WO2007009352A1

WO2007009352A1 - A v-switch access method, apparatus and v-switch system

Info

Publication number: WO2007009352A1
Application number: PCT/CN2006/001654
Authority: WO
Inventors: Lingyuan Fan; Yuepeng Chen; Dengchao Wu
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2005-07-15
Filing date: 2006-07-12
Publication date: 2007-01-25

Abstract

The core idea of a new V-SWITCH access method, apparatus and V-SWITCH system in the V-SWITCH technology is that a V-SWITCH access method based on a variety of policies (including port, MAC address, IP address and so on) is supported on the edge DRE, by configuring the policies on the edge ingress DRE to establish the mapping relationship from the user data packets/frames to the V-SWITCH channels, which mainly includes the mapping relationship from port to V-SWITCH channel, from IP layer address information to V-SWITCH channel, or from Ethernet layer address information to V-SWITCH channel, when the packets from user come into the V-SWITCH domain, the ingress DRE selects a suitable V-SWITCH channel for the packets according to the policies above to transmit the traffic stream. The present invention consummates the architecture of the V-SWITCH system, enriches the V-SWITCH access method, thereby enhances the access capacity of the V-SWITCH technology, expands the application area of the V-SWITCH technology, and reduces the cost of establishing network.

Description

The present invention relates to the field of network communication technologies, and in particular, to a virtual switching access method, apparatus, and virtual switching system. Background technique

As the scale of the Internet continues to increase, a variety of network services continue to emerge, and advanced multimedia systems have emerged. Since real-time services are sensitive to network transmission delays and delay jitter, there are bursty file transfer protocols (FTP, FILE Transfer Protocol) or hypertext transfer protocols (HTTP, Hyber Text) with image files. In the case of services such as Transport Protocol, real-time services are greatly affected. On the other hand, multimedia services take up a large amount of bandwidth, so that the critical services to be guaranteed by existing networks are difficult to obtain reliable transmission. As a result, various QoS (Quality Of Service) technologies emerged. The Network Engineering Task Force (IETF) has proposed a number of service models and mechanisms to meet the needs of QoS.

Portal-based applications and services as well as broadband multimedia services, including rich video/Audio streaming, video on demand (VOD, Video On Demand), video multicast, multimedia interaction, high bandwidth demand online games for ordinary residential users, Providing video conferencing, distance education, virtual private network (VPN, Virtue Private Network), QoS-guaranteed data line, and IPHotel (Internet Protocol Hotel) for commercial users, it is an important part of broadband operation.

Operators and enterprises have a high level of awareness of Ethernet technology and end-to-end Ethernet technology. Ethernet technology has become one of the main technologies for building a triple play and metropolitan area network in the future. The Ethernet service will have great development in the future market.

For these commercial users, virtual switching (V-Switch, Virtue-Switch), Generic Routing Encapsulation (GRE), Layer 2 Tunneling Protocol (L2TP), and multi-protocol label switching have been proposed. (MPLS, Multiple Protocols Label Switch) and other methods of virtual private network / virtual private dial-up network (VPN / VPDN, Virtue Private Dial-up Network) dedicated line solution.

Virtual Switching (V-Switch) technology is mainly used to build stable, practical, and economical carrier-grade Metro Ethernet can implement QoS guarantee, network security protection, carrier-class network maintenance and management functions, and has core business management capabilities such as number-based user management, certain mobility, and centralized management of business open management billing. Provides services such as intelligent Layer 2 traffic scheduling, local area network (LAN), IP traffic planning, and other services. The emergence of virtual switching (V-Switch) fills the gaps in the current network's pure Layer 2 capabilities. The construction of the metropolitan area network has an important role.

The virtual switch (V-Switch) architecture has complete Ethernet VLAN switching and scheduling functions, flexible service scheduling and establishment and adjustment methods, and rich and scalable Layer 2 service provision capabilities. And information.

The logical hierarchy and functional model of the virtual switching (V-Switch) system are shown in Figure 1. In Figure 1, the virtual switching (V-Switch) system is divided into three layers: the V-Switch service control layer, the V-Switch connection control layer, and the V-Switch bearer capability layer.

The V-Switch connection control layer maintains the switching resources in the Data Retransmission Entity (DRE), such as devices, ports, links, and VLANs, and receives the V-Switch service control layer service control entity (SCR, Service Control Register). The V-Switch establishes a request, selects a service flow path for the V-Switch connection, allocates bandwidth and a virtual local area network (VLAN) resource, and sends the control information to the data forwarding entity DRE device through which the service flow passes. on.

The DRE is in the V-Switch bearer capability layer. The DRE completes the service flow forwarding in the Ethernet frame format according to the VLAN switch entry set by the V-Switch connection control layer.

The contents of the VLAN switch entry are shown in Table 1.

Parameter parameter description

1 port 1 service flow forwarding port 1, such as GE 1/0/0

2 vlan id 1 VLAN ID carried in the Ethernet frame format on port 1

3 port 2 service flow forwarding port 2, such as GE 1/0/1

4 VLAN ID of the vlan id 2 service flow carried in the Ethernet frame format on port 2.

5 Bandwidth Service Flow Bandwidth Limit

5. 1 Upstream maximum bandwidth service flow uplink (received from port 1, sent from port 2) Maximum bandwidth 5. 2 Downstream Maximum Bandwidth Service Flow Downstream (received from port 2, sent from port 1)

Large bandwidth

6 QoS parameters Service flow QoS parameter requirements

6. 1 delay

6. 2 Delay jitter

6. 3 packet loss rate

The DRE uses the VLAN switching table as the routing basis for service flow forwarding. The DRE forwards the service flow data according to the VLAN exchange table as shown in Figure 2:

In Figure 2, the DRE in the V-Switch domain includes: an edge DRE and an intermediate DRE. The edge DRE is the edge node of the V-Switoh domain, which is connected to the access user, and the intermediate DRE is connected to the access user through the edge DRE. Both edge DRE and intermediate DRE only support "port + Vlan" exchange. For example, DRE sends the vlan id 1 Ethernet frame received on port 1 to port 2, and converts vlan id 1 to vlan id 2; DRE will The Ethernet frame of vlan id 2 received in port 2 is sent to port 1 and vlan id 2 is converted to vlan id 1.

