Fibre Channel Protocol Architecture
Fibre Channel Layered Model
Fibre Channel Layered Model
Fibre Channel is to SANs what Ethernet is to LANs
…and more…
– Ethernet just defines
how data is physically
Applications Applications
transmitted
– Protocols like IP define
SCSI and
the logical network
other protocols
TCP and
other protocols
– Fibre Channel defines
Logical
IP and
both physical and other protocols
logical aspects
Fibre Channel
Physical
Ethernet
Why does Fibre Channel use a layered model?
Each layer has a specific function that it performs on the data as it
passes between the application and operating system, the FC network,
and the storage devices.
Separating functions into layers allows future standards to change the
functionality implemented in one layer without impacting the other layers.
This allows vendors to adapt more quickly to evolving standards, and it
facilitates backwards compatibility.
The tiered architecture simplifies debugging by allowing the analyst to
first determine the layer at which the failure occurred, and then focus on
the components of that layer.
The FC-4 layer interfaces with the upper layer protocols and maps to
Fibre Channel in order to use Fibre Channel’s physical transport.
The lower three layers are specified in the FC-PH set of standards
documents.