The VIPRION® system is a complete traffic management solution that offers high performance, reliability, scalability, and ease of management. Based on chassis and blade technology, this system is designed to meet the needs of large, enterprise networking environments that normally require multiple BIG-IP® systems to process large volumes of application traffic.
The VIPRION system includes multiple blades that work together as a powerful cluster to process application traffic. When traffic comes into a single virtual server, the system distributes that traffic over multiple blades using the full multi-processing capacity of each blade. This ensures that other blades can complete the processing of the request if one unexpectedly becomes unavailable.
This illustration shows a typical VIPRION system with a four-slot cluster processing traffic destined for virtual server vs_http. In this example, the virtual server resides on all blades in the cluster, due to a process known as cluster synchronization. The primary blade receives the client traffic and then uses the power of all blades in the cluster to process the traffic before sending the traffic to the appropriate server.
Example of a basic VIPRION system
The VIPRION® system includes SuperVIP® cluster technology, the core feature that coordinates all of the blades into a single high-performance system. A cluster is a group of active slots in the VIPRION system chassis. The size of the cluster depends on the number of running blades installed in the chassis. Cluster technology provides the processing power of multiple blades, but you manage the entire cluster as a single system. When you install a blade in a slot and power on the blade, the slot automatically becomes a member of the cluster.
Each slot in the cluster represents a cluster member, and the blades in the slots of a cluster work together to process application traffic. Moreover, the blades can be configured to mirror each other's connections so that if a blade is taken out of service or becomes unavailable for some reason, any in-process connections remain intact.
One of the tasks you performed as part of the hardware installation was to assign a unique cluster IP address to the primary slot in the cluster. This cluster IP address is a floating management IP address used to access the blade in the primary slot to manage the system. If the blade in the primary slot becomes unavailable for any reason, the primary designation moves to a different slot, and the cluster IP address floats to that slot.
The VIPRION® system automatically performs cluster synchronization, an internal process that causes the primary blade to automatically propagate the BIG-IP ®software configuration to all secondary blades, even when a new blade is introduced into the cluster. Cluster synchronization allows all blades in the cluster to work together to process incoming traffic, and ensures that you can always access the cluster using the cluster IP address, even when the blade in the primary slot changes.
The number of slots in a chassis varies depending on the chassis model. For example, while some chassis models contain two or four slots, the VIPRION® C4800 Series contains eight slots.
Each chassis model requires a specific blade type. For example, the VIPRION C4800 chassis uses VIPRION B4300 Series blades. For specific information on blade types compatible with your chassis, consult the platform guide for your chassis series.
If you need multi-tenancy, you can optionally provision the VIPRION® system for virtual Clustered Multiprocessing (vCMP®). Provisioning vCMP creates a hypervisor and allows you to create guests on the system for multi-tenant processing.
This illustration shows a basic vCMP system with a host and four guests. Note that each guest has a different set of modules provisioned, depending on the guest's particular traffic requirements.
Example of a four-guest vCMP system on a VIPRION chassis
For more information, see the vCMP product documentation at F5 Networks® knowledge web site http://support.f5.com.
There are several cluster-related terms that are helpful to understand.
|cluster||The group of active slots in the chassis. The blades in the cluster work together as one powerful system to process application traffic. Also known as a SuperVIP® cluster.|
|cluster member||An enabled physical slot (or a virtual slot) that contains an active blade.|
|cluster IP address||The floating management IP address of the slot designated as the primary slot. You normally assign the cluster IP address during VIPRION chassis installation.|
|primary slot||The slot containing the blade that initially accepts application traffic. The floating cluster IP address is assigned to the primary slot. If the blade in the primary slot becomes unavailable, the cluster IP address automatically floats to another cluster member and the slot of the new cluster member becomes the primary slot.|
|primary blade||The blade residing in the primary slot.|
|secondary slot||Any slot that is not the primary slot and therefore does not have the floating cluster IP address assigned to it.|
|secondary blade||Any blade residing in a secondary slot.|
|cluster member IP address||The static management IP address assigned to a cluster member.|
|cluster synchronization||An ongoing, internal process by which the primary blade automatically propagates the BIG-IP® system configuration to all secondary blades when powered-on. Cluster synchronization allows all blades in the cluster to work together to process network traffic.|