Through the above forwarding form, a virtual channel of a VLAN can be established in the entire network. The virtual channel can be described as: (device 1, port 1, vlan id 1) - (device 1, port 2, vlan id 2) - (device 2, port 3, vlan id 2) - (device 2, port 1, vlan id 3) ....

Because the current V-Switch does not fully consider the service requirements of the edge DRE, the access technology on the edge DRE is single, and only the access mode based on "port + VLAN" is required, and the user equipment is required to send to the access DRE. The Ethernet tag must be marked with an Ethernet frame. When the client device does not support VLANs, you need to add a Layer 2 (L2) device between the client device and the access DRE. The VLAN is encapsulated in the Ethernet frame sent by the client device to the DRE. The application of technology is very inconvenient, and it also increases the equipment overhead of building a network. .

In addition, when the V-Switch carries an Ethernet frame to implement point-to-multipoint interworking, the "port + Vlan" of the Ethernet frame that receives the edge of the Ethernet frame received from the user side must be different, as shown in Figure 4. As shown in the figure, if the data forwarding entity DRE whose medium access control address is MAC (Medium Access Control) 1 needs to communicate with the data forwarding entity DRE whose media access control address is MAC2, MAC3, and MAC4 respectively, the V-Switch domain needs to be Establish three virtual channels of V-Switch for these three service flows, and the access address is MAC1. The port + Vlan of the Ethernet frame received by the edge DRE from the user side must be different, so that it can be switched to different V-Switch virtual channels. Therefore, the requirements on the user side device are very high, requiring each for each Possible destination MAC addresses establish independent routes, and different "ports + VLANs" are forwarded. This is difficult to implement in actual networking.

A session connection can be established between any two IP addresses in the IP protocol. For example, any IP address in the Next Generation Network (NGN) service can establish a session connection with multiple IP addresses at the same time. When the V-Switch carries IP services, the "port + Vlan" of the Ethernet frame carrying the IP flow received by the edge DRE of the access IP service from the user side must be different, as shown in Figure 3, if IP1 needs both IP2 and IP2. , IP3, IP4 communication, the V-Switch domain needs to establish three virtual channels of V-Switeh for the three IP flows, and the port of the Ethernet frame that receives the IP flow received by the edge DRE of the IP1 from the user side. + Vlan, must be different, so that it can be switched to different V-Switch virtual channels. Therefore, when the V-Switch carries IP services, the Layer 3 routing device on the user side is required for each possible destination IP address. Establish independent routes and take different interfaces, which is almost impossible to implement in actual networking.

The present invention aims to provide a virtual switching access method and apparatus, and a virtual switching system. The present invention improves the virtual switching access method based on multiple policies (including ports, MAC addresses, IP addresses, etc.) on the edge DRE. The architecture of the virtual switching system enriches the virtual switching access method, thereby improving the access capability of the virtual switching technology, expanding the application range of the virtual switching technology, and reducing the cost of forming the network.

According to a first aspect of the present invention, a virtual switching access method is provided, which is applied to a virtual switching system having an edge data forwarding entity configured with a virtual switching policy, the method comprising:

Establish a mapping relationship between user data packets and virtual switching channels;

The edge data forwarding entity performs virtual exchange data transmission according to the mapping relationship.

Also includes:

According to the configured virtual switching policy, the edge data forwarding entity maps the user data packet to the corresponding virtual switching channel;

The data packet is encapsulated according to the virtual switching channel Ethernet frame format, and the encapsulated data packet is transmitted along the virtual switching channel. The virtual switching policy includes:

The mapping between the port and the virtual switch channel is established based on the port, or the mapping between the port and the virtual local area network VLAN ID and the virtual switch channel is established.

Establish a mapping relationship between the MAC address and the virtual switching channel according to the media access control MAC address; or

According to the IP address or IP quintuple, the mapping relationship between the IP address or the IP quintuple and the virtual switching channel is established.

Another virtual switching access method provided by the present invention includes:

Establishing a mapping relationship between the port and the virtual switching channel in the edge data forwarding entity;

The data forwarding entity performs virtual exchange data transmission according to the mapping relationship of the received Ethernet frame.

The establishing a mapping relationship includes:

Establishing a correspondence between the first port and the second port and the VLAN identifiers of the first port and the second port, where the correspondence is saved in the VLAN switch table.

The virtual exchange data transmission includes:

The Ethernet frame received from the first port is encapsulated by the VUN protocol according to the second port virtual local area network VLAN identifier in the corresponding relationship, and the encapsulated Ethernet frame is transmitted through the second port;

The Ethernet frame encapsulated by the second port VLAN identifier according to the VLAN protocol is decapsulated, and the decapsulated Ethernet frame is transmitted through the first port.

When the VLAN ID corresponding to the first port in the record of the VLAN switch is invalid, and the VLAN ID corresponding to the second port is valid, the information stored in the record is the correspondence between the first port and the second port and the VLAN identifier. ; or

When the VLAN IDs of the two ports in the record of the VLAN switch are invalid VLAN IDs, the information stored in the record is the correspondence between the first port and the second port.

When the predetermined field in a record of the VLAN switch table is the first predetermined value, the information stored in the record is a correspondence between the first port and the second port and the second port VLAN identifier;

When a predetermined field in a record of the VLAN switch table is a second predetermined value, the record is in the record The stored information is a correspondence between the first port and the second port.

According to a second aspect of the present invention, a virtual switching access device is provided, including: a storage module: a mapping relationship between a storage port and a virtual switching channel;

The switching module: performs the virtual exchange data transmission according to the mapping relationship stored in the storage module.

A further virtual switching access method according to the present invention includes:

Establish a mapping relationship between IP address information and virtual switching channels.

The data forwarding entity determines the location according to the IP layer address information of the IP packet and the mapping relationship.

IP · 虚对应 corresponding virtual exchange channel;

The data forwarding entity carries the IP packet according to the virtual switching channel corresponding to the IP packet for transmission.

The establishing a mapping relationship includes:

Establish the mapping relationship between the IP address information of the IP layer or the quintuple information of the IP layer and the outgoing port of the virtual switching channel and the VL AN identifier.

More suitably, further comprising:

The IP packet of the access virtual switching domain received from the Ethernet layer is delivered to the IP layer;

The IP layer determines the egress port and the outbound VLAN identifier of the virtual switch channel corresponding to the IP packet according to the mapping relationship, the destination address information of the IP packet to be accessed in the virtual switching domain, or the quintuple information of the IP layer.

