WO2014169439A1 - Method, device and system for realizing fcoe - Google Patents

Abstract

Provided are a method, device and system for realizing an FCoE. The method comprises: a first FEF acquiring a first fibre channel identifier (FC-ID) range allocated by an FCS; receiving forwarding information sent by the FCS, wherein the forwarding information comprises: a correlation between an MAC address of another FEF and the FC-ID range allocated to the other FEF by the FCS; and the first FEF updating a message forwarding table in accordance with the forwarding information, wherein the message forwarding table comprises a correlation between the media access control (MAC) address and the FC-ID range. The present invention enables the realization of an FCoE network to be much simpler.

Description

 Method, device and system for implementing FCoE

TECHNICAL FIELD The present invention relates to communications technologies, and in particular, to a method, apparatus, and system for implementing a fiber channel over ethernet (FCoE). Background technique

 By using FCoE technology, Fibre Channel can be mapped to Ethernet, Fibre Channel information can be inserted into Ethernet packets, allowing Fibre Channel requests and data from server-SAN storage devices to be transmitted over Ethernet connections without the need for specialized Fibre Channel structure.

 In the FCoE network architecture, the FCoE forwarder (FCF) is responsible for forwarding FCoE packets. The FCoE packets sent by the source device in the network need to pass the FCF. The FCF is used to forward the packets. The routing table of the text finds the forwarding path of the packet and forwards the packet to the destination device. Since an FCF usually needs to use a separate domain identifier, and the number of domain identifiers (IDs) is limited, the entire network size is also limited by the number of FCFs, resulting in poor scalability of the FCF networking. To this end, a new functional entity, FCoE data-plane forwarder FCoE (FDF), is used in the prior art to be forwarded by the FCF, and the combination of FCF and FDF can be used. To reduce the number of FCFs in this way, the FCF performs route calculation and sends the related information to the FDF. The FDF forwards the FCoE packets from the source device to the destination device according to the routing table.

 However, whether it is the networking of FCF or the combination of FCF and FDF, the FCoE packet forwarding and routing are very complicated, that is, the control plane of FCoE is highly complex. Summary of the invention

 The present invention provides a method, apparatus and system for implementing FCoE to reduce the implementation complexity of the FCoE control plane.

In a first aspect, a method for implementing Fibre Channel over Ethernet FCoE is provided for FCoE The FCoE network includes: a Fibre Channel Control Plane Server FCS and at least one Fibre Channel Edge Forwarder FEF, wherein the at least one FEF includes a first FEF; the method includes: the first FEF receiving the FCS allocation The first Fibre Channel identifies the FC-ID range; and further receives the forwarding information sent by the FCS, where the forwarding information includes: a medium access control MAC address of another FEF and an FC-ID allocated by the FCS to the other FEFs Corresponding relationship of the range; the first FEF, according to the forwarding information, updating a packet forwarding table of the first FEF, where the packet forwarding table includes a correspondence between a MAC address and an FC-ID range;

 And updating, by the first FEF, the packet forwarding table of the first FEF according to the forwarding information, specifically: the packet forwarding table of the first FEF according to the forwarding information by the first FEF The correspondence between the MAC address of the other FEF and the FC-ID range is recorded.

 With reference to the first aspect, in a first possible implementation manner, after the first FEF acquires the first FC-ID range of the FCS allocation, the method further includes: the first FEF receiving the first local a connection request sent by the node; the first FEF allocates a first node FC-ID to the first local node from the first FC-ID range according to the connection request; the FEF is according to the first And updating, by the node FC-ID, the packet forwarding table of the first FEF, where the FEF updates the packet forwarding table of the first FEF according to the first node FC-ID, specifically: The first FEF records the correspondence between the MAC address of the first local node and the first node FC-ID in the packet forwarding table of the first FEF according to the first node FC-ID.

 With reference to the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner, the method further includes: the first FEF receiving the exit information sent by the FCS, the exiting The information includes a MAC address of the second FEF, where the first FEF updates the packet forwarding table of the first FEF according to the exit information, where the first FEF updates the first information according to the exit information. The packet forwarding table of the FEF includes: deleting, by the first FE F, the forwarding entry corresponding to the MAC address of the second FEF from the packet forwarding table of the first FEF according to the exit information.

With reference to any one of the first aspect to the second possible implementation manner of the first aspect, in a third possible implementation, the first Fibre Channel identifier FC of the FCS allocation is obtained in the first FEF Before the ID range, the method further includes: the first FEF establishing a connection with the FCS; the establishing, by the first FEF, the FCS is configured to: the first FEF is statically configured with an FCS address, and according to Establishing a connection between the FCS address and the FCS; or: the first FEF The dynamic discovery mode establishes a connection with the FCS.

 With reference to the first aspect to any one of the third possible implementation manners of the first aspect, in a fourth possible implementation, the method further includes: the first FEF receiving an FCoE report from a local node Obtaining a destination FC-ID from the FCoE packet; the first FEF searches the packet forwarding table of the first FEF according to the destination FC-ID, and obtains a packet corresponding to the destination FC-ID. The MAC address of the destination FEF; the first FEF sets the MAC address of the destination FEF to the destination MAC address of the FCoE packet, and sends the FCoE packet.

 In a second aspect, a method for implementing an Ethernet Fibre Channel FCoE is provided, which is applied to an FCoE network, where the FCoE network includes a Fibre Channel Control Plane Server FCS and at least one Fibre Channel Edge Forwarder FEF, where the at least one FEF includes a An FEF; the method includes: the FCS assigning a first Fibre Channel identification FC-ID range to the first FEF; the FCS transmitting the first FC-ID range, and forwarding information to the first And the FEF is configured to update the packet forwarding table of the first FEF according to the forwarding information, where the forwarding information includes: a media access control MAC address of another FEF, and the FCS is allocated to the other Correspondence of the FC-ID range of the FEF; the correspondence table of the MAC address includes the correspondence between the MAC address and the FC-ID range.

 With reference to the second aspect, in a first possible implementation, after the FCS allocates the first FC-ID range to the first FEF, the method further includes: the FCS corresponding to the first FEF The first forwarding information is sent to the other FEF, so that the other FEFs update the packet forwarding table of the other FEF according to the first forwarding information, where the first forwarding information includes: the first FEF Correspondence between the MAC address and the first FC-ID range.

 With reference to the second aspect or the first possible implementation manner of the second aspect, in a second possible implementation manner, the method further includes: sending, by the FCS, the exit information to the first FEF, the exit information The MAC address of the second FEF is included in the first FEF, so that the first FEF updates the packet forwarding table of the first FEF according to the exit information.

 In a third aspect, a Fibre Channel edge repeater FEF is provided, including:

An information receiving unit, configured to receive a first Fibre Channel identification FC-ID range allocated by the Fibre Channel control plane server FCS; and further receive forwarding information sent by the FCS, where the forwarding information includes: a medium access control MAC address of another FEF Corresponding relationship between the FC-ID range allocated by the FCS to the other FEFs; a forwarding table update unit, configured to update, according to the forwarding information, a packet forwarding table of the FEF, where the packet forwarding table includes a correspondence between a MAC address and an FC-ID range; Corresponding relationship between the MAC address of the other FEF and the FC-ID range is recorded in the message forwarding table of the FEF.

 With reference to the third aspect, in a first possible implementation, the method further includes: a request receiving unit, configured to receive a connection request sent by the first local node, and an address allocation unit, configured to: according to the connection request, The first local node is allocated a first node FC-ID in an FC-ID range; the forwarding table update unit is further configured to update the FEF packet forwarding table according to the first node FC-ID Specifically, the mapping between the MAC address of the first local node and the first node FC-ID is recorded in the packet forwarding table of the FEF according to the first node FC-ID.

 With the third aspect or the first possible implementation manner of the third aspect, in a second possible implementation manner, the information receiving unit is further configured to receive the exit information sent by the FCS, where the exit information is The forwarding table update unit is further configured to: update the packet forwarding table of the first FEF according to the exit information; specifically, according to the exit information, from the first The forwarding entry corresponding to the MAC address of the second FEF is deleted in the packet forwarding table of the FEF.

 With reference to any one of the third aspect to the second possible implementation of the third aspect, in a third possible implementation, the method further includes: a connection processing unit, configured to establish a connection with the FCS; And establishing a connection with the FCS according to the statically configured FCS address; or acquiring the FCS address by using a dynamic discovery manner, and establishing a connection with the FCS.

 With reference to any one of the third aspect to the third possible implementation manner of the third aspect, in a fourth possible implementation, the method further includes: a packet forwarding unit, where the packet forwarding unit is configured to receive Obtaining an FCoE packet from the local node, and obtaining a destination FC-ID from the FCoE packet; searching for a packet forwarding table of the FEF according to the destination FC-ID, and obtaining a packet corresponding to the destination FC-ID The MAC address of the destination FEF, the MAC address of the destination FEF is set to the destination MAC address of the FCoE packet, and the FCoE packet is sent.

 In a fourth aspect, a Fibre Channel control plane server FCS is provided, including:

An address allocation unit, configured to allocate a first Fibre Channel identification FC-ID range to the first FEF, and an information sending unit, configured to send the first FC-ID range, and forwarding information to the first FEF, so that Updating, by the first FEF, the packet of the first FEF according to the forwarding information a forwarding table, where the forwarding information includes: a media access control MAC address of another FEF, and a correspondence between the FC-ID range allocated by the FCS to the other FEF, where the forwarding table includes a MAC address and an FC - The correspondence of the ID range.

 With reference to the fourth aspect, in a first possible implementation, the information sending unit is further configured to send the first forwarding information corresponding to the first FEF to the other FEF, so that the other FEF is configured according to The first forwarding information is used to update the packet forwarding table of the other FEF, where the first forwarding information includes: a correspondence between the MAC address of the first FEF and the first FC-ID range.

 With reference to the fourth aspect, or the first possible implementation manner of the fourth aspect, in the second possible implementation, the information sending unit is further configured to send the exit information to the first FEF, where the exit information is The MAC address of the second FEF is included in the first FEF, so that the first FEF updates the packet forwarding table of the first FEF according to the exit information.

 In a fifth aspect, a system for implementing Fibre Channel over Ethernet FCoE is provided, comprising at least one Fibre Channel edge repeater FEF of the present invention, and a Fibre Channel Control Plane Server FCS of the present invention.

