM relies on BIG-IQ 6.0.1 and BIG-IP 13.1.1 images to provide services such as, scaling services and resources, load-balancing, and high availability (HA).
When finished deploying your VNFM, you will have the following system:
F5 VNFM resides in the MANO layer. Once you upload BIG-IQ, BIG-IP, and the VNFM images into OpenStack, and deploy the VNFM blueprint, the BIG-IQ license manager, BIG-IPs deployed by the blueprint, and your VNFM will comprise the management network.
Traffic will pass through the disaggregation (DAG) tier and the predefined-throughput service layers, connecting your subscriber packet gateway with the Internet (Gi LAN). The service layers will auto-scale out or add virtual machine instances, depending on your traffic demands. BIG-IPs in the DAG tier load-balance the subscriber traffic across BIG-IP VEs providing the virtualized network functions (VNF) inside the VNF service layer. This diagram depicts an F5 blueprint setting for the Gilan solution, auto_last_hop set to disabled (default), routing the return traffic through the DAG tier. However, this is a configurable setting in the F5 blueprint, which you can enable, and route return traffic directly back to the VNF service layer (bypassing the DAG tier).
Processes like, your policy engines, your subscriber service-charging functions, and signaling will occur on the control network.
What’s Next?