The message transmission includes: an out VLAN identifier encapsulating and transmitting an Ethernet frame of the IP >3⁄4 text; or

The Ethernet layer of the egress data forwarding entity transmits the IP packet that needs to be forwarded out of the virtual switching domain to the IP layer.

The IP layer performs IP packet transmission according to the destination address of the IP packet.

Another virtual switching access device according to the present invention includes:

The storage module is configured to: store the mapping relationship between the IP address information and the virtual switch channel; the channel determining module: receives the IP packet that is connected to the virtual switching domain, and according to the mapping relationship stored in the storage module, the IP carried in the IP packet Layer address information determines that the IP packet corresponds to Virtual exchange channel

The encapsulation and forwarding module: encapsulates and transmits the IP packet according to the virtual switching channel determined by the virtual switching channel module.

According to a third aspect of the present invention, a virtual switching system is provided, including: an ingress data forwarding entity and an egress data forwarding entity;

The ingress data forwarding entity determines the virtual switching channel corresponding to the IP packet according to the mapping relationship between the IP layer address information and the virtual switching channel, and the IP layer address information of the IP packet accessing the virtual switching domain, and the IP address is After the packet is encapsulated, it is transmitted on the virtual switched channel.

The egress packet forwarding entity: The IP packet that needs to be forwarded out of the virtual switching domain is transmitted according to the IP layer destination address it carries.

The ingress data forwarding entity includes:

The storage module is configured to: store the mapping relationship between the IP address information and the virtual switch channel; the channel determining module: receives the IP packet that is connected to the virtual switching domain, and according to the mapping relationship stored in the storage module, the IP carried in the IP packet The layer address information determines a virtual switching channel corresponding to the IP packet.

Encapsulating and forwarding module: The IP packet is encapsulated and transmitted according to the determined virtual switching channel. The virtual switching system further includes: at least one intermediate data forwarding entity;

The intermediate data forwarding entity transmits IP packets in the virtual switching domain to other intermediate data forwarding entities or egress data forwarding entities according to the VLA exchange table.

According to a fourth aspect of the present invention, a virtual switching access method is provided, including: establishing a mapping relationship between Ethernet layer address information and a virtual switching channel;

Determining, according to the Ethernet layer address information of the Ethernet frame, the mapping relationship, the virtual switching channel corresponding to the Ethernet frame; the establishing the mapping relationship includes:

The mapping between the destination MAC address information and the port ID and VLAN ID is established. The mapping between the destination MAC address information and the source MAC address information of the virtual switch channel and the outbound VLAN ID is established.

More suitably, the method further comprises: The Ethernet layer of the ingress data forwarding entity determines the egress port and the outbound virtual switching channel corresponding to the Ethernet frame according to the mapping relationship, the destination MAC address information of the Ethernet frame that accesses the virtual switching domain, or the destination MAC address and the source MAC address information. VLAN ID.

The Ethernet frame transmission includes:

The port data forwarding entity encapsulates and transmits the Ethernet frame corresponding to the port corresponding to the Ethernet frame and the VLAN identifier;

The intermediate data forwarding entity transmits the Ethernet frame in the virtual switching domain it receives according to the VLA exchange table;

The egress data forwarding entity that forwards the Ethernet frame out of the virtual switching domain receives the Ethernet frame in the virtual switching domain and transmits according to the destination MAC address of the Ethernet frame.

According to the present invention, there is still another virtual switching access device, including:

The storage module is configured to: store a mapping relationship between the Ethernet layer address information and the virtual switching channel; and the channel determining module: receiving an Ethernet frame that accesses the virtual switching domain, and according to the mapping relationship stored in the storage module, the Ethernet layer address carried in the Ethernet frame The information determines a virtual switching channel corresponding to the Ethernet frame;

Encapsulating and forwarding module: The Ethernet frame is encapsulated and transmitted according to the virtual switching channel determined by the virtual switching channel module.

According to still another virtual switching system of the present invention, the method includes: an ingress data forwarding entity and an egress data forwarding entity;

The ingress data forwarding entity: determining a virtual switching channel corresponding to the Ethernet frame according to the mapping relationship between the Ethernet layer address information and the virtual switching channel, and the Ethernet layer address information of the Ethernet frame of the access virtual switching domain, and encapsulating the Ethernet frame Post transmission

Egress data forwarding entity: The Ethernet frame transmitted in the virtual switching domain is transmitted according to the destination MAC address it carries.

The ingress data forwarding entity includes:

The storage module is configured to: store a mapping relationship between the Ethernet layer address information and the virtual switching channel; and the channel determining module: receiving an Ethernet frame that accesses the virtual switching domain, and according to the mapping relationship stored in the storage module, the Ethernet layer address carried in the Ethernet frame The information determines a virtual switching channel corresponding to the Ethernet frame; Encapsulating and forwarding module: The Ethernet frame is encapsulated and transmitted according to the virtual switching channel determined by the virtual switching channel module.

The virtual switching system further includes: at least one intermediate data forwarding entity;

Intermediate data forwarding entity: The Ethernet frame of the virtual switching domain is transmitted to other intermediate data forwarding entities or egress data forwarding entities according to the VLAN switching table.

The port-based policy is to establish a mapping relationship between the port and the virtual switching channel on the edge DRE, so that the edge DRE can perform virtual switch transmission of the common Ethernet frame according to the port. For example, the ingress DRE can transmit the client device without passing the 802.1Q. The encapsulated ordinary Ethernet frame is directly encapsulated in 802.1Q and transmitted through the virtual switch channel corresponding to the receiving port. The egress DRE can convert the 802.1Q encapsulated Ethernet frame transmitted from the virtual switched domain into a normal Ethernet frame and directly pass the corresponding output. The port is sent. For example, the ingress DRE can directly send the normal Ethernet frame received from the port through the corresponding other port. Therefore, when the user equipment does not support the VLAN, the L2 device does not need to be added, and the network is reduced. The cost of the virtual switch is increased. The virtual switching method is improved by setting the mapping relationship in the VLAN switching table, such as making full use of the existing fields in the VLAN switching table. Access capability, expand the application range of virtual switching technology, reduce the group The purpose of building network costs.