 The technical effect of the method, device and system for implementing FCoE provided by the present invention is: a first FEF in at least one FEF in the FCoE network, the first FEF can receive the first FC-ID range allocated by the FCS, and also receive the FCS transmission The forwarding information includes: a MAC address of another FEF and a correspondence between the FCS and the FC-ID range allocated by the FCS; the first FEF can update the packet forwarding table according to which the FCoE packet is forwarded. Recording the correspondence between the MAC address and the FC-ID range in the packet forwarding table; the manner of forming the forwarding table is very simple and fast, compared with the prior art master FCS every network update. To recalculate the entire network route, the formation of the packet forwarding table is simplified, and the implementation of the control plane of the FCoE network is simpler, which reduces the implementation complexity of the FCoE control plane. DRAWINGS

 1 is a schematic diagram of a networking architecture of an FCoE network according to an embodiment of the present invention; FIG. 2 is a flowchart of a method for implementing FCoE according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of another networking architecture of an FCoE network according to an embodiment of the present invention; FIG. 4 is a flowchart of another method for implementing FCoE according to an embodiment of the present invention; FIG. 5 is a flowchart of a method for implementing FCoE according to an embodiment of the present invention; FIG. 6 is a schematic diagram of signaling interaction for implementing FCoE according to an embodiment of the present invention; FIG. 7 is a schematic diagram of a method for implementing FCoE according to an embodiment of the present invention; FIG. 8 is a schematic diagram of packet changes in a method for implementing FCoE according to an embodiment of the present invention; FIG. 9 is a schematic structural diagram of an FEF according to an embodiment of the present invention;

 FIG. 10 is another schematic structural diagram of an FEF according to an embodiment of the present disclosure;

 FIG. 11 is still another schematic structural diagram of an FEF according to an embodiment of the present invention;

 FIG. 12 is a schematic structural diagram of an FCS according to an embodiment of the present disclosure;

 FIG. 13 is a schematic structural diagram of an entity structure of an FEF according to an embodiment of the present disclosure;

 FIG. 14 is a schematic diagram of an entity structure of an FCS according to an embodiment of the present invention. detailed description

 FIG. 1 is a schematic diagram of a networking architecture of an FCoE network according to an embodiment of the present invention. The embodiment of the present invention provides a method for implementing FCoE, which may be based on the FCoE network networking architecture shown in FIG. As shown in FIG. 1 , the FCoE network includes two functional entities: a Fibre Channel Control Plane Server (FCCS) and an FC Edge Forwarder (FEF).

 The FCS participates in the operation of the control plane and the management plane of the FCoE network, and does not participate in the forwarding of the data plane of the FCoE network. It is used to centrally control and manage the FEF, and send relevant signaling and parameters to the FEF.

 The FEF is used to receive the control and management of the FCS, and receive the relevant signaling and parameters from the FCS. It is also used to generate a packet forwarding table, and forward the FCoE packet according to the packet forwarding table, that is, complete the data plane of the FCoE network. Forward.

In Figure 1, four FEFs are managed under the FCS1. However, the number and topology of the FEFs are not limited in the specific implementation. The FCS1 can be connected to each FEF through a common Ethernet switch or other device. Both the server 1 (Server1) and the storage 2 (storage2) are traditional FCoE end nodes (ENodes), and the ENode nodes are node devices in the network that forward FCoE in the FCoE mode, for example, the server controller local area network. (controller area network, abbreviation: CAN) NIC, FCoE switch, FCoE-enabled storage devices, etc. When the ENode node is connected to the FEF, the port of the ENode node is called a VN port (VN port), and the port of the FEF is called a VF port (VF port). The FCoE initialization protocol can be adopted between the VN port and the VF port. (FCoE initialization protocol, FIP for short) Initialize the connection.

 The method for implementing the FCoE according to the embodiment of the present invention includes: Each FEF, specifically FEF1, FEF2, FEF3, and FEF4, establishes a connection with the FCS1 responsible for the main control, and is based on the FCS1. The related information is sent to generate a packet forwarding table for forwarding packets, and then the FCoE packet is forwarded according to the packet forwarding table. Therefore, the FCoE network shown in Figure 1 can perform the function of forwarding FCoE packets. If a new FEF is added to the FCoE network shown in Figure 1, for example, the FEF5 is required to be inbound, and the packet forwarding table of each FEF in the network, specifically FEF1, FEF2, FEF3, and FEF4, is updated; If a new ENode is added, the packet forwarding table of the FEF connected to the new ENode needs to be updated, and the packet is forwarded according to the updated forwarding table.

 Next, in the embodiment of the present invention, a schematic diagram of the networking architecture of the FCoE network shown in FIG. 1 is used to describe how to implement FCoE, including a control plane and a data plane; for example, a packet forwarding table of the control plane is generated, and data is generated. Forwarding of flat FCoE packets.

 2 is a flowchart of a method for implementing FCoE according to an embodiment of the present invention. As shown in FIG. 2, the method may include:

 201. The first FEF receives the first fiber channel identifier (FC-ID) range allocated by the FCS to the first FEF; and further receives the forwarding information sent by the FCS, where the forwarding information includes: another FEF a media access control (MAC) address and a correspondence between the FCS and the FC-ID range allocated by the FCS to the other FEFs;

 In this embodiment, any FEF of the FCoE network receives the FC-ID range allocated by the FCS when it newly joins the FCoE network; the FC-ID range obtained by each FEF is mainly used for allocation to the access. The ENode of the FEF; the FC-ID is mainly used for addressing and forwarding FCoE packets in the network.

 Optionally, the foregoing FC-ID range may be represented by multiple types, for example, by masking ( 1.1.0/16 ), or by consecutive n starting from a certain FC-ID (n> 0) , for example, 殳 an FC-ID range is ( 1.1.1 10 ), which means that the FC-ID range is 10 consecutive from the FC-ID of 1.1.1, ie 1.1.1, 1.1 .2, 1.1.3 1.1.10. There may be other representations in the specific implementation, and are not limited to the above examples.

In addition, assuming that there are multiple FEFs in the FCoE network, the FCS allocates different FC-ID ranges for each FEF, that is, the FC-ID ranges to which different FEFs are allocated cannot overlap. More specifically, when there is an FCS in the FCoE network, the FC-ID range of each FEF under the FCS cannot be assigned. When there are multiple FCSs in the FCoE network, the FC-ID ranges controlled by different FCSs cannot overlap, that is, the FC-ID ranges assigned to the FEFs belonging to different FCSs cannot be overlapped. The FC-ID The non-overlapping range means that the whole network is unique.

 There are several ways to achieve non-overlapping FC-ID ranges. The following are several examples, but the specific implementation is not limited to this: For example, with an FCS as an example, which FC-ID range can be recorded on the FCS, Then, when the FC-ID range is assigned to other FEFs next time, these used ones will not be allocated again, so a single FCS is still easy to control without overlapping. In addition, taking two FCSs in the network as an example, if FCS1 and FCS2 exist, in order to ensure that the FC-ID ranges allocated by the two FCSs do not overlap, the following methods can be used:

 For example, the FC-ID resources that can be controlled by each of them can be pre-configured on the two FCS1 and FCS2. For example, if the entire FC-ID resource is not a resource pool, FCS1 can be configured to control the first part of the FC in the resource pool. -ID resource, the subsequent FCS1 allocates the FC-ID range to each of its subordinate FEFs, and selects the allocation from the first part of the FC-ID resource; and the configuration FCS2 is responsible for controlling the second part of the FC-ID in the resource pool. For the resource, the subsequent FCS2 allocates the FC-ID range for each of its subordinate FEFs, and then selects the allocation from the second part of the FC-ID resource, so that there is no overlap between the two FCSs.

 For example, the interaction signaling between the FCS1 and the FCS2 may be shared, and the FC-ID range allocation information is shared by the interaction signaling. For example, after the FCS1 allocates the FC-ID range to a certain FEF of the subordinate. The first FC-ID range will be notified to FCS2 by interactive signaling, so that FCS2 can know that the first FC-ID range has been used in FCS1, and FCS2 will not use the first one when it is allocated to its subordinate FEF. A range of FC-IDs. Similarly, FCS2 will also notify FCS1 of its assigned FC-ID range, and the information exchange between the two FCSs can avoid the repeated allocation of the FC-ID range.

 For another example, suppose FCS 1 and FCS2 select an FC-ID range allocation from the FC-ID resource pool (for example, an FC-ID database) when allocating FC-ID ranges for their respective FEFs, then FCS1 is allocated. The FC-ID range that has already been assigned can be marked later to indicate that the FC-ID range has been assigned, and FCS2 will view the FC-ID range to be assigned when selecting from the FC-ID resource pool. The tag, if it is already marked, will not be used, it will only be allocated using the un-allocated FC-ID range.

In this implementation, when the first FFE is added to the FCoE network, the FCS allocates a first FC-ID range to the first FEF, and the FCS records the first forwarding information corresponding to the first FEF. The first forwarding information includes a correspondence between a MAC address of the first FEF and the first FC-ID range.

 The first FC-ID range is controlled by the first FEF, and is used to allocate to the local node that accesses the first FEF, that is, the ENode that accesses the first FEF; specifically, the first FEF For each local node accessing the first FEF, an FC-ID is selected from the first FC-ID range and allocated to the local node.

 In addition, the first FC-ID range allocated by the FCS to the first FEF may be randomly selected by the FCS, or may be selected according to a certain rule policy, which is not limited in the embodiment of the present invention.

 In this implementation, the first FEF further receives the forwarding information sent by the FCS, where the forwarding information includes: a MAC address of another FEF and a correspondence between the FCS and the FC-ID range allocated by the FCS to the other FEF.

 Specifically, when the first FEF is newly added to the FCoE network, if there are other FEFs in the FCoE network, the first FEF receives the FCS transmission in addition to receiving the first FC-ID range allocated by the FCS. Forwarding information, the forwarding information may be stored in the FCS, and the FCS sends the forwarding information to the first FEF when the first FEF newly joins the network; the forwarding information and the first An FC-ID range may be carried in the same message and sent to the first FEF, or may be sent to the first FEF through different messages. In addition, after the first FEF is added to the FCoE network, if a new FEF is added to the FCoE, the first FEF receives the forwarding information sent by the FCS, that is, receives the forwarding corresponding to the new FEF sent by the FCS. The information includes a MAC address of the new FEF and a correspondence of the FC-ID range allocated by the FCS to the new FEF.

For example, in Figure 1, H殳FCS1 knows that FEF1 is newly added to the FCoE network, and the FC-ID range allocated for FEF1 is "1.1.0/16", and the FC-ID range allocated to FCS1 for other FEFs is as follows. As shown in Figure 1, the forwarding information received by FEF1 includes "FEF2-MAC, 1.2.0/16", "FEF3-MAC, 1.3.0/16" and "FEF4-MAC, 1.4.0/16", that is, FCS1 notifies FEF 1 of the FC-ID range corresponding to other FEFs in the network. Of course, the FC-ID range of FEF2~FEF4 is also allocated by FCS1 when FEF2~FEF4 is added to the network. FCS1 will record the FC-ID range assigned to each FEF, that is, record the MAC address of FEF and FC-. The correspondence between ID ranges is shown in Table 1 below. FEF and its FC-ID range

 202. The first FEF updates a packet forwarding table of the first FEF according to the forwarding information, where the packet forwarding table includes a correspondence between a medium access control MAC address and an FC-ID range.

