Overview of Fibre Channel
Fibre Channel (FC) is a high-speed network technology that interconnects network elements and allows them to communicate with one another. The International Committee for Information Technology Standards (INCITS) T11 Technical Committee sets FC standards.
FC networks provide high-performance characteristics such as lossless transport combined with flexible network topology. FC is primarily used in storage area networks (SANs) because it provides reliable, lossless, in-order frame transport between initiators and targets. FC components include initiators, targets, and FC-capable switches that interconnect FC devices and may also interconnect FC devices with Fibre Channel over Ethernet (FCoE) devices. Initiators originate I/O commands. Targets receive I/O commands. For example, a server can initiate an I/O request to a storage device target.
The Juniper Networks QFX3500 Switch has native FC ports as well as Ethernet access ports, and can function as an FCoE-FC gateway or as an FCoE transit switch. All other QFX Series switches and EX4600 switches have Ethernet access ports and can function as an FCoE transit switch.
FCoE transports native FC frames over an Ethernet network by encapsulating the unmodified frames in Ethernet. It also provides protocol extensions to discover FCoE devices through the Ethernet network. FCoE requires that the Ethernet network support data center bridging (DCB) extensions that ensure lossless transport and allow the Layer 2 Ethernet domain to meet the requirements of FC transport.
The FCoE-FC gateway functionality is a licensed feature on the QFX Series that is available only on QFX3500 switches. As an FCoE-FC gateway, the switch connects FCoE devices on an Ethernet network to a SAN FC switch.
You do not need a license to use the switch as an FCoE transit switch. As an FCoE transit switch, the switch:
Is a Layer 2 data center bridging (DCB) switch that can transport FCoE frames.
Implements FCoE Initialization Protocol (FIP) snooping.
Connects multiple FCoE endpoints to the FC network.
Standalone switches support FCoE. Virtual Chassis (VC) and mixed-mode Virtual Chassis Fabric (VCF) configurations do not support FCoE. Pure QFX5100 switch VCFs (consisting of only QFX5100 switches) support FCoE.
This topic describes:
Fibre Channel Transport Protocol
The Fibre Channel Protocol is a transport protocol that consists of five layers as shown in Table 1:
FC Protocol Layer |
Description |
|---|---|
FC-0 |
Physical (cabling, connectors, and so on) |
FC-1 |
Data link layer |
FC-2 |
Network layer (defines the main protocols) |
FC-3 |
Common services |
FC-4 |
Protocol mapping |
The FC protocol layers are generally split into three groups:
FC-0 and FC-1 are the physical layers.
FC-2 is the protocol layer, similar to OSI Layer 3.
FC-3 and FC-4 are the services layers.
The FCoE-FC gateway operates the physical layers and the protocol layer, and provides FIP and service redirection at the services layer.
How FC Works on the Switch
The switch connects devices that support FC and Ethernet (such as FCoE servers on an Ethernet network) to an FC SAN, thus converging the Ethernet and FC networks on a single physical network infrastructure. The switch provides the class-of-service (CoS) features needed to handle the different types of traffic appropriately.
To converge FC and Ethernet networks, you can configure the switch as an:
FCoE-FC Gateway
When the switch functions as an FCoE-FC gateway, the switch aggregates FCoE traffic and performs the encapsulation and de-encapsulation of native FC frames in Ethernet as it transports the frames between FCoE devices in the Ethernet network and the FC switch. In effect, the switch translates Ethernet to FC and FC to Ethernet.
The gateway receives FC frames encapsulated in Ethernet from FCoE devices through an FCoE VLAN interface composed of one or more 10-Gigabit Ethernet interfaces. The gateway removes the Ethernet encapsulation from the FC frames, and then sends the native FC frames to the FC switch through a native FC interface.
The gateway receives native FC frames from the FC switch on the gateway’s native FC interfaces. The gateway encapsulates the native FC frames in Ethernet, and then sends the encapsulated frames to the appropriate FCoE device through the FCoE VLAN interface.
To FCoE devices, the gateway behaves like an FC switch and can present multiple virtual F_Ports (VF_Ports) on a single interface. To an FC switch, the gateway behaves like an FC node that is doing N_Port ID virtualization (NPIV).