The IP address-based policy provides a virtual switch access method for the IP packet. The mapping between the IP address information and the virtual switch channel is established on the edge DRE, so that the ingress DRE can directly obtain the address information of the IP packet. The virtual switch channel corresponding to the IP packet is determined, so that the V-Switch does not require the port + VLAN of the Ethernet frame to carry the IP service flow. Therefore, the port must be different through the same port. The IP packet of the information can be transmitted through the different ports + Vlan, and the virtual switch access method is enriched. The egress packet of the present invention can transmit IP packets to the virtual switch domain according to the IP address of the IP packet. Forwarding; thus, the technical solution provided by the present invention achieves the purpose of improving the carrying capacity of the virtual switching to the IP service and expanding the application range of the virtual switching technology.

The MAC address-based policy provides a new virtual switching access method for the Ethernet frame. By establishing the mapping relationship between the Ethernet layer address information and the virtual switching channel on the edge DRE, the ingress DRE can directly according to the address information of the Ethernet frame. Determining the virtual switching channel corresponding to the Ethernet frame, When the V-Switch is carrying an Ethernet frame, the "port + VLAN" of the Ethernet frame must not be required to be different, so that the Ethernet frame with different Ethernet layer address information accessed through the same port or the same "port + VLAN" is allowed. The virtual switch access method can be enriched by different "port + VLAN"transmission; the egress DRE of the present invention can forward the Ethernet frame transmitted by the virtual switch domain according to the destination MAC address of the Ethernet frame; thus, provided by the present invention The technical solution achieves the purpose of improving the access capability of the virtual switching technology and expanding the application range of the virtual switching technology. DRAWINGS

Figure 1 is a schematic diagram of the logical hierarchy and functional model of the virtual switching system;

2 is a schematic diagram of a method for implementing virtual switching in the prior art;

3 is a schematic diagram of a virtual switch access implementation principle of an IP service in the prior art; FIG. 4 is a schematic diagram of a virtual switch access principle based on an Ethernet frame in the prior art;

5 is a schematic diagram of an implementation principle of virtual switching access according to the present invention;

6 is a schematic diagram of a virtual-switching access principle based on "port-port-port, VLAN" according to a first embodiment of the present invention;

7 is a schematic diagram of a virtual port access principle based on "port-one port" according to a second embodiment of the present invention;

8 is a schematic diagram of a virtual switch access device according to the present invention;

9 is a schematic diagram of an IP-based virtual switch access implementation principle according to the present invention; FIG. 10 is a schematic diagram of implementing IP layer point-to-multipoint service based on IP-based virtual switch access according to the present invention;

11 is a schematic diagram of a virtual switch access principle based on an Ethernet frame according to the present invention; FIG. 12 is a schematic diagram showing the principle of implementing Ethernet layer-to-multipoint service based on virtual switch access of an Ethernet frame according to the present invention. detailed description

The present invention proposes a new virtual switching access method, device and virtual switching system in the V-Switch technology. The core idea is to establish a mapping relationship between a user data packet/frame and a virtual switching channel by configuring a policy at the edge entry DRE. It mainly includes the mapping between the port and the virtual switching channel, the mapping between the IP address information and the virtual switching channel, or the Ethernet layer address information and the virtual switching channel. 06 001654

- 11- mapping relationship, when the data packet/frame from the user enters the virtual switching domain, the ingress DRE selects a suitable virtual switching channel to transmit the service flow according to the above policy.

The present invention establishes a mapping relationship between a user data packet/frame and a virtual switching channel by configuring a virtual switching policy at the edge DRE. When a user data packet/frame from a user edge device (CE, Customer Edge) enters the virtual switching domain, the ingress DRE selects a suitable virtual switching channel to transmit the service flow according to the policy, and the service flow passes through the virtual switching channel. The exchange of several intermediate DREs finally reaches the edge exit DRE, the exit DRE terminates the tunnel, obtains the user data packet/frame, and then forwards it to the CE. FIG. 5 is a schematic diagram showing the principle of virtual switching access according to the present invention, with reference to FIG. 5.

At the ingress DRE, the user data packet/data frame is mapped to the virtual switching channel according to the configured policy, and is encapsulated according to the virtual switching channel Ethernet frame format, and then the encapsulated data packet is sent out from the outbound port.

The intermediate DRE forwards the data packet according to the port and the VLAN;

At the egress DRE, the virtual switched channel is terminated to restore the user packet/frame and forwarded to the CE device. The virtual exchange policy of the above-mentioned entry DRE is flexible, and can be based on port, port +

VLAN ID, MAC address (destination MAC, destination MAC + source MAC), IP address (such as S IP address, IPv4 quintuple). The policy of "Port + Vlan," is supported by the existing technology. Other strategies are as follows:

According to the port policy-based virtual switching access method provided by the present invention, in the process of performing virtual switching access, the edge of the virtual switching access may be only a port, and the other end may be only a port or a port. +VLAN" and so on. When both ends of the virtual switch access are ports, the Ethernet frame transmission channel between the port and the port is a special case in the virtual switch channel.

In order to make the principles, features and advantages of the present invention more apparent, the invention will be further described in accordance with the accompanying drawings.

This embodiment adopts a virtual port access method based on "port one port, VLAN". First, the mapping between port 1 and port 2 and VLAN ID 2 is required. Therefore, the DRE can encapsulate the normal Ethernet frame transmitted by the client device received from port 1 according to the VLAN identifier in the corresponding relationship, such as VLAN. The identifier 2 is encapsulated in the 802.1Q, and the encapsulated Ethernet frame is transmitted according to the virtual switching channel in the corresponding relationship; at the same time, the DRE receives the encapsulated VLAN 2 from the virtual switched channel in the corresponding relationship. The Ethernet frame is converted into a normal Ethernet frame, and the converted normal Ethernet frame is directly transmitted through the port 1 according to the correspondence. In this way, when the user equipment does not support the VLAN, the L2 device does not need to be added between the user equipment and the ingress DRE, or between the user equipment and the egress DRE, which reduces the equipment overhead of the network and increases the application of the virtual exchange. Scenes.

6 is a schematic diagram of virtual port access based on "port-port, VLAN" according to the present invention, with reference to FIG. 6,

The normal Ethernet frame on the user side enters the V-Switch domain through port 1 of the ingress DRE. When port 1 of the ingress DRE corresponds to port 2 and VLAN ID 2, the Ethernet layer of the ingress DRE performs 802.1Q on the common Ethernet frame according to the VLAN ID 2. Encapsulation, and sending the 802.1Q encapsulated Ethernet frame according to port 2 and VLAN identifier 2.