 The packet forwarding table includes a correspondence between a MAC address and an FC-ID range. Further, the packet forwarding table may further correspond to an egress port. And updating, by the first FEF, the packet forwarding table of the first FEF according to the forwarding information, specifically: the first FEF is in the first FEF according to the forwarding information The correspondence between the MAC address of the other FEF and the FC-ID range is recorded in the publication. Further, an egress port corresponding to the other FEFs may also be included.

 In addition, when the correspondence between the MAC address of each FEF recorded by the FCS and the FC-ID range changes, the FCS will also notify all FEFs in the network. Specifically, when the first FEF is newly added to the FCoE network, the FCS sends the first forwarding information corresponding to the first FEF to other FEFs other than the first FEF in the FCoE network, so that other FEFs are according to the foregoing The forwarding information of the other FEF is updated by the forwarding information, and the first forwarding information includes a correspondence between the MAC address of the first FEF and the first FC-ID range. In this way, each FEF in the entire network can know the FC-ID range that is responsible for each, and this notification is specifically handled by the FCS.

 For example, as shown in Figure 1, after FEF1 is added to the network and FCS1 allocates the corresponding FC-ID range for the FEF1, FCS1 also forwards the forwarding information corresponding to FEF1, that is, the MAC address of FEF1 and its FC-ID range. , notify other FEFs, such as FEF2, FEF3, and FEF4, so that each FEF in the entire network knows the FC-ID range that each is responsible for, and this notification is handled by FCS1.

Further, after the first FEF receives the first FC-ID range that is allocated by the FCS, the method further includes: the first FEF receiving a connection request sent by a first local node; The connection request, the first node is assigned a first node FC-ID from the first FC-ID range; the FEF updates the first FEF according to the first node FC-ID a packet forwarding table. Specifically, the first FEF records, according to the first node FC-ID, a MAC address of the first local node in the packet forwarding table of the first FEF, and the first Corresponding relationship of a node FC-ID; further, an egress port corresponding to the first local node may also be recorded.

 The server 1 connected to the FEF1 in FIG. 1 is taken as an example. When the server 1 accesses the FEF1, a connection request is sent to the FEF1, for example, a FIP FLOGI (clothing login) message, and the connection request is received. FEF1 will select an unassigned FC-ID from the FC-ID range allocated by FCS1 to FEF1, and assign it to server 1. The FC-ID assigned to server 1 can be referred to as the node FC-ID. In addition to the node FC-ID, the server 1 has an identifier, which is a world wide name (WWN). The WWN is 64 bits, and its role is for identity and security control, similar to media access control ( Media access control, referred to as MAC), but not for forwarding addressing.

 Further, when the second FEF exits the FCoE network, the FCS also notifies the remaining FEFs in the FCoE network, so that the remaining FEFs update the message forwarding table. Optionally, the method further includes: the first FEF receiving the exit information sent by the FCS, where the exit information includes a MAC address of the second FEF, and the first FEF is updated according to the exit information. Deleting a packet forwarding table of the first FEF; specifically, the first FEF deletes, according to the exit information, a forwarding entry corresponding to the MAC address of the second FEF from the packet forwarding table of the first FEF .

 Optionally, before step 201, the first FEF establishes a connection with the FCS. Since the FCS is responsible for controlling and managing each FEF in the FCoE network, each newly joined FEF must first establish a connection with the FCS. Specifically, if the first FEF is newly added to the FCoE network, the first FEF first establishes a connection with the FCS, and then receives the first FC-ID range that the FCS allocates for the first FEF. Specifically, the first FEF may pass a transmission control protocol.

 (Transmission control protocol, abbreviated as: TCP) or a user datagram protocol (abbreviation: UDP) establishes a connection relationship with the FCS.

 This connection can be established in a variety of ways, including:

 Statically configuring the first FEF, where the first FEF establishes a connection with the FCS according to the address of the statically configured FCS; or

Obtaining, by the dynamic discovery mode, the address of the FCS, and establishing a connection with the FCS; specifically, sending, by the first FEF, a broadcast or a multicast packet, where the FCS responds; or, the FCS may periodically Sending a multicast/broadcast notification to the network to control the FEF it controls, the FCS Address or logo.

 It should be noted that the connection established between the first FEF and the FCS is usually a logical connection, and may of course be a physical direct connection.

As shown in Figure 1, the FEF1 new port is added to the FCoE network, and the FCS1^ ¹ other FEFs, such as FEF2, FEF3, and FEF4, are sent to FEF1, so that FEF1 updates the FEF1 message according to the forwarding information of FEF2, FEF3, and FEF4. Forwarding table. In addition, FEF1 receives the local node, that is, the connection request of Server 1, and FEF 1 assigns the node FC-ID to Server 1 from its FC-ID range, that is, 1.1.0/16, for example, 1.1.1, and then updates the FEF1 message. Forwarding table. Table 2 below shows the message forwarding table of FEF1:

 Table 2 FEF1 4 text forwarding table

参见上述的表 2所示， 第 2〜5行中的每一行都可以称为一条转发表项， 每个转发表项包括： 其他 FEF的 MAC地址、 FC-ID范围和出端口的对应关 系。 例如， 第 2行中的 "1.2.0/16 FEF2-MAC 1" , 其中的 1.2.0/16是

As shown in Table 2 above, each of the rows 2 to 5 can be referred to as a forwarding entry. Each forwarding entry includes: MAC address of another FEF, FC-ID range, and correspondence of outgoing ports. For example, in the second line, "1.2.0/16 FEF2-MAC 1", where 1.2.0/16 is

The FC-ID range of FEF2, FEF2-MAC is the MAC address of FEF2, and 1 is the outgoing port for outputting the message to FEF2. See circle 1 on FEF1 in Figure 1 for the outgoing port. The forwarding entry in line 5 of Table 2 is generated by FEF1 based on the node FC-ID assigned to the local node, Server1. The forwarding entry includes: the MAC address of the local node, the correspondence between the node FC-ID and the egress port, as shown in the fifth line in Table 2, "1.1.1 Serverl-MAC 100".

 Usually, the output port in the above packet forwarding table is learned by the FEF through the data plane address. If the out port is not learned, it can be empty. In this case, the sent packets will follow the usual Layer 2 forwarding technology and will be treated as unknown unicast, usually using a flooding method.

In addition, each of the FEFs in the network shown in FIG. 1 generates a respective forwarding table, for example, For the FEF2, when it is newly added to the network, it may receive the forwarding information about the FEF3 and the FEF4 sent by the FCS1. Therefore, the forwarding entry corresponding to FEF3 and FEF4 is generated in the packet forwarding table on the FEF2. After FEF1 is added to the network, FCS1 notifies FEF2 of the forwarding information of FEF1 (including the MAC address of FEF1 and its corresponding FC-ID range). FEF2 adds the forwarding entry corresponding to FEF 1 in its packet forwarding table. can. When there is no local node connected to the FEF2, the forwarding entry corresponding to the local node may not be available in the packet forwarding table.

 Further, each FEF can forward FCoE packets according to its own packet forwarding table. The method can also include:

 The first FEF receives the FCoE packet from the local node, and obtains the destination FC-ID from the FCoE packet;

 The first FEF searches the packet forwarding table of the first FEF according to the destination FC-ID, and obtains a MAC address of the destination FEF corresponding to the destination FC-ID;

 The first FEF sets the MAC address of the destination FEF to the destination MAC address of the FCoE packet, and sends the FCoE packet.

 For example, referring to FIG. 1, the server 1 sends an FCoE message to the memory 2 as an example to describe how each FEF in the network performs FCoE packet forwarding. The server 1 is a node connected to the FEF1, FEF1. The FC-ID assigned to this server 1 is 1.1.1, as shown in Table 2,

In the packet forwarding table of the FEF1, the forwarding entry includes "1.1.1 Server 1 - MAC 100; and the memory 2 is a node connected to the FEF4. It is assumed that the FC-ID allocated by the FEF4 to the memory 2 is 1.4.1; In the packet forwarding table on FEF4, its forwarding entries include

"1.4.1 - storge2-MAC - 200. First, the server 1 sends the FCoE message to the FEF1, and the FEF 1 obtains the destination MAC address from the FCoE message, and finds that the destination MAC address is the MAC address of the FEF1. The FEF1 decapsulates the FCoE packet, and obtains the destination FC-ID from the FCoE packet. The destination FC-ID is 1.4.1. The FEF1 is based on the destination FC-ID, that is, 1.4.1. The FEF1 packet forwarding table is searched for, and the destination MAC address corresponding to the destination FC-ID is FEF4-MAC, and the corresponding egress port is port 100, and FEF 1 uses FEF4-MAC as the FCoE packet. The MAC address is encapsulated, and then the FCoE packet is sent from the port 1. After receiving the FCoE packet, the FEF4 obtains the destination MAC address of the FCoE packet, and finds that the destination MAC address is the MAC address of the FEF4, and the FEF4 pair The FCoE packet is decapsulated, and the destination FC-ID is obtained from the FCoE packet as 1.4.1; and the FEF4 is configured according to the destination FC-ID. That is, 1.4.1, query the message forwarding table of FEF4, and obtain the destination MAC address corresponding to the destination FC-ID, namely 1.4.1, which is storge2-MAC, and the corresponding outgoing port is port 200, then FEF4 uses storge2- The MAC is used as the destination MAC address of the FCoE packet and is FCoE encapsulated. Then, the FEF4 sends the FCoE packet from the port 200 to the memory 2.

 Further, the FEF and the FCS may also pass through some or some of the FEFs, that is, the topology of the FEF cascading may be included in the networking, for example, as shown in FIG. 3, which is another FCoE network provided by the embodiment of the present invention. A schematic diagram of a networking architecture. FEF4 is connected to the FCS through FEF5. For example, the server 1 that accesses the FEF1 needs to communicate with the memory 2 that accesses the FEF4. Then, the FCoE message sent by the server 1 to the memory 2 needs to pass through the FEF5 to reach the FEF4, and the FEF5 can be called the intermediate FEF. When the FEF5 receives the FCoE packet, the FEF5 obtains the destination MAC address of the FCoE packet, and is the FEF4-MAC. The FEF5 determines that the destination MAC address of the FCoE packet is not the MAC address of the FEF5, and the FEF5 does not report the FCoE packet. The packet is decapsulated, and the FCoE packet is directly forwarded to FEF4.

