BIG-IP Virtual Edition (VE) is a version of the BIG-IP system that runs as a virtual machine in specifically-supported hypervisors. BIG-IP VE virtualizes a hardware-based BIG-IP system running a VE-compatible version of BIG-IP software.
Each time there is a new release of BIG-IP Virtual Edition (VE) software, it includes support for additional hypervisor management products. The Virtual Edition and Supported Hypervisors Matrix on the AskF5 website, http://support.f5.com, details which hypervisors are supported for each release.
The KVM virtual machine guest environment for the BIG-IP Virtual Edition (VE), at minimum, must include:
For production licenses, F5 Networks suggests using the maximum configuration limits for the BIG-IP VE system. For lab editions, required reserves can be less. For each virtual machine, the KVM virtual machine guest environment permits a maximum of 10 network adapters. You can either deploy these as a management port and 9 dataplane ports or a management port, 8 dataplane ports, and an HA port.
There are also some maximum configuration limits to consider for deploying a BIG-IP VE virtual machine, such as:
The size of the image that you choose to download determines both the number of slots and the number and type of modules that are supported on the VE instance.
|Allocated disk space||Supported module combinations||Module specific concerns|
|7 GB||LTM only on a single slot||You cannot install upgrades or hotfixes to this version.|
|31 GB||LTM only on two slots.||This option can be extended and upgraded with new versions and hot fix updates. It does not allow installing any modules besides LTM, GTM, or LTM + GTM.|
|100 GB||Supports all modules. Two slots are supported with potential room to install a third.||
This option can be extended and upgraded with new versions and hot fix updates. It allows installing any combination of other modules supported by the current version of BIG-IP VE software.
The general memory requirement recommendation for BIG-IP Virtual Edition (VE) is 2 GB per virtual CPU. Additionally, the following memory guidelines may be helpful in setting expectations based on which modules are licensed on VE guests.
|Provisioned memory||Supported module combinations||Module specific concerns|
|12 GB or more||All module combinations are fully supported.||N/A|
|8 GB||Provisioning more than three modules together is not supported.||GTM and Link Controller do not count toward the module-combination limit.|
|More than 4 GB, but less than 8 GB||Provisioning more than three modules together is not supported. (See module-specific concerns relating to AAM).||
Application Acceleration Manager (AAM) cannot be provisioned with any other module; AAM can only be provisioned as Standalone.
GTM and Link Controller do not count toward the module-combination limit.
|4 GB or less||Provisioning more than two modules together is not supported.||AAM can only be provisioned as Dedicated.|
If you want to disable support for TCP Segmentation Offloading (TSO), you must submit a tmsh command, because the TSO feature is enabled by default. Note that enabling TSO support also enables support for large receive offload (LRO) and Jumbo Frames.
You must have the Admin user role to enable or disable TSO support for a hypervisor.
If you want support for SR-IOV, in addition to using the correct hardware and BIOS settings, you must configure hypervisor settings before you set up the guests.
You must have an SR-IOV-compatible network interface card (NIC) installed, and the SR-IOV BIOS enabled before you can configure SR-IOV support.
Refer to the documentation included with your hypervisor operating system for information on support and configuration for SR-IOV.