Converged networks with Fibre Channel over Ethernet and Data Center Bridging

7
The ENode (Figure 4) consists of these components:
• FCoE Controller uses FCoE Initialization Protocol (FIP) to discover the SAN fabrics through the FCFs and
provisions the virtual N_Ports (VN_Ports) and FCoE Link End Points (LEPs).
• FCoE LEPs convert FC frames to FCoE frames on the transmit side, and convert FCoE frames to FC frames
on the receive side. There is one LEP for each VN_Port established in the ENode.
• VN_Ports instantiate virtual N_Ports with N_Port ID Virtualization (NPIV) capability similar to a traditional
FC HBA. The VN_Ports in an ENode include information about the MAC address to WWN translations
required for proper communications with FCFs in a converged network.
• FC Function is the traditional logic implemented in an FC HBA. It handles storage discovery, storage
connection management, error recovery, and host bus (PCIe) interface interoperation to upper layer
driver/SCSI drivers. Again, this function behaves so much like an FC HBA function that CNAs and HBAs
from the same vendors typically use the same storage drivers in the host operating systems to control
them. This makes deploying both converged and non-converged systems in a data center very easy
during the transition to a converged infrastructure.
Figure 4. FCoE architecture components
FCoE and Ethernet
FCoE requires DCB-enabled Ethernet. The IEEE is working to enhance the IEEE 802 network standards to
allow FC, or any TC requiring lossless behavior, to run efficiently over many types of IEEE 802 compliant,
MAC layer protocols, including Ethernet. We expect the FCoE standard ratification in late 2010. It is
important to understand that FCoE will not work on legacy Ethernet networks because it requires a lossless
form of Ethernet.
FC cannot handle dropped frames as Ethernet allows today. It is possible to create a
lossless Ethernet network using existing IEEE 802.3x flow control mechanisms. If the network carries multiple