After receiving the 802.1Q encapsulated Ethernet frame, the Ethernet layer of the intermediate DRE still transmits in the "port + VLAN" mode according to the VLAN exchange table.

After the egress Ethernet layer receives the 802.1Q encapsulated Ethernet frame through the port 2, when the port 1 of the egress DRE corresponds to the port 2 and the VLAN ID 2, the Ethernet layer of the egress DRE converts the 802.1 Q encapsulated Ethernet frame into an ordinary ether. Frame, and send the normal Ethernet frame directly through port 1.

This embodiment adopts a virtual port access method based on "port one port".

First, it is necessary to establish a correspondence between two ports, that is, a correspondence between port 1 and port 2, so that the DRE can directly transmit the normal Ethernet frame received from one port to another port according to the correspondence, that is, It is not necessary to VLAN-identify and encapsulate the Ethernet frame transmitted between the two ports, and directly perform virtual exchange transmission. In this way, when the user equipment does not support the VLAN, the L2 device does not need to be added between the user equipment and the ingress DRE, and between the user equipment and the egress DRE, which reduces the equipment overhead of the network and increases the application of the virtual exchange. Scenes.

7 is a schematic diagram of a virtual port access principle based on "port-port" according to the present invention, referring to FIG. 7,

The ingress DRE and the egress DRE are the same DRE, and the normal Ethernet frame on the user side enters the V-Switch domain through the port 1 of the ingress DRE. When the port 1 of the ingress DRE corresponds to the port 2, the Ethernet layer of the ingress DRE associates the ordinary Ether frame. Send directly through port 2.

The correspondence in the above embodiment may be set in the VLAN switching table. In the VLAN The method of setting the above correspondence in the exchange table is flexible, and can be set to the form of Table 2.

Table 2

In Table 2, the vlan id 1 of port 1 is an invalid VLAN ID, and the vlan id2 of port 2 is a valid VLAN ID, or the vlan id 1 of port 1 is a valid VLAN ID, and the vlan id of port 2 is 2 If the VLAN ID is invalid, or if the VLAN ID of port 1 is invalid, and the VLAN ID of port 2 is invalid, the DRE needs to perform the port-based virtual switch according to the entry. Into the method.

If the vlan id1 of port 1 is a valid VLAN ID and the vlan id2 of port 2 is also a valid VLAN ID, it indicates that the DRE needs to perform the virtual switch based on "port + Vlan" in the prior art according to the entry. Into the method.

When the corresponding relationship in the present invention is set to the form of Table 2, the specific process of DRE implementing virtual switch access is:

After receiving the Ethernet frame from port 1, the DRE needs to determine the entry of the VLAN switch table corresponding to the Ethernet frame, and then determine the VLAN ID vlan idl and VLAN in the entry. Whether the value of ID vlan id2 is a valid value.

If the vlan id1 is a valid value and the vlan id2 is an invalid value, the packet received by the ingress DRE is an 802.1Q encapsulated Ethernet frame of the vlan id1, and the ingress DRE needs to convert the 802.1Q encapsulated Ethernet frame into an ordinary one. The Ethernet frame is sent directly through port 2.

If vlan id2 is a valid value and vlan idl is an invalid value, the packet received by the ingress DRE is a normal Ethernet frame. The DRE needs to convert the ordinary Ethernet frame into an 802.1 Q encapsulated Ethernet frame of vlan id2. And send it out through port 2.

If the vlan id1 is invalid and the vlan id2 is invalid, the packet received by the DRE is an ordinary Ethernet frame. The ingress DRE only needs to send the ordinary Ethernet frame directly through port 2.

If vlan id2 is a valid value and vlan idl is also a valid value, the packet received by the ingress DRE is the 802.1Q encapsulated Ethernet frame of the VLAN ID vlan idl, and the ingress DRE receives the vlan idl of the port 1. The 802.1Q encapsulated Ethernet frame is sent to port 2, and the vlan idl is converted to vlan id2. -

The setting, such as adding a new switching mode type in the VLAN switching table, indicating the VLAN switching mode used by the DRE by using different values of the switching mode type, and adding two fields in the VLAN switching table, The different values of the two fields indicate which VLAN switching mode the DRE uses.

A virtual switching access device provided by the present invention is shown in FIG.

Referring to FIG. 8, the virtual switch access device includes: a storage module and a switch module.

The storage module is mainly used to map the relationship between the storage port and the virtual switch channel, including: the mapping between port 1 and port 2 and VLAN ID 2, the mapping between port 1 and port 2, and port 1 and VLAN ID 1 and port 2 and VLAN. The correspondence of the identifier 2. These correspondences can be stored in the form of a VLAN exchange table.

The switching module is configured to receive an Ethernet frame, determine an entry in the VLAN switch table corresponding to the Ethernet frame, and determine a virtual switch access mode of the Ethernet frame according to the content in the entry, and perform virtual switching according to the Ethernet frame. The access mode sends an Ethernet port, such as based on "port one port, Virtual switching access method of VLAN", or virtual port access method based on "port-port", or virtual port exchange based on "port, VLAN-port, VLAN". Switch module for virtual frame of Ethernet The exchange transmission method is as described in the above method and will not be described in detail herein.

On the other hand, the virtual address access method based on the IP address policy provided by the present invention: the IP layer address information in the present invention may be the IP layer destination address information, or may be the IP layer quintuple (source IP address, destination IP address) Address, protocol number, source port number, destination port number) information. Therefore, the mapping relationship between the IP address information and the virtual switch channel in the edge DRE is as follows: The mapping between the IP address of the IP layer and the outbound port and the outbound VLAN of the virtual switch channel is established, or the IP layer quintuple is established. The mapping between the information and the outgoing port of the virtual switching channel and the outgoing VLA identifier.

The above mapping relationship can be stored in the ingress DRE.

FIG. 9 is a schematic diagram of an IP-based virtual switch access implementation according to the present invention. The transmission process of IP packets in the virtual switch domain is described in detail below in conjunction with FIG. 9.

