| Fibre Channel | |
|---|---|
| Layer 4. Protocol mapping | |
| LUN masking | |
| Layer 3. Common services | |
| Layer 2. Network | |
| Fibre Channel fabric Fibre Channel zoning Registered state change notification | |
| Layer 1. Data link | |
| Fibre Channel 8b/10b encoding | |
| Layer 0. Physical |
Switched fabric orswitching fabric is anetwork topology in which networknodes interconnect via one or morenetwork switches[1] (particularlycrossbar switches). Because a switched fabric network spreadsnetwork traffic across multiple physical links, it yields higher totalthroughput thanbroadcast networks, such as the early10BASE5 version ofEthernet and mostwireless networks such asWi-Fi.
The generation of high-speedserial data interconnects that appeared in 2001–2004 which providedpoint-to-point connectivity between processor andperipheral devices are sometimes referred to as fabrics; however, they lack features such as amessage-passing protocol.[citation needed] For example,HyperTransport, the computer processor interconnect technology, continues to maintain aprocessor bus focus even after adopting a higher speed physical layer. Similarly,PCI Express is just a serial version of PCI; it adheres to PCI's host/peripheral load/storedirect memory access (DMA)-based architecture on top of a serial physical and link layer.
In theFibre ChannelSwitched Fabric (FC-SW-6) topology, devices are connected through one or moreFibre Channel switches. While this topology has the bestscalability of the three FC topologies (the other two areArbitrated Loop andpoint-to-point),[2] it is the only one requiring switches, which are costly hardware devices.
Visibility among devices (callednodes) in a fabric is typically controlled withFibre Channel zoning.
Multiple switches in a fabric usually form amesh network, with devices being on the "edges" ("leaves") of the mesh. Most Fibre Channel network designs employ two separate fabrics forredundancy. The two fabrics share the edge nodes (devices), but are otherwise unconnected. One of the advantages of such a setup is the capability offailover, meaning that in case one link breaks or a fabric goes out of order,datagrams can be sent via the second fabric.
The fabric topology allows the connection of up to the theoretical maximum of about 16 million devices, limited only by the available address space (224).
239 domains * 256 areas * 256 ports = 15,663,104[3]