 It should be noted that the first-level cascading structure shown in FIG. 3 is only used as an example. In the actual deployment, a two-level or multi-level cascading structure may be used. The present invention does not limit the number of cascading stages. In the method of implementing FCoE in this embodiment, when the first FEF is newly added to the FCoE network, the first FEF only needs to receive the FCS to allocate the FC-ID range for the first FEF, and receive forwarding information of other FEFs in the network, where the first The FEF can establish a packet forwarding table accordingly. For other FEFs, the FCS only needs to notify other FEFs of the first FC-ID range allocated to the first FEF newly added to the FCoE network. Other FEFs are added on the forwarding table. The forwarding entry corresponding to the first FEF can be used. Compared with the prior art, the generation of the packet forwarding table is simpler and faster, and the routing efficiency is improved, and the FCoE implementation is simplified.

 FIG. 4 is a flowchart of another method for implementing FCoE according to an embodiment of the present invention. The method for implementing the FCoE in this embodiment is performed by the FCS, and the embodiment is only briefly described. As shown in FIG. 4, the method may include:

 401. The FCS allocates a first FC-ID range to the first FEF.

 In this embodiment, when any FEF is newly added to the FCoE network, the FCS allocates an FC-ID range, and the FC-ID range obtained by each FEF is mainly used to allocate to the ENode accessing the FEF. The FC-ID is mainly used for addressing and forwarding FCoE messages in the network.

Specifically, when the first FEF is newly added to the FCoE network, the FCS allocates the first number to the first FEF. The FC-ID range is controlled by the first-FEF, and is used for the local node that accesses the first FEF, that is, the ENode that accesses the first FEF. Specifically, the first The FEF allocates an FC-ID from the first FC-ID range to the local node for each local node accessing the first FEF.

 The FC-ID range can be expressed in various ways, for example, by masking ( 1.1.0/16 ), or by n consecutive n (n > 0) from a certain FC-ID, for example, assuming An FC-ID range (1.1.10 10), which means that the FC-ID range is 10 consecutive from the FC-ID of 1.1.1, ie 1.1.1, 1.1.2, 1.1.3 1.1 .10. There may be other representations in the specific implementation, and are not limited to the above examples.

 In addition, assuming that there are multiple FEFs in the FCoE network, the FC-ID range allocated by the FCS for each FEF is different, that is, the FC-ID ranges of different FEFs cannot be overlapped. More specifically, when there is an FCS in the FCoE network, the FC-ID ranges allocated by the FEFs under the FCS cannot overlap; when there are multiple FCSs in the FCoE network, the FC-IDs controlled between different FCSs The range cannot overlap, that is, the FC-ID ranges assigned to the FEFs belonging to different FCSs cannot be overlapped, and the FC-ID range not overlapping means that the entire network is unique.

 There are a plurality of ways in which the FC-ID ranges are not overlapped. For details, refer to 201, which is not described here.

 The FCS sends the first FC-ID range and the forwarding information to the first FEF, so that the first FEF updates the packet forwarding table of the first FEF according to the forwarding information; The information includes: a correspondence between the MAC address of the other FEF and the FC-ID range allocated by the FCS to the other FEFs; the packet forwarding table includes a correspondence between the medium access control MAC address and the FC-ID range.

Further, when there are other FEFs other than the first FEF in the FCoE network, the FCS sends the forwarding information corresponding to the other FEFs to the first FEF; the forwarding information includes the MAC addresses of other FEFs and Corresponding relationship of the FC-ID range allocated by the FCS to the other FEFs; the forwarding information may be stored in the FCS, and the FCS sends the forwarding information to the first when the first FEF is newly added to the network. An FEF. The other FEF may be an FEF that joins the network before the first FEF, or may be an FEF that joins the network after the first FEF; for example, after the first FEF joins the FCoE network, if a new FEF joins the FCoE The forwarding information sent by the FCS received by the first FEF is the forwarding information newly added to the FEF. For example: Referring to Figure 1, when FCS1 learns that FEF1 is newly added to the FCoE network, it allocates the corresponding FC-ID range of FEF1 to "1.1.0/16" and other FEFs in the FCoE network (such as FEF2, FEF3, and FEF4). The corresponding forwarding information (such as the MAC address and the FC-ID range it is responsible for) is sent to the FEF1, so that the FEF1 generates a packet forwarding table according to the packet, and forwards the FCoE packet according to the packet forwarding table.

 In addition, the FCS also sends the first forwarding information corresponding to the first FEF that is newly added to the network to the other FEF, so that the other FEFs update the packet forwarding table of the other FEF according to the first forwarding information. The first forwarding information includes a correspondence between a MAC address of the first FEF and the first FC-ID range. In this way, each FEF in the entire network can know the FC-ID range that is responsible for each, and this notification is specifically handled by the FCS.

 For example, FCS1 in Figure 1 also notifies the FEF1's MAC address and its corresponding FC-ID range to other FEFs in the FCoE network, so that other FEFs update their own packet forwarding table, such as in the packet forwarding table. Add a forwarding entry corresponding to FEF1.

 Further, before 401, the FCS establishes a connection with the first FEF;

 For example, the first FEF newly joined to the FCoE network must establish a connection with the FCS responsible for the master. Since the FCS is responsible for controlling and managing each FEF in the FCoE network, each newly joined FEF must first establish a connection with the FCS. When the first FEF joins the FCoE network, it first establishes a connection with the FCS, and then receives the first FC-ID range allocated by the FCS for the first FEF.

 Specifically, a connection relationship may be established between the FCS and the first FEF by using TCP or UDP. The connection relationship may be established in a plurality of manners, including but not limited to: the first FEF is statically configured with an FCS address, and a connection is established with the FCS according to the FCS address; or the first FEF is passed. Dynamically discovering the address of the FCS, and establishing a connection with the FCS; for example, sending, by the first FEF, a broadcast or a multicast message, the FCS responding, or at the same time, the FCS may periodically send The multicast/broadcast notification informs the network of the FEF it controls, the address or identity of the FCS.

 It should be noted that the connection established between the first FEF and the FCS is usually a logical connection, and may of course be a physical direct connection.

In the method of implementing FCoE in this embodiment, when the first FEF is newly added to the FCoE network, the FCS allocates a first FC-ID range to the first FEF, and notifies the first FEF of the forwarding information of other FEFs in the stored network. The first FEF can establish a message forwarding table according to this, for other FEFs. In addition, the FCS only needs to notify other FEFs of the first FC-ID range of the first FEF that is newly added to the FCoE network, and the other FEFs may add the forwarding entry corresponding to the first FEF to the forwarding table. When there is a technology that needs to recalculate the entire network route, the generation of the packet forwarding table is simpler and faster, which improves the routing efficiency and simplifies the implementation of FCoE.

 Further, FIG. 5 is a flowchart of another method for implementing FCoE according to the embodiment of the present invention. On the basis of the method for implementing FCoE shown in FIG. 2, this embodiment details how to forward FCoE packets in the FCoE network of the present invention. , as shown in Figure 5, including:

 501. The first FEF receives an FCoE packet from the local node.

 Since the FCoE message is from the local node of the first "FEF", the first "FEF" is equivalent to the starting point FEF of the FCoE message.

 502. The first FEF obtains a destination FC-ID from the FCoE packet.

 The FCoE packet includes a source FC-ID, a destination FC-ID, a source MAC address, and a destination MAC address. The source FC-ID is the FC-ID of the local node of the first FEF, the source MAC address is the MAC address of the local node of the first FEF, and the destination FC-ID is the MAC address of the destination ENode, and the destination MAC address. The address is the MAC address of the first FEF.

 503. The first FEF searches the packet forwarding table of the first FEF according to the destination FC-ID, and obtains a MAC address of the destination FEF corresponding to the destination FC-ID.

 The starting FEF will find the destination FEF corresponding to the destination FC-ID according to the destination FC-ID. For example, taking FIG. 3 as an example, it is assumed that the destination node for message transmission is memory 2, and the destination FC-ID corresponding to the memory 2 is 1.4.1; the starting point FEF can be found according to the 1.4.1 memory 2 is under FEF4. Node, so the destination FEF is FEF4.

 504. The first FEF sets the MAC address of the destination FEF to the destination MAC address of the FCoE packet, and sends the FCoE packet.

In this embodiment, the starting point FEF has directly set the destination MAC address of the packet to the MAC address of the destination FEF, which is different from the prior art processing. For example, as shown in FIG. 3, when the server 1 When the memory 2 sends a packet, the FEF1 (the FCF device in the prior art) decapsulates the MAC address of the next hop FEF through the destination FC-ID, and sets the destination MAC address of the FCoE packet to FEF5. The MAC address is sent to the FEF5, and then the FEF5 is decapsulated to find the MAC address of the next hop FEF through the destination FC-ID, and the destination MAC address of the FCoE packet is set to the MAC address of the FEF4, and sent to the FEF4; , FCoE message from the entrance FEF, that is, FEF1 to the egress FEF, that is, each hop that the FEF4 passes, that is, each intermediate FEF (the FCF device in the prior art) needs to be decapsulated, routed using the next hop MAC address, and repackaged. Packet forwarding processing is costly and inefficient. In the embodiment of the present invention, at the ingress FEF, that is, the FEF1, the MAC address of the egress FEF can be found by the destination FC-ID, and the destination MAC address of the FCoE packet is directly set as the MAC address of the egress FEF, that is, The MAC address of the FEF4, the intermediate FEF only forwards the packet, does not modify the destination MAC address of the packet, and does not perform packet decapsulation and encapsulation processing. For example, at FEF5, only the destination MAC address in the packet, that is, FEF4-MAC, is required. The message can be directly output from the corresponding port 300, so that FCoE packets can be easily and quickly forwarded. Therefore, in the method for implementing the FCoE in the embodiment, the FEF improves the forwarding processing efficiency of the packet when the FCoE packet is forwarded compared with the prior art.

 In this embodiment, the network structure shown in FIG. 3 will be described, and the entire process from FEF1 being added to the network as the new device, the server 1 being added to the FEF1, and the server 1 transmitting the FCoE message to the memory 2 will be described. 6 is a schematic diagram of signaling interaction of a method for implementing FCoE according to an embodiment of the present invention. Since FEF2 and FEF3 are similarly processed, only FEF2 is shown in FIG. 6. FIG. 7 is a schematic diagram of packet forwarding according to a method for implementing FCoE according to an embodiment of the present invention.

 Specifically, as shown in FIG. 6, the signaling interaction process of the FCoE method in this embodiment is as follows:

 601, FEF1 establishes a connection with FCS1;

 For example, FEF1 is statically configured with the FCS1 address and establishes a connection with FCS1 accordingly.

 602. The FCS1 allocates an FC-ID range to the FEF1.

 For example, the FCS allocates an FC-ID range of 1.1.0/16 for FEF1.

 603. The FCS1 sends the FEF forwarding information to the FEF1.

 For example, the forwarding information of FCS1, FEF2, FEF4, and FEF5 is sent to FEF1, including: the correspondence between the MAC address of FEF2 and the FC-ID range of FEF2, and the correspondence between the MAC address of FEF3 and the FC-ID range of FEF3. Relationship, the correspondence between the MAC address of FEF4 and the FC-ID range of FEF4, and the correspondence between the MAC address of FEF5 and the FC-ID range of FEF5; the FC-ID range of FEF2, FEF3, FEF4, and FEF5 is also allocated by FCS .