In Figure 8, when the IP service flow enters the V-Switch domain, the link layer of the ingress DRE submits the IP packet to the IP layer for processing. If the mapping relationship stored in the ingress DRE is the IP layer quintuple information and the virtual switch channel. In the mapping relationship, the IP layer should obtain the corresponding address information of the IP packet, and match the address information with the mapping relationship stored in the ingress DRE to determine the egress port and out VLAN ID of the virtual switching channel corresponding to the packet. . If the mapping relationship between the IP address and the virtual switch channel is the mapping between the IP address and the virtual switch channel, the IP address of the IP address is matched with the destination address of the IP address. The outbound port and outbound VLAN ID of the virtual switch channel corresponding to the IP packet are determined.

After the outbound port and the outgoing VLAN ID of the virtual switch channel corresponding to the IP packet are determined, the Ethernet layer of the ingress DRE uses the outbound VLAN identifier to encapsulate the IP packet in 802.1Q, and sends the packet according to the determined outbound port.

When the intermediate DRE Ethernet layer receives the 802.1 Q-encapsulated IP packet, it still transmits the packet according to the VLAN switch table as "port + VLAN".

After the egress Ethernet layer receives the 802.1Q encapsulated IP packet, the egress Ethernet interface terminates the virtual switch channel and sends the IP packet to the IP layer for processing. The IP layer forwards the packet according to the destination address of the IP packet. 10 is a schematic diagram of implementing IP layer point-to-multipoint service based on IP-based virtual switching access according to the present invention. The virtual switching process of establishing a session connection between an IP address and multiple IP addresses is described in detail below with reference to FIG. .

In Figure 10, if IP1 needs to communicate with IP2, IP3, and IP4 at the same time, the V-Switch domain will establish three V-Switch channels for the three IP service flows. Since the IP address information of the three IP service flows received by the access DRE1 of the access IP1 is different from the IP address, the IP header addresses are IP2, IP3, and IP4 respectively. Therefore, the DRE1 can be in the IP address. After the IP address of the IP address is encapsulated, the three IP service flows are encapsulated and switched to different V-Switch channels. After the three IP service flows are encapsulated, they are respectively identified according to different egress ports and VLAN IDs. Transfer to DRE2 and DRE4.

When receiving the two IP service flows transmitted by DRE1, DRE2 transmits one of the IP service flows to DRE3 according to the VLAN switching table, and submits one IP service flow to the IP layer for processing, and the IP layer of DRE2 is based on the IP. The destination address of the packet forwards the IP traffic flow route to IP2.

DRE3 submits the IP service flow transmitted by DRE2 to the IP layer for processing. The IP layer of DRE3 forwards the IP service flow route to IP3 according to the destination address of the IP packet.

When receiving an IP service flow from DRE1, DRE4 transmits it to DRE5 according to the VLAN switch table. DRE5 submits the IP service flow transmitted by DRE4 to the IP layer for processing. The IP layer of DRE5 forwards the IP service flow route to IP4 according to the destination address of the IP packet.

Thereby completing a virtual exchange process in which an IP address establishes a session connection with multiple IP addresses at the same time.

It can be seen from the above description that when the virtual packet transmission method of the present invention is used for virtual switching transmission of IP packets, there is no special requirement for the routing device on the user side, and only the routing device between the user side and the ingress DRE is only Need a port connection.

A virtual switching system provided by the present invention includes: an ingress data forwarding entity and an egress data forwarding entity. Of course, the virtual switching system may also include an intermediate data forwarding entity, and the number of intermediate data forwarding entities may be one or more. The following describes an example of a virtual switching system including an ingress data forwarding entity, an intermediate data forwarding entity, and an egress data forwarding entity.

The number of the ingress data forwarding entity, the intermediate data forwarding entity, and the egress data forwarding entity in the system may be one or more.

The ingress data forwarding entity is mainly used to map the IP layer address information to the virtual switching channel. The IP address information of the IP address of the IP address of the virtual switch domain is determined, and the virtual switch channel corresponding to the IP packet is determined, and the IP packet is encapsulated and transmitted to the intermediate data forwarding entity or the egress data forwarding entity.

The ingress data forwarding entity is an IP-based virtual switching access device, which includes a storage module, a virtual switching channel module, and a package forwarding module.

The storage module is mainly used for storing the mapping relationship between the IP address information and the virtual switching channel, such as the mapping between the IP address of the IP layer and the outgoing port of the virtual switching channel, and the outbound VLAN identifier. Five-tuple information, etc.

The virtual switching channel determining module is configured to receive an IP packet that is connected to the virtual switching domain, such as receiving an IP packet transmitted by the routing device on the user side, and receiving the IP packet according to the mapping relationship stored in the storage module. The IP address information carried in the IP address determines the virtual switch channel corresponding to the IP packet. For example, the outbound port and the outbound VLAN identifier of the virtual switch channel corresponding to the IP packet are determined according to the destination address of the IP packet.

The encapsulation and forwarding module is configured to encapsulate the IP packet and transmit it to the intermediate DRE or the egress DRE according to the information determined by the port and the VLAN identifier determined by the virtual switching channel module.

The intermediate data forwarding entity is configured to receive the IP packet transmitted by the encapsulating and forwarding module of the ingress DRE or other intermediate DRE, and transmit the IP packet received by the inbound to the other intermediate data forwarding entity or the egress data forwarding entity according to the VLAN switching table. .

The egress packet forwarding entity is mainly used to transmit the IP packet transmitted by the ingress data forwarding entity or the intermediate data forwarding entity according to the destination address of the bearer.

In another aspect, the virtual address access method based on the MAC address policy provided by the present invention: the Ethernet layer address information in the present invention may be the destination MAC address information, or may be the destination MAC address information and the source MAC address information. Therefore, the mapping relationship between the Ethernet layer address information and the virtual switching channel is established in the edge DRE as follows: establishing a mapping relationship between the destination MAC address information and the outbound port and the outgoing VLAN identifier of the virtual switching channel, or establishing the destination MAC address information and source. The mapping between the MAC address information and the egress port and outbound VLAN ID of the virtual switching channel.

The above mapping relationship can be stored in the ingress DRE.

