White Paper

Send documentation comments to [e-mail]
1-3
Cisco Fabric Manager Server Federation Deployment Guide
OL-21765-01, Cisco MDS NX-OS Release 5.x
Chapter 1 Cisco Fabric Manager Server Federation Overview
Fabric Manager Servers and the Fabric Manager Web Clients in the cluster share a common database.
The HTTP load balancer interfaces with the web servers to handle requests from web clients and the
Fabric Manager desktop client will access Fabric Manager Servers in a cluster through the cluster proxy.
Figure 1-2 Federated Server Architecture
The relationships between physical server, logical server, and the fabrics are maintained by a centralized
database. The mappings from logical to physical server as well as the mappings with the fabrics are
managed by each Fabric Manager Server.
• Logical Server—Describes the functional layout – logically they appear to be different servers. In
network topologies, a logical topology describes the paths that data can take across a network
irrespective of how they are connected to each other.
• Physical Server— A physical server describes how the system is connected together in the physical
world.
• Fabric— A fabric is similar to a network segment in a local area network. A typical Fibre Channel
SAN fabric is made up of a number of Fibre Channel switches.
You can move fabrics from one logical server to another logical server. The logical servers also can be
moved from one physical machine to another physical machine as well depending upon your resource
constraints or requirements as shown in Figure 1-3. The mapping table entries should remain
synchronized with the changes. This design helps to redistribute workloads, saves from manual failover,
helps to optimize performance or scalability, and avoids collision of server IDs.