 604. The FCS1 notifies the FEF1 of the forwarding information of the FEF1.

For example, FCS1 notifies FEF1, FEF3, FEF4, and FEF5 of FEF1's MAC address and corresponding FC-ID range, so that FEF2, FEF3, FEF4, and FEF5 update their respective documents. A forwarding table, that is, a forwarding entry corresponding to FEF1 is added to the respective packet forwarding table, including the correspondence between the MAC address of FEF1 and the FC-ID range of FEF1. In Fig. 6, only FCS1 is shown to notify the above information to FEF2, but actually FCS1 also notifies FEF3, FEF4 and FEF5.

 In addition, it should be noted that the implementation of this step may be various. For example, you can unicast each FEF or use a specific multicast address to multicast to all FEFs. The notification of the content can be incremental, that is, only the changed part is notified, or it can be completely typed, that is, all the information is delivered. Different implementation methods may be different in specific steps, which are not listed here. The purpose is to enable each FEF to dynamically obtain the forwarding information of the current network dynamically.

 It should be noted that the embodiment does not limit the execution order of 603 and 604. For example, the two steps may be performed separately or simultaneously.

 605. The FEF1 generates a forwarding entry of the packet forwarding table according to the forwarding information of the other FEFs. For the content of the specific forwarding entry, refer to the foregoing embodiment, and details are not described herein.

 606. The server 1 sends a FIP VLAN discovery to the FEF1.

 607, FEFl replies to the server 1 FIP FCF discovery;

 608. The server 1 sends a FIP FLOGI to the FEF1.

 Among them, 606~608, and the subsequent 609 are the normal processes when the server 1 newly establishes a connection with FEF1, and will not be described in detail. For example, any node needs to send a FIP FLOGI message when establishing a connection with FEF. It should be noted that the method for implementing FCoE in the embodiment of the present invention does not require any improvement at the node (ENode), so the flow of the execution of the server 1 is the same as the prior art.

 609. The FEF1 replies to the FIP FLOGI ACC to the server 1 and carries the FC-ID and other information delivered by the server 1;

 For example, FEF1 will allocate an FC-ID to server 1 from the FC-ID range that FCS1 allocates for itself, that is, 1.1.0/16, assuming that the assigned server 1 has an FC-ID of 1.1.1. FEF1 carries the FC-ID in the FIP FLOGI ACC and sends it to the server 1.

 610. The FEF1 generates a forwarding entry in the packet forwarding table according to the FC-ID allocated to the server 1.

 For example, the forwarding entry includes "1.1.1 - Server 1 -MAC - 100".

The above 601 to 610 perform the flow of the control plane and the management plane. The new node (server 1) and the new FEF (FEF1) are added to the network, and the FCS1 assigns the FC-ID to the new FEF1. The FEF1 allocates the FC-ID to the server 1 from the FC-ID range. The packet forwarding table is generated or updated on each FEF, and the FCoE packet is ready to be forwarded.

 Further, as shown in FIG. 7, after the completion of the control plane in FIG. 6, the FCoE packet is forwarded. Specifically, the packet processing process in which the server 1 sends the FCoE packet to the memory 2 in the FCoE network shown in FIG. As shown in the figure, the packet forwarding process of the FCoE method in this embodiment is as follows:

 701. The server 1 sends an FCoE packet to the FEF1.

 The format and mechanism of the FCoE packet sent by the server 1 are the same as those in the prior art, and are not described in detail. The packet includes: a source MAC address (that is, a MAC address of the server 1) and a destination MAC address (adjacent The FEF is the MAC address of FEF1), the source FC-ID (that is, the FC-ID of the server 1, 1.1.1), and the destination FC-ID (that is, the FC-ID of the memory 2, 1.4.1).

 FIG. 8 is a schematic diagram of packet changes in a method for implementing FCoE according to an embodiment of the present invention, and FIG. 8 shows a part of content of a header in a FCoE in a forwarding process.

 702. The FEF1 re-encapsulates the destination MAC address in the packet to the MAC address of the FEF4. When the FEF1 receives the packet sent by the server 1, the FEF1 strips the outer Ethernet encapsulation and obtains the destination FC-ID in the packet. , ie 1.4.1; FEF1 looks up its own message forwarding table, and knows that the FC-ID range including 1.4.1 is 1.4.0/16, and the MAC address corresponding to 1.40/16 is

FEF4-MAC, the purpose of the message is that the destination node corresponding to the FC-ID is located under FEF4. Therefore, FEF1 re-encapsulates the FCoE, sets the FEF4-MAC to the destination MAC address of the packet, and sets the source MAC address of the packet to the MAC address of FEF 1 itself.

 703. The FEF1 sends the packet to the FEF4.

 FEF1 sends the packet from port 1 to FEF4, and may pass other devices, including FEF5. As the destination MAC address of the packet is not FEF5, the FEF5 performs normal Layer 2 forwarding and continues to send the packet to FEF4. Taking Figure 3 as an example, FEF5 sends the message from port 300 to FEF4. Therefore, from Figure 8, the content of the message header does not change, that is, FEF5 does not make any changes to the message header.

704. The FEF4 changes the destination MAC address of the packet to the MAC address of the memory 2. When the packet reaches the FEF4, the FEF4 strips the outer Ethernet encapsulation and obtains the destination FC-ID address of the inner FC packet. Based on the destination FC-ID of the packet, FEF4 searches for the forwarding entry in its own packet forwarding table to obtain the MAC address and local output port corresponding to the FC-ID. Take In FIG. 3, for example, FEF4 will obtain the MAC address of the destination MAC which is storage2, and the local output port is port 200, that is, the output of the message from the port 200 will reach the memory 2.

 FEF4 re-encapsulates the FCoE and changes the source MAC address of the packet to its own MAC address, that is, FEF4-MAC, and changes the destination MAC address of the packet to the MAC address of the memory 2, that is, the memory 2-MAC. See Figure 8.

 705. The FEF4 sends the message to the memory 2.

 Among them, FEF4 sends the message to the memory 2 through the port 200. It should be noted that a network composed of multi-hop Ethernet switches can be used between FEF1 and FEF4 in this embodiment.

 Further, the FCS of the FCoE network may also send the following partition information, where the partition information is mainly used to notify a certain node in the FEF partition to allow communication with a certain node, for example, as shown in FIG. Zone X contains Serverl and storage2, then FCS will issue (1.1.1, 1.4.1, allowed) to FEF1 and FEF4 after the partition is activated. The FEF may generate an access control list (ACL) according to the partition information that is sent. The ACL is used to record the partition information, and is used to view the ACL when the FEF receives the packet. The transmitting node and the receiving node are the nodes that can communicate in the same partition recorded in the ACL, and then continue to forward the packet. Otherwise, the packet cannot be forwarded. For example, when FEF1 receives a packet from Server1 and looks up the packet forwarding table for forwarding, it needs to check whether the source and destination of the packet can be communicated according to the ACL generated by the partition information. Only when communication is allowed, Will be forwarded.

 FIG. 9 is a schematic structural diagram of an FEF according to an embodiment of the present invention, which is used to implement the method for implementing FCoE according to any embodiment of FIG. 2, FIG. 3, FIG. 6, and FIG. The foregoing method embodiments. As shown in FIG. 9, the first FEF includes: an information receiving unit 91 and a forwarding table updating unit 92;

 The information receiving unit 91 is configured to receive a first FC-ID range allocated by the FCS, and further receive, by the FCS, the forwarding information, where the forwarding information includes: a MAC address of another FEF, and the FCS is allocated to the other FEF Correspondence of the FC-ID range;

The forwarding table updating unit 92 is configured to update the packet forwarding table according to the forwarding information, where the packet forwarding table includes a correspondence between a medium access control MAC address and an FC-ID range; specifically, according to the forwarding information, Recording, in the packet forwarding table of the first FEF, a correspondence between a MAC address of the other FEF and an FC-ID range. The FCS allocates a first FC-ID range for the first FEF when the first FEF is newly added to the FCoE network; the first FC-ID range is controlled by the first FEF, and is used for allocation and connection. a local node that is in the first FEF, that is, an ENode that accesses the first FEF. Specifically, the first FEF is a local node that accesses the first FEF from the first FC-ID. In the range, an FC-ID is selected and assigned to the local node, and the FC-ID is mainly used for addressing and forwarding FCoE packets in the network. Optionally, the foregoing FC-ID range may be represented by multiple types, for example, by masking (1.1.0/16), or by consecutive n (from a certain FC-ID). 0).

 In addition, assuming that there are multiple FEFs in the FCoE network, the FCS allocates different FC-ID ranges for each FEF, that is, the FC-ID ranges of different FEFs cannot be overlapped; the FC-ID ranges do not overlap. It is the only one in the whole network. For details, see 201, and details are not described here.

 The first FEF, in addition to receiving the first FC-ID range allocated by the FCS, may further receive forwarding information sent by the FCS, where the forwarding information includes other The correspondence between the MAC address of the FEF and the FC-ID range assigned by the FCS to the other FEFs.

 It can be understood that the forwarding information received by the information receiving unit 91 may be the first

When the FEF is added to the FCoE network, the forwarding information of other FEFs that are stored in the FCS and existing in the FCoE network may also be the forwarding information of the new FEF in the FEF when the first FEF joins the FCoE network. .

 Further, FIG. 10 is another schematic structural diagram of an FEF according to an embodiment of the present invention. As shown in FIG. 10, on the basis of the structure of FIG. 9, the first FEF further includes a request receiving unit 93 and an address allocating unit 94;

 The request receiving unit 93 is configured to receive a connection request sent by the first local node;

 The address allocation unit 94 is configured to allocate, according to the connection request, the first FC-ID of the node from the first FC-ID range to the first local node;

 The forwarding table update unit 92 is further configured to update the packet forwarding table according to the first node FC-ID, where the FEF updates the first FEF according to the first node FC-ID The packet forwarding table specifically includes: the first FEF records, in the packet forwarding table of the first FEF, a correspondence between a MAC address of the first local node and a first node FC-ID.

Specifically, the correspondence between the MAC address of the first local node and the first node FC-ID is recorded in the packet forwarding table of the first FEF, and may further include the first local node. Corresponding outgoing port.

 Further, the information receiving unit 91 is further configured to receive the exit information sent by the FCS, where the exit information includes a MAC address of the second FEF;

 The forwarding table update unit 92 is further configured to: update the packet forwarding table of the first FEF according to the exit information; specifically, according to the exit information, from the packet forwarding table of the first FEF The forwarding entry corresponding to the MAC address of the second FEF is deleted.