11 is a schematic diagram of a virtual switch access principle based on an Ethernet frame according to the present invention. The transmission process of an Ethernet frame in a virtual switched domain is described in detail below with reference to FIG. 11. In Figure 11, if the mapping relationship stored in the ingress DRE is the mapping between the destination MAC address, the source MAC address information, and the outbound port and outbound VLAN identifier of the virtual switching channel, the ingress DRE Ethernet layer should obtain the destination MAC address of the Ethernet frame and The source MAC address information is matched, and the destination MAC address and the source MAC address information are matched with the mapping relationship stored in the ingress DRE to determine the outbound port and outbound VLAN identifier of the virtual switching channel corresponding to the Ethernet frame. If the mapping between the destination MAC address and the outbound port of the virtual switch channel and the port of the outbound VLAN are mapped, the destination DRE Ethernet layer should acquire the Ethernet frame when the Ethernet frame enters the V-Switch domain. The MAC address is matched with the mapping relationship between the destination MAC address and the ingress DRE to determine the outbound port and outbound VLAN identifier of the virtual switching channel corresponding to the Ethernet frame.

After the outbound port and the outbound VLAN identifier of the virtual switch channel corresponding to the Ethernet frame are determined, the ingress DRE layer uses the outbound VLAN identifier to encapsulate the Ethernet frame in 802.1Q, and sends the packet according to the determined out port.

When the Ethernet layer of the intermediate DRE receives the 802.1Q encapsulated Ethernet frame, it still transmits in the "port + VLAN" according to the VLAN switch table.

After the Ethernet layer of the egress DRE receives the 802.1Q encapsulated Ethernet frame, the Ethernet layer of the egress DRE terminates the virtual switch channel and forwards according to the destination MAC address of the Ether frame.

FIG. 12 is a schematic diagram of the principle of implementing Ethernet layer-to-multipoint service based on virtual switching access of an Ethernet frame according to the present invention. The virtual switching process of establishing a session connection between a MAC address and multiple MAC addresses simultaneously is performed in conjunction with FIG. 11. A detailed description.

In Figure 12, if MAC1 needs to communicate with MAC2, MAC3, and MAC4 at the same time, the V-Switch domain establishes three V-Switch channels for three Ethernet frame traffic flows. The Ethernet layer address information of the three Ethernet frame service flows received from the user side is different, for example, the destination MAC address is MAC2, MAC3, and MAC4. Therefore, the DRE1 can be based on the destination MAC address. The address or the three Ethernet frame service flows are encapsulated according to the destination MAC address and the source MAC address, and then switched to different V-Switch channels, that is, after the three Ethernet frame service flows are encapsulated, according to different outgoing ports and VLANs. The identifiers are transmitted to DRE2 and DRE4, respectively.

When DRE2 receives two Ethernet frame traffic flows from DRE1, it is based on VLAN The table exchanges one of the Ethernet frame traffic to DRE3, and the other Ethernet frame traffic forwards the Ethernet frame traffic to MAC2 according to the destination MAC address of the Ethernet frame or according to the destination MAC address and the source MAC address. DRE3 receives the Ethernet frame traffic transmitted by DRE2, and forwards the Ethernet frame traffic to MAC3.

When receiving the Ethernet frame traffic from DRE1, DRE4 transmits it to DRE5 according to the VLAN switch table. DRE5 receives the Ethernet frame traffic transmitted by DRE4 and forwards it to MAC4 according to the ATM traffic.

Thus, a virtual exchange process in which a MAC address establishes a session connection with multiple MAC addresses simultaneously is completed.

It can be seen from the above description that when the virtual switching access method of the present invention is used for virtual switching access transmission of the Ethernet frame, there is no special requirement for the Layer 2 or Layer 3 routing device on the user side, and the second layer of the user side. Or only one port connection is required between the Layer 3 routing device and the ingress DRE.

Another virtual switching system provided by the present invention mainly includes: an ingress data forwarding entity and an egress data forwarding entity. Of course, the virtual switching system may also include an intermediate data forwarding entity, and the number of intermediate data forwarding entities may be one or more. The virtual switching system includes an ingress data forwarding entity, an intermediate data forwarding entity, and an egress data forwarding entity as an example.

The ingress data forwarding entity is configured to determine the virtual switching channel corresponding to the Ethernet frame according to the mapping relationship between the Ethernet layer address information and the virtual switching channel, and the Ethernet layer address information of the Ethernet frame of the virtual switching domain, and encapsulate the Ethernet frame. , to the intermediate data forwarding entity or the export data forwarding entity.

The ingress data forwarding entity is a virtual switching access device based on an Ethernet frame, and includes a storage module, a virtual switching channel module, and a package forwarding module.

The storage module is mainly used for storing the mapping relationship between the Ethernet layer address information and the virtual switching channel, such as the mapping between the destination MAC address and the outgoing port of the virtual switching channel, and the outgoing VLAN identifier. Of course, the Ethernet layer address information can also serve the purpose. MAC address and source MAC address information, etc.

The virtual switching channel determining module is mainly used for receiving an Ethernet frame that accesses the virtual switching domain, such as Receiving an Ethernet frame transmitted by the Layer 2 or Layer 3 routing device on the user side, and determining the virtual switching channel corresponding to the Ethernet frame according to the mapping relationship stored in the storage module and the Ethernet layer address information carried in the received Ethernet frame, such as The outbound port and the outbound VLAN identifier of the virtual switching channel corresponding to the Ethernet frame are determined according to the destination MAC address of the Ethernet frame.

The encapsulation and forwarding module is mainly used to encapsulate the Ethernet frame and transmit it to the intermediate DRE or the outlet DRE according to the information determined by the port and the VLAN identifier determined by the virtual switching channel module.

The intermediate data forwarding entity is mainly used for receiving an encapsulated forwarding module of the ingress DRE or other Ethernet frames transmitted by the intermediate DRE, and transmitting the received Ethernet frame to other intermediate data forwarding entities or egress data forwarding entities according to the VLA exchange table.

The egress data forwarding entity is mainly used to transmit the Ethernet frame transmitted by the ingress data forwarding entity or the intermediate data forwarding entity according to the destination MAC address it carries.

The invention has been described above by way of examples, but the invention is not limited thereto. It will be apparent to those skilled in the art that the present invention is susceptible to various modifications and alternatives without departing from the spirit and scope of the invention.

Claims

Rights request

A virtual switching access method, which is applied to a virtual switching system having an edge data forwarding entity configured with a virtual switching policy, and includes:

The virtual switching access method of claim 1, further comprising: mapping, by the edge data forwarding entity, the user data packet to the corresponding virtual switching channel according to the configured virtual switching policy;

The data packet is encapsulated according to the virtual switching channel Ethernet frame format, and the encapsulated data packet is transmitted along the virtual switching channel.