FCoE Transit Switch
When the switch functions as an FCoE transit switch, it forwards traffic (including FCoE traffic) based on Layer 2 media access control (MAC) forwarding and is a normal DCB-enabled Layer 2 switch that also performs FIP snooping. The switch aggregates FCoE traffic and passes it through to an FCF. The switch does not remove the Ethernet encapsulation from the FC frames, but it does preserve the class of service (CoS) required to transport FC frames.
The switch inspects (snoops) FIP information in order to create filters that permit only valid FCoE traffic to flow through the switch between FCoE devices and the FCF. The switch does not use native FC ports because the FC frames are encapsulated in Ethernet when they flow between the FCoE devices and the FCF. Virtual point-to-point links between each FCoE device and the FCF pass transparently through the switch, so the switch is not seen as a terminating point or an intermediate point by FCoE devices or by the FCF.
FCoE VLANs
All FCoE traffic must travel in a VLAN dedicated to transporting only FCoE traffic. Only FCoE interfaces should be members of an FCoE VLAN. Ethernet traffic that is not FCoE or FIP traffic must travel in a different VLAN.
The same VLAN cannot be used in both transit switch mode and FCoE-FC gateway mode.
FCoE VLANs (any VLAN that carries FCoE traffic) support only Spanning Tree Protocol (STP) and link aggregation group (LAG) Layer 2 features.
FCoE traffic cannot use a standard LAG because traffic might be hashed to different physical LAG links on different transmissions. This breaks the (virtual) point-to-point link that Fibre Channel traffic requires. If you configure a standard LAG interface for FCoE traffic, FCoE traffic might be rejected by the FC SAN.
QFabric systems support a special LAG called an FCoE LAG, which enables you to transport FCoE traffic and regular Ethernet traffic (traffic that is not FCoE traffic) across the same link aggregation bundle. Standard LAGs use a hashing algorithm to determine which physical link in the LAG is used for a transmission, so communication between two devices might use different physical links in the LAG for different transmissions. An FCoE LAG ensures that FCoE traffic uses the same physical link in the LAG for requests and replies in order to preserve the virtual point-to-point link between the FCoE device converged network adapter (CNA) and the FC SAN switch across the QFabric system Node device. An FCoE LAG does not provide load balancing or link redundancy for FCoE traffic. However, regular Ethernet traffic uses the standard hashing algorithm and receives the usual LAG benefits of load balancing and link redundancy in an FCoE LAG.
IGMP snooping is enabled by default on all VLANs in all software versions before Junos OS R13.2. Disable IGMP snooping on FCoE VLANs if you are using software that is older than 13.2.
You can configure more than one FCoE VLAN, but any given virtual link must be in only one FCoE VLAN.
All 10-Gigabit Ethernet interfaces that connect to FCoE devices must have a native VLAN configured in order to transport FIP traffic, because FIP VLAN discovery and notification frames are exchanged as untagged packets.
Only FCoE traffic is permitted on the FCoE VLAN. A native VLAN might need to carry untagged traffic of different types and protocols. Therefore, it is a good practice to keep the native VLAN separate from FCoE VLANs.
Supported FC Features and Functions
The following features and functionality are supported:
As an FCoE-FC gateway:
DCB, including Data Center Bridging Capability Exchange protocol (DCBX), priority-based flow control (PFC), enhanced transmission service (ETS), and 10-Gigabit Ethernet interfaces
FCoE Initialization Protocol (FIP)
Proxy for FCoE devices when communicating with FC switches and acts as a proxy for FC switches when communicating with FCoE devices
Up to 12 native FC interfaces per QFX3500 switch (each interface can be configured as a 2-Gigabit, 4-Gigabit, or 8-Gigabit Ethernet interface)
As an FCoE transit switch:
DCB functions
FIP snooping
Transparent Layer 2 MAC forwarding of FCoE frames
Lossless Transport Support
Up to six lossless forwarding classes are supported. For lossless transport, you must enable PFC on the IEEE 802.1p code point of lossless forwarding classes. The following limitations apply to support lossless transport:
The external cable length from a standalone switch or QFabric system Node device to other devices cannot exceed 300 meters.
The internal cable length from a QFabric system Node device to the QFabric system Interconnect device cannot exceed 150 meters.
For FCoE traffic, the interface maximum transmission unit (MTU) must be at least 2180 bytes to accommodate the packet payload, headers, and checks.