 Further, the FEF may further include: a connection processing unit 95, configured to establish a connection with the FCS before the information receiving unit receives the first FC-ID range of the FCS allocation; specifically, according to the static configuration The address of the FCS is connected to the FCS; or the address of the FCS is obtained through a dynamic discovery manner, and a connection is established with the FCS.

 Among them, since the FCS is responsible for controlling and managing each FEF in the FCoE network, each FEF newly joined to the network must first establish a connection with the FCS. Specifically, if the first FEF is newly added to the FCoE network, the first FEF first establishes a connection with the FCS, and then receives the first FC-ID range allocated by the FCS for the first “FEF”. Specifically, the first FEF may establish a connection relationship with the FCS by using TCP or UDP. The connection relationship may be established in a plurality of manners, including but not limited to: statically configuring the first FEF, where the first FEF establishes a connection with the FCS according to the statically configured address of the FCS; Obtaining an address of the FCS and establishing a connection with the FCS. It should be noted that the connection established between the first FEF and the FCS is usually a logical connection relationship, and may also be a physical direct connection.

 In this embodiment, the first FEF obtains the FC-ID range allocated by the FCS, and other

For the forwarding information of the FEF, a packet forwarding table can be established. For other FEFs, only the first FC-ID range allocated by the FCS to the FEF is obtained, and the corresponding forwarding entry can be added to the forwarding table. Compared with the prior art, it is required to recalculate the entire network route, and the generation of the packet forwarding table is simpler and faster, which improves the routing efficiency and simplifies the implementation of FCoE.

 FIG. 11 is still another schematic structural diagram of an FEF according to an embodiment of the present invention, which is shown in FIG. 9 or 10.

On the basis of the FEF, the first FEF may further include: a message forwarding unit 96; the message forwarding unit 96 is configured to receive an FCoE message from the local node, and obtain the destination FC from the FCoE message. -ID; searching the packet forwarding table to obtain a MAC address of the destination FEF corresponding to the FC-ID according to the destination FC-ID; setting the MAC address of the destination FEF as the destination of the FCoE packet MAC address; and sending the FCoE message; Specifically, FIG. 11 is based on the FEF structure shown in FIG. 10. As shown in FIG. 11, the packet forwarding unit 96 may specifically include: an address obtaining subunit 961, a destination finding subunit 962, and a message processor. Unit 963 and message transmitting subunit 964; wherein

 The address obtaining subunit 961 is configured to receive an FCoE packet from the local node, and obtain a destination FC-ID from the FCoE packet.

 Since the FCoE message is from the local node of the first "FEF", the first "FEF" is equivalent to the starting point FEF of the FCoE message.

 The destination search sub-unit 962 is configured to search, according to the destination FC-ID, the MAC address of the destination FEF corresponding to the FC-ID by using the "3⁄4 text forwarding table";

 The FCoE packet includes a source FC-ID, a destination FC-ID, a source MAC address, and a destination MAC address. The source FC-ID is the FC-ID of the local node of the first FEF, the source MAC address is the MAC address of the local node of the first FEF, and the destination FC-ID is the MAC address of the destination ENode, and the destination MAC address. The address is the MAC address of the first FEF.

 The text processing sub-unit 963 is configured to set a MAC address of the destination FEF as a destination MAC address of the FCoE packet;

 The message sending subunit 964 is configured to send the FCoE message.

In this embodiment, the starting point FEF has directly set the destination MAC address of the packet to the MAC address of the destination FEF, which is different from the prior art processing. For example, as shown in FIG. 3, when the server 1 When the memory 2 sends a packet, the FEF1 (the FCF device in the prior art) decapsulates the MAC address of the next hop FEF through the destination FC-ID, and sets the destination MAC address of the FCoE packet to FEF5. The MAC address is sent to the FEF5, and then the FEF5 is decapsulated to find the MAC address of the next hop FEF through the destination FC-ID, and the destination MAC address of the FCoE packet is set to the MAC address of the FEF4, and sent to the FEF4; The FCoE packet needs to be decapsulated from the ingress FEF, that is, the FEF1 to the egress FEF, that is, each hop that the FEF4 passes, that is, each intermediate FEF (the FCF device in the prior art), using the next hop MAC address. Routing, re-encapsulation process, packet forwarding processing is costly and inefficient. In the embodiment of the present invention, at the ingress FEF, that is, the FEF1, the MAC address of the egress FEF can be found by the destination FC-ID, and the destination MAC address of the FCoE packet is directly set as the MAC address of the egress FEF, that is, The MAC address of the FEF4, the intermediate FEF is only forwarded, and the destination MAC address of the packet is not modified. The decapsulation and encapsulation processing of the packet is not performed. For example, in the FEF5, only the destination MAC address in the packet is required. FEF4-MAC directly outputs the packets from the corresponding port 300, so that FCoE packets can be easily and quickly forwarded.

 In this embodiment, the first FEF can obtain the packet forwarding table by acquiring the first FC-ID range allocated by the FCS and the forwarding information of other FEFs in the network. For other FEFs, only the FCS is required to be allocated to the new join. In the first FC-ID range of the first FEF of the FCoE network, the corresponding forwarding entry can be added to the packet forwarding table. Compared with the prior art, the control plane of the FCoE network is implemented more. Fast and simple, reducing the implementation complexity of the control plane. In addition, in the embodiment of the present invention, the inbound FEF of the FCoE packet finds the MAC address of the egress FEF according to the destination FC-ID, and directly sets the destination MAC address of the FCoE packet, and the intermediate FEF only forwards the FCoE packet. The packet forwarding processing of the FEF in this embodiment is low in cost and high in efficiency, and is not required to perform de-encapsulation and encapsulation processing on the FCoE packet.

 FIG. 12 is a schematic structural diagram of an FCS according to an embodiment of the present invention. The method for implementing FCoE according to the embodiment of FIG. 4 is implemented. For the working principle, refer to the foregoing method embodiment. As shown in Figure 12, the FCS includes:

 The address allocation unit 1201 is configured to allocate a first FC-ID range to the first FEF.

 The information sending unit 1202 is configured to send the first FC-ID range and the forwarding information to the first FEF, so that the first FEF updates the packet forwarding of the first FEF according to the forwarding information. The forwarding information includes: a MAC address of another FEF and a correspondence between the FCS and the FC-ID range allocated by the FCS to the other FEF, where the forwarding table includes a MAC address and an FC-ID range. Correspondence relationship. And updating, by the first FEF, the packet forwarding table of the first FEF according to the forwarding information, specifically: the first FEF, according to the forwarding information, in a forwarding table of the first FEF Corresponding relationships between the MAC addresses of the other FEFs and the FC-ID ranges of the other FEFs are recorded.

The FCS allocates a first FC-ID range for the first FEF when the first FEF is newly added to the FCoE network; the first FC-ID range is controlled by the first FEF, and is used for allocation and connection. a local node that is in the first FEF, that is, an ENode that accesses the first FEF. Specifically, the first FEF is a local node that accesses the first FEF from the first FC-ID. In the range, an FC-ID is selected and assigned to the local node, and the FC-ID is mainly used for addressing and forwarding FCoE packets in the network. Optionally, the foregoing FC-ID range may be represented by multiple types, for example, to cover The way of the code indicates (1.1.0/16), or consecutive η (η>0) from a certain FC-ID. In addition, assuming that there are multiple FEFs in the FCoE network, the FCS allocates different FC-ID ranges for each FEF, that is, the FC-ID ranges of different FEFs cannot be overlapped; the FC-ID ranges do not overlap. It is the only one in the whole network. For details, see 201, and details are not described here.

 The first FEF, in addition to receiving the first FC-ID range allocated by the FCS, may receive the forwarding information sent by the FCS, where the forwarding information includes The correspondence between the MAC address of the other FEF and the FC-ID range assigned by the FCS to the other FEFs. It can be understood that the forwarding information received by the information receiving unit 91 may be forwarding information of other FEFs existing in the FCoE network that are saved by the FCS when the first FEF is newly added to the FCoE network, and may also be the foregoing After an FEF joins the FCoE network, when a new FEF joins, the forwarding information of the new FEF is added.

 Further, the information sending unit 1202 is further configured to send the first forwarding information corresponding to the first FEF to the other FEF, so that the other FEFs update the other FEF according to the first forwarding information. The packet forwarding table includes: a correspondence between a MAC address of the first FEF and the first FC-ID range. Specifically, after the address allocation unit 1201 allocates the corresponding first Fibre Channel identifier FC-ID range to the first FEF, the information sending unit 1202 sends the first forwarding information corresponding to the first FEF to Said other FEF.

 When the correspondence between the MAC address of each FEF recorded by the FCS and the FC-ID range changes, the FCS will also notify all FEFs in the network. Specifically, when the first FEF is newly added to the FCoE network, the FCS sends the first forwarding information corresponding to the first FEF to other FEFs other than the first FEF in the FCoE network, so that other FEFs are according to the foregoing The forwarding information of the other FEF is updated by the forwarding information, and the first forwarding information includes a correspondence between the MAC address of the first FEF and the first FC-ID range. In this way, each FEF in the entire network can know the FC-ID range that is responsible for each, and this notification is specifically handled by the FCS.

Further, the information sending unit 1202 is further configured to send the exit information to the first FEF, where the exit information includes a MAC address of the second FEF, so that the first FEF updates the location according to the exit information. A packet forwarding table of the first FEF. The first FEF, according to the exit information, updating the packet forwarding table of the first FEF, specifically: the first FEF deleting the second packet from the packet forwarding table of the first FEF The forwarding entry corresponding to the MAC address of the FEF. The FCS of the embodiment sends a corresponding first FC-ID range to the first FEF, and sends forwarding information of other FEFs in the network to the first FEF, so that the first FEF can update the forwarding table according to the first FEF. The corresponding forwarding entry is added to the packet forwarding table. Compared with the prior art, the control plane of the FCoE network is faster and simpler, and the implementation complexity of the control plane is reduced.

 FIG. 13 is a schematic structural diagram of a hardware structure of an FEF according to an embodiment of the present invention. The method for implementing FCoE according to any of the embodiments of FIG. 2, FIG. 3, FIG. 6 and FIG. 8 is implemented. The foregoing method embodiments. As shown in FIG. 13, the first FEF may include a memory 1301, a processor 1302, a bus 1303, and a communication interface 1304. The processor 1302 and the memory 1301 and the communication interface 1304 are connected by a bus 1303 and complete communication with each other. The processor 1302 may be a single-core or multi-core central processing unit (CPU), or an application specific integrated circuit (ASIC), or one configured to implement an embodiment of the present invention. Or multiple integrated circuits. The memory 1301 may be a high speed RAM memory or a non-volatile memory such as at least one disk memory.