The virtual switching access method according to claim 1 or 2, wherein the virtual switching policy comprises:

The mapping between the port and the virtual switching channel is established based on the port, or the mapping between the port and the virtual local area network VLAN ID and the virtual switching channel is established.

Establishing a mapping relationship between the MAC address and the virtual switching channel according to the media access control MAC address; or

The mapping relationship between the IP address or the IP quintuple and the virtual switching channel is established according to the IP address or the IP quintuple.

4. A virtual switching access method, characterized in that:

The virtual switch access method according to claim 4, wherein the establishing the mapping relationship comprises:

The virtual switching access method according to claim 5, wherein the virtual exchange data transmission comprises:

The Ethernet frame received from the first port is according to the second port virtual office in the corresponding relationship The VLAN ID of the domain network is encapsulated by the VLAN protocol, and the encapsulated Ethernet frame is transmitted through the second port.

Decapsulating the Ethernet frame encapsulated by the VLAN protocol with the second port VLAN identifier received from the second port, and transmitting the decapsulated Ethernet frame through the first port.

The virtual switch access method according to claim 5 or 6, wherein when the VLAN identifier corresponding to the first port in the record of the VLAN switch table is invalid and the VLAN identifier corresponding to the second port is valid, The information stored in the record is a correspondence between the first port and the second port and the VLAN identifier; or

When the VLAN IDs of the two ports in the record of the VLAN switch are invalid VLAs, the information stored in the records is the correspondence between the first port and the second port.

The virtual switch access method according to claim 5 or 6, wherein when the predetermined field in a record of the VLAN switch table is a first predetermined value, the information stored in the record is the first port and Corresponding relationship between the second port and the second port VLAN identifier exchange;

When the predetermined field in a record of the VLAN switch table is a second predetermined value, the information stored in the record is a correspondence between the first port and the second port.

A virtual switching access device, comprising:

Storage module: The mapping relationship between the storage port and the virtual switch channel.

10. A virtual switching access method, characterized in that:

The data forwarding entity determines the virtual switching channel corresponding to the IP packet according to the IP layer address information of the BP packet and the mapping relationship;

The virtual switch access method according to claim 10, wherein the establishing the mapping relationship comprises:

The mapping between the IP address of the IP address and the outbound port and outbound VLAN of the virtual switch channel is established.

The method of accessing the virtual switch according to claim 11, further comprising:

The IP layer determines the egress port and the outbound VLAN identifier of the virtual switch channel corresponding to the IP packet according to the mapping relationship, the destination address information of the IP packet that accesses the virtual switching domain, or the IP layer quintuple information.

The virtual switch access method according to claim 12, wherein the message transmission comprises:

The Ethernet layer of the ingress data forwarding entity encapsulates and transmits the Ethernet frame according to the outbound port and the outbound VLAN identifier of the virtual switching channel determined by the IP layer; or

The Ethernet layer of the egress data forwarding entity transmits the IP packets that need to be forwarded out of the virtual switching domain to the IP layer.

14. A virtual switching access device, comprising:

The storage module: stores a mapping relationship between the IP layer address information and the virtual switching channel.

The channel determining module is configured to: receive the IP packet corresponding to the virtual switching domain, and determine the virtual switching channel corresponding to the packet according to the mapping relationship stored in the storage module and the IP layer address information carried in the packet;

A virtual switching system, comprising: an ingress data forwarding entity and an egress data forwarding entity;

The ingress data forwarding entity: determining a virtual switching channel corresponding to the IP packet according to the mapping relationship between the IP layer address information and the virtual switching channel, and the IP layer address information of the IP packet accessing the virtual switching domain, and the IP address After the packet is encapsulated, it is transmitted on the virtual switched channel.

The egress packet forwarding entity: IP packets that need to be forwarded out of the virtual switching domain according to the bearer

The IP layer destination address is transmitted.

The virtual switching system according to claim 15, wherein the ingress data forwarding entity comprises:

The channel determining module: receives an IP packet that is connected to the virtual switching domain, and stores the packet according to the storage module. The mapping relationship between the storage and the IP address information carried in the IP packet determines the virtual switching channel corresponding to the IP packet;

Encapsulating and forwarding module: The IP packet is encapsulated and transmitted according to the determined virtual switching channel.

The virtual switching system according to claim 15, wherein the virtual switching system further comprises: at least one intermediate data forwarding entity;

The intermediate data forwarding entity transmits the IP packet in the virtual switching domain to another intermediate data forwarding entity or an egress data forwarding entity according to the VLAN switching table.

18. A virtual switching access method, comprising:

Establish a mapping relationship between the Ethernet layer address information and the virtual switching channel;

Determining, according to the Ethernet layer address information of the Ethernet frame, the mapping relationship, the virtual switching channel corresponding to the Ethernet frame;

The virtual switch access method according to claim 18, wherein the establishing the mapping relationship comprises:

The mapping between the destination MAC address information and the port ID and VLAN ID is established. The mapping between the destination MAC address information and the source MAC address information and the outgoing port of the virtual switch channel and the outgoing VLAN ID are established.

The virtual switch access method of claim 19, further comprising:

The Ethernet layer of the ingress data forwarding entity determines the egress port and the outbound virtual switching channel corresponding to the Ethernet frame according to the mapping relationship, the destination MAC address information of the Ethernet frame accessing the virtual switching domain, or the destination MAC address and the source MAC address information. VLAN ID.

The virtual switch access method according to claim 20, wherein the Ethernet frame transmission comprises:

The ingress data forwarding entity encapsulates and transmits the Ethernet frame according to the port and the VLAN identifier corresponding to the Ethernet frame.

The intermediate data forwarding entity transmits the Ethernet frame in the virtual switching domain it receives according to the VLAN switching table;

The egress data forwarding entity that forwards the Ethernet frame out of the virtual switching domain receives the virtual switching domain. Too frame, and ~ according to the destination MAC address of the Ethernet frame for transmission.

22. A virtual switching access device, comprising:

The virtual switching system according to claim 23, wherein the ingress data forwarding entity comprises:

The virtual switching system according to claim 23, wherein the virtual switching system further comprises: at least one intermediate data forwarding entity;