 The memory 1301 is configured to store program code, where the program code includes a computer operation instruction;

 The processor 1302 executes a computer operation instruction to: receive, by using the communication interface 1304, a first FC-ID range allocated by the FCS and forwarding information sent by the FCS, where the forwarding information includes: a MAC address of another FEF and a Determining, by the FCS, the correspondence between the FC-ID ranges of the other FEFs, and updating the packet forwarding table of the first FEF according to the forwarding information, where the packet forwarding table includes a MAC address and an FC-ID range. Correspondence relationship;

 The processor 1302 is specifically configured to record, according to the forwarding information, a correspondence between a MAC address of the other FEF and an FC-ID range in a packet forwarding table of the first FEF. The memory 1301 may be further configured to store a packet forwarding table of the first FEF, as shown in Table 2.

The FCS allocates a first FC-ID range for the first FEF when the first FEF is newly added to the FCoE network; the first FC-ID range is controlled by the first FEF, and is used for allocation and connection. a local node that is in the first FEF, that is, an ENode that accesses the first FEF. Specifically, the first FEF is a local node that accesses the first FEF from the first FC-ID. Select an FC-ID to be assigned to the local node in the range, and the FC-ID is mainly used for FCoE packets in the network. Address forwarding. Optionally, the foregoing FC-ID range may be represented by multiple types, for example, by masking (1.1.0/16), or by consecutive n (from a certain FC-ID). 0).

 In addition, assuming that there are multiple FEFs in the FCoE network, the FCS allocates different FC-ID ranges for each FEF, that is, the FC-ID ranges of different FEFs cannot be overlapped; the FC-ID ranges do not overlap. It is the only one in the whole network. For details, see 201, and details are not described here.

 The first FEF may receive the forwarding information sent by the FCS, where the forwarding information includes the other ones, in the FCoE network, the first FEF, in addition to receiving the first FC-ID range allocated by the FCS. The MAC address of the FEF and the correspondence between the FCS and the FC-ID range assigned by the FCS to the other FEFs. It can be understood that the forwarding information received by the information receiving unit 91 may be forwarding information of other FEFs existing in the FCoE network that are saved by the FCS when the first FEF is newly added to the FCoE network, and may also be the foregoing After a FEF joins the FCoE network, when a new FEF joins, the new FEF forwards information.

 In addition, when the correspondence between the MAC address of each FEF recorded by the FCS and the FC-ID range changes, the FCS will also notify all FEFs in the network. Specifically, when the first FEF is newly added to the FCoE network, the FCS sends the first forwarding information corresponding to the first FEF to other FEFs other than the first FEF in the FCoE network, so that other FEFs are The first forwarding information is used to update the forwarding table of the other FEFs, where the first forwarding information includes a correspondence between a MAC address of the first FEF and the first FC-ID range. In this way, each FEF in the entire network can know the FC-ID range that is responsible for each, and this notification is specifically handled by the FCS.

 Further, the processor 1302 is further configured to receive, by using the communication interface 1304, a connection request sent by the first local node, and allocate, according to the connection request, the first local node from the first FC-ID range. The node FC-ID is updated according to the first node FC-ID, and the processor 1302 is configured to record the first local in the packet forwarding table of the first FEF. Correspondence between the MAC address of the node and the first node FC-ID.

 Further, the processor 1302 is further configured to receive, by using the communication interface 1304, the exit information sent by the FCS, where the exit information includes a MAC address of the second FEF, and update the report of the first FEF according to the exit information. The forwarding table of the first FEF is updated according to the exit information, and the first FEF is deleted from the packet forwarding table of the first FEF. The forwarding entry corresponding to the MAC address of the second FEF.

Further, the processor 1302 is further configured to acquire the first FC-ID range allocated by the FCS The connection to the FCS is established through the communication interface 1304. The method includes: establishing a connection with the FCS according to the statically configured address of the FCS; or acquiring the address of the FCS by using a dynamic discovery manner, and the FCS establish connection.

 Among them, since the FCS is responsible for controlling and managing each FEF in the FCoE network, each FEF newly joined to the network must first establish a connection with the FCS. Specifically, if the first FEF is newly added to the FCoE network, the first FEF first establishes a connection with the FCS, and then receives the first FC-ID range allocated by the FCS for the first “FEF”. Specifically, the first FEF may establish a connection relationship with the FCS by using TCP or UDP. The connection relationship may be established in a plurality of manners, including but not limited to: statically configuring the first FEF, where the first FEF establishes a connection with the FCS according to the statically configured address of the FCS; Obtaining an address of the FCS and establishing a connection with the FCS. It should be noted that the connection established between the first FEF and the FCS is usually a logical connection relationship, and may also be a physical direct connection.

 Further, the processor 1302 is further configured to: receive, by the communication interface 1304, an FCoE packet from the local node, and obtain a destination FC-ID from the FCoE packet; and find the 4艮 according to the destination FC-ID. The forwarding table obtains the MAC address of the destination FEF corresponding to the destination FC-ID; sets the MAC address of the destination FEF to the destination MAC address of the FCoE packet, and sends the FCoE packet.

 In addition, the processor 1302 can be used to execute the computer instructions in addition to the above-described method flow, so as to implement other steps in the method embodiment of the present invention, and details are not described herein again.

 In this embodiment, the first FEF can obtain the packet forwarding table by acquiring the first FC-ID range allocated by the FCS and the forwarding information of other FEFs in the network. For other FEFs, only the FCS is required to be allocated to the new join. In the first FC-ID range of the first FEF of the FCoE network, the corresponding forwarding entry can be added to the packet forwarding table. Compared with the prior art, the control plane of the FCoE network is implemented more. Fast and simple, reducing the implementation complexity of the control plane. In addition, in the embodiment of the present invention, the inbound FEF of the FCoE packet finds the MAC address of the egress FEF according to the destination FC-ID, and directly sets the destination MAC address of the FCoE packet, and the intermediate FEF only forwards the FCoE packet. The packet forwarding processing of the FEF in this embodiment is low in cost and high in efficiency, and is not required to perform de-encapsulation and encapsulation processing on the FCoE packet.

FIG. 14 is a schematic structural diagram of hardware of an FCS according to an embodiment of the present disclosure, where The method for implementing FCoE shown in FIG. 4, the working principle of which is described in the foregoing method embodiment. As shown in FIG. 14, the FCS may include: a memory 1401, a processor 1402, a bus 1403, and a communication interface 1404. The processor 1402 and the memory 1301 and the communication interface 1404 are connected by a bus 1403 and complete communication with each other. The processor 1402 may be a single-core or multi-core central processing unit (CPU), or an application specific integrated circuit (ASIC), or a one configured to implement an embodiment of the present invention. Or multiple integrated circuits. The memory 1401 may be a high speed RAM memory or a non-volatile memory such as at least one disk storage device.

 The memory 1401 is configured to store program code, where the program code includes a computer operation instruction;

 The processor 1402 executes computer operation instructions to: allocate a first FC-ID range for the first FEF, and send the first FC-ID range and forwarding information to the first FEF via the communication interface 1404 And the packet forwarding table of the first FEF is updated by the first FEF, where the forwarding information includes: a MAC address of another FEF and a corresponding relationship between the FCS and the FC-ID range allocated by the FCS to the other FEFs. The corresponding forwarding table includes a correspondence between a MAC address and an FC-ID range.

 The memory 1401 is further configured to store the first FC-ID range.

 The first FEF updates the packet forwarding table of the first FEF according to the forwarding information, specifically: the first FEF records the other FEF in the first FEF forwarding table. Correspondence between the MAC address and the FC-ID range.

 The FCS allocates a first FC-ID range for the first FEF when the first FEF is newly added to the FCoE network; the first FC-ID range is controlled by the first FEF, and is used for allocation and connection. a local node that is in the first FEF, that is, an ENode that accesses the first FEF. Specifically, the first FEF is a local node that accesses the first FEF from the first FC-ID. In the range, an FC-ID is selected and assigned to the local node, and the FC-ID is mainly used for addressing and forwarding FCoE packets in the network. Optionally, the foregoing FC-ID range may be represented by multiple types, for example, by masking (1.1.0/16), or by consecutive n (from a certain FC-ID). 0).

In addition, assuming that there are multiple FEFs in the FCoE network, the FCS allocates different FC-ID ranges for each FEF, that is, the FC-ID ranges of different FEFs cannot be overlapped; the FC-ID ranges do not overlap. It is the only one in the whole network. For details, see 201, and details are not described here. The first FEF, in addition to receiving the first FC-ID range allocated by the FCS, may receive the forwarding information sent by the FCS, where the forwarding information includes The correspondence between the MAC address of the other FEF and the FC-ID range assigned by the FCS to the other FEFs. It can be understood that the forwarding information received by the information receiving unit 91 may be forwarding information of other FEFs existing in the FCoE network that are saved by the FCS when the first FEF is newly added to the FCoE network, and may also be the foregoing After an FEF joins the FCoE network, when a new FEF joins, the forwarding information of the new FEF is added.

 Further, the processor 1402 is further configured to send, by using the communication interface 1404, the first forwarding information corresponding to the first FEF to the other FEF, so that the other FEF updates the other according to the first forwarding information. ? The first forwarding information includes: a correspondence between a MAC address of the first FEF and the first FC-ID range.

 When the correspondence between the MAC address of each FEF recorded by the FCS and the FC-ID range changes, the FCS will also notify all FEFs in the network. Specifically, when the first FEF is newly added to the FCoE network, the FCS sends the first forwarding information corresponding to the first FEF to other FEFs other than the first FEF in the FCoE network, so that other FEFs are according to the foregoing The forwarding information of the other FEF is updated by the forwarding information, and the first forwarding information includes a correspondence between the MAC address of the first FEF and the first FC-ID range. In this way, each FEF in the entire network can know the FC-ID range that is responsible for each, and this notification is specifically handled by the FCS.

 Further, the processor 1402 is further configured to send the exit information to the first FEF, where the exit information includes a MAC address of the second FEF, so that the first FEF updates the first FEF according to the exit information. The packet forwarding table, wherein the first FEF updates the packet forwarding table of the first FEF according to the exit information, specifically: the packet forwarding table of the first FEF from the first FEF The forwarding entry corresponding to the MAC address of the second FEF is deleted.

 In addition, the processor 1402 can be used to execute the computer instructions in addition to the above-described method flow, so as to implement other steps in the method embodiment of the present invention, and details are not described herein again.

The FCS of the present embodiment, by allocating the first FC-ID range to the first FEF, and transmitting the forwarding information of other FEFs in the network to the first FEF, so that the first FEF can update the text forwarding table according to the The corresponding forwarding entry is added to the forwarding table. Compared with the prior art, the control plane of the FCoE network is faster and simpler, and the implementation complexity of the control plane is reduced. An embodiment of the present invention further provides a system for implementing Ethernet Fibre Channel FCoE, where the system includes two functional entities, FCS and FEF. The FCS participates in the operation of the control plane and the management plane of the FCoE network, and does not participate in the forwarding of the data plane of the FCoE network. It is used to centrally control and manage the FEF, and send relevant signaling and parameters to the FEF. The FEF is used to receive the control and management of the FCS, and receive the relevant signaling and parameters from the FCS. It is also used to generate a packet forwarding table. The FCoE packet is forwarded according to the packet forwarding table, that is, the data plane of the FCoE network is completed. Forward.

 Specifically, the function and structure of the FEF are as described in any of the embodiments of Figs. 9 to 11 and Fig. 13 of the present invention; the function and structure of the FCS are as described in any of Figs. 12 and 14 of the present invention. The FCS controls and manages the FEF, for example, to form Figure 1 or Figure 3 or other networking forms. The structure and working principle of the FEF and the FCS can be combined with reference to the method embodiment and the device embodiment of the present invention, and details are not described herein again. As shown in FIG. 1 , the first FEF newly added to the FCoE network establishes a connection with the FCS1 responsible for the main control, and obtains the first FC-ID range delivered by the FCS1, and also receives the forwarding information delivered by the FCS1, and the forwarding is performed. The information includes: a MAC address of another FEF and a correspondence between the FCS and the FC-ID range allocated by the FCS; and the message forwarding table used for forwarding the message according to the forwarding information, specifically: The first FEF records the correspondence between the MAC address of the other FEF and the FC-ID range in the packet forwarding table of the first FEF; and then forwards the FCoE packet according to the forwarding table. The FCS1 may also send the first forwarding information corresponding to the first FEF to other FEFs other than the first FEF in the FCoE network, so that other FEFs update their message forwarding table according to the information, for example, adding the newly added ― A forwarding entry corresponding to the FEF, where the forwarding entry includes a correspondence between a MAC address of the first FEF and a first FC-ID range.

The FCoE is implemented by the system. The first FEF can obtain the packet forwarding table by obtaining the FC-ID range allocated by the FCS and the forwarding information of other FEFs in the network. For other FEFs, only the FCS is required to be added to the new join. The first FC-ID range of the first FEF of the FCoE network can add a corresponding forwarding entry in the packet forwarding table. Compared with the prior art, the control plane of the FCoE network is implemented more. It is fast and simple, and reduces the implementation complexity of the control plane. In addition, in the embodiment of the present invention, the FEF of the FCoE packet finds the MAC address of the egress FEF according to the destination FC-ID, and directly sets the FCoE packet as the destination. The MAC address, the intermediate FEF forwards the FCoE packet only, and does not perform decapsulation and encapsulation processing. Compared with the prior art, the FCoE packet is hop-by-hop, decapsulated, and encapsulated. Text forwarding processing is costly and efficient. A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The method includes the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

 Finally, it should be noted that the above embodiments are only for explaining the technical solutions of the present invention, and are not intended to be limiting thereof; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and the modifications or substitutions do not deviate from the technical solutions of the embodiments of the present invention. range.

Claims

claims 1. A method for implementing Fiber Channel over Ethernet FCoE, applied to FCoE networks, characterized in that the FCoE network includes a Fiber Channel control plane server FCS and at least one Fiber Channel edge forwarder FEF, and the at least one FEF includes The first FEF; the method includes: the first FEF receives the first fiber channel identification FC-ID range allocated by the FCS; and also receives the forwarding information sent by the FCS, and the forwarding information includes: other FEF's The corresponding relationship between the media access control MAC address and the FC-ID range assigned by the FCS to the other FEF; The first FEF updates the message forwarding table of the first FEF based on the forwarding information, and the first FEF updates the message forwarding table of the first FEF. The text forwarding table includes the corresponding relationship between the MAC address and the FC-ID range; Wherein, the first FEF updates the message forwarding table of the first FEF according to the forwarding information, which specifically includes: the first FEF updates the message forwarding table of the first FEF according to the forwarding information. The corresponding relationship between the MAC addresses and FC-ID ranges of other FEFs is recorded in . 2. The method according to claim 1, characterized in that, after the first FEF obtains the first FC-ID range allocated by the FCS, the method further includes: The first FEF receives the connection request sent by the first local node; The -th FEF allocates the first node FC-ID to the -th local node from the -th FC-ID range according to the connection request; The FEF updates the text forwarding table of the first FEF based on the first node FC-ID; wherein, the FEF updates the text forwarding table of the first FEF based on the first node FC-ID. The text forwarding table specifically includes: the first FEF records the MAC address of the first local node and the first local node in the text forwarding table of the first FEF according to the first node FC-ID. Correspondence of node FC-ID. 3. The method according to claim 1 or 2, characterized in that the method further includes: the first FEF receiving the exit information sent by the FCS, and the exit information includes the MAC address of the second FEF; The first FEF updates the message forwarding table of the first FEF according to the exit information; wherein the first FEF updates the message forwarding table of the first FEF according to the exit information, specifically including: : The first FEF deletes the forwarding table entry corresponding to the MAC address of the second FEF from the packet forwarding table of the first FEF according to the exit information. 4. The method according to any one of claims 1 to 3, characterized in that, in the first FEF Before obtaining the first fiber channel identification FC-ID range allocated by the FCS, the method further includes: establishing a connection between the first FEF and the FCS; establishing a connection between the first FEF and the FCS specifically includes: The first FEF is statically configured with an FCS address, and establishes a connection with the FCS according to the FCS address; or, The first FEF obtains the FCS address through dynamic discovery and establishes a connection with the FCS. 5. The method according to any one of claims 1 to 4, characterized in that the method further includes: the -th FEF receiving an FCoE message from the local node; Obtain the destination FC-ID from the FCoE message; The first FEF searches the message forwarding table of the first FEF according to the destination FC-ID, and obtains the MAC address of the destination FEF corresponding to the destination FC-ID; The first FEF sets the MAC address of the destination FEF as the destination MAC address of the FCoE message, and sends the FCoE message. 6. A method for implementing Fiber Channel over Ethernet FCoE, applied to FCoE networks, characterized in that the FCoE network includes a Fiber Channel control plane server FCS and at least one Fiber Channel edge repeater FEF, and the at least one FEF includes the first FEF; the method includes: the FCS allocates a first fiber channel identification FC-ID range to the first FEF; the FCS sends the first FC-ID range and forwarding information to the first FEF. A FEF, so that the first FEF updates the packet forwarding table of the first FEF according to the forwarding information; the forwarding information includes: the media access control MAC addresses of other FEFs and the FCS assigned to the Correspondence between FC-ID ranges of other FEFs; The text forwarding table includes a correspondence between MAC addresses and FC-ID ranges. 7. The method according to claim 6, wherein after the FCS allocates the first FC-ID range to the first FEF, the method further includes: The FCS sends the first forwarding information corresponding to the first FEF to the other FEFs, so that the other FEFs update the message forwarding tables of the other FEFs according to the first forwarding information, and the first The forwarding information includes: the correspondence relationship between the MAC address of the -th FEF and the -th FC-ID range. 8. The method according to claim 6 or 7, characterized in that, the method further includes: The FCS sends exit information to the first FEF, where the exit information includes the MAC address of the second FEF, so that the first FEF updates the packet forwarding table of the first FEF according to the exit information. 9. A fiber channel edge repeater FEF, which is characterized by including: An information receiving unit, configured to receive the first fiber channel identification FC-ID range assigned by the fiber channel control plane server FCS; and also receive the forwarding information sent by the FCS, where the forwarding information includes: the media access control MAC address of other FEFs and the corresponding relationship between the FC-ID ranges assigned by the FCS to the other FEFs; A forwarding table update unit, configured to update the message forwarding table of the FEF according to the forwarding information. The message forwarding table includes the correspondence between the MAC address and the FC-ID range; specifically used to update the message forwarding table according to the forwarding information. , record the corresponding relationship between the MAC address and the FC-ID range of the other FEF in the message forwarding table of the FEF. 10. The FEF according to claim 9, further comprising: A request receiving unit, configured to receive a connection request sent by the first local node; An address allocation unit, configured to allocate a first node FC-ID from the first FC-ID range to the first local node according to the connection request; The forwarding table update unit is also used to update the message forwarding table of the FEF according to the first node FC-ID; specifically used to update the message forwarding table of the FEF according to the first node FC-ID. The forwarding table records the corresponding relationship between the MAC address of the first local node and the FC-ID of the first node. 11. The FEF according to claim 9 or 10, characterized in that, The information receiving unit is also configured to receive exit information sent by the FCS, where the exit information includes the MAC address of the second FEF; The forwarding table update unit is also configured to update the message forwarding table of the first FEF based on the exit information; specifically, delete the message forwarding table of the first FEF based on the exit information. The forwarding table entry corresponding to the MAC address of the second FEF. 12. The FEF according to any one of claims 9 to 11, further comprising: a connection processing unit, used to establish a connection with the FCS; specifically used to establish a connection with the FCS according to a statically configured FCS address. ; Or, obtain the FCS address through dynamic discovery and establish a connection with the FCS. 13. The FEF according to any one of claims 9 to 12, further comprising: message transfer The sending unit is used to receive the FCoE message from the local node and obtain the destination FC-ID from the FCoE message; according to the destination FC-ID, search the message forwarding table of the FEF to obtain the message forwarding table of the FEF. The MAC address of the destination FEF corresponding to the destination FC-ID; Set the MAC address of the destination FEF as the destination MAC address of the FCoE message, and send the FCoE message. 14. A fiber channel control plane server FCS, characterized by including: An address allocation unit, configured to allocate a first fiber channel identification FC-ID range to the first FEF; an information sending unit, configured to send the first FC-ID range and forwarding information to the first FEF, so that The first FEF updates the packet forwarding table of the first FEF according to the forwarding information. The forwarding information includes: the media access control MAC address of other FEFs and the FC- assigned by the FCS to the other FEFs. The corresponding relationship between the ID range and the text forwarding table includes the corresponding relationship between the MAC address and the FC-ID range. 15. The FCS according to claim 14, characterized in that, The information sending unit is also configured to send the first forwarding information corresponding to the first FEF to the other FEFs, so that the other FEFs update the message forwarding of the other FEFs according to the first forwarding information. Posted, the first forwarding information includes: a correspondence relationship between the MAC address of the first FEF and the first FC-ID range. 16. The FCS according to claim 14 or 15, characterized in that, The information sending unit is also configured to send exit information to the first FEF, where the exit information includes the MAC address of the second FEF, so that the first FEF updates the -th FEF according to the exit information. message forwarding table. 17. A system for implementing Fiber Channel over Ethernet FCoE, characterized in that it includes at least one Fiber Channel edge repeater FEF according to any one of claims 9 to 13, and a fiber channel edge forwarder FEF according to any one of claims 14 to 16. Fiber Channel control plane server FCS.