CN116668374A

CN116668374A - Communication method and device

Info

Publication number: CN116668374A
Application number: CN202310687833.9A
Authority: CN
Inventors: 游君平; 汪洪远; 翟颖颖; 李玉涛; 赵海峰
Original assignee: New H3C Technologies Co Ltd
Current assignee: New H3C Technologies Co Ltd
Priority date: 2023-06-10
Filing date: 2023-06-10
Publication date: 2023-08-29

Abstract

The application provides a communication method and a device, wherein the method comprises the following steps: receiving a second service message, wherein the second service message and the first service message belong to different service flows; when the line card receiving the second service message is determined to be the second LC, selecting a second SFC connected with the second LC by using a global SFC DLB table, wherein the second SFC is the SFC with the minimum current bearing service flow in all SFCs connected with the second LC; acquiring a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in a flow setting table; selecting a first member port from the member ports according to the first current load state and the second current load state; and sending the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

Description

Communication method and device

Technical Field

The present application relates to the field of communications technologies, and in particular, to a communications method and apparatus.

Background

In general, an ASIC chip dedicated to handling service message forwarding is included in the switch, and the handling procedure of the ASIC chip for handling service messages is as follows: analyzing fields such as message header data, MAC/IP address, protocol type and the like; after looking up a table according to the processing logic of the ASIC chip, forwarding decision information, for example, a message processing priority, performing two-layer forwarding or three-layer forwarding, etc., is generated. The above-mentioned forwarding decision information is intermediate information, and is generally referred to as a "descriptor". In the multi-chip exchange system, the descriptor and the data message are sent to the downstream processing chip together to further guide the table look-up and forwarding of the downstream processing chip.

In current mainstream frame switches, distributed forwarding architecture is commonly employed. That is, the ASIC chip is configured within the line card. The line cards are connected through the switching network, and the switching chip for executing the message processing of the switching network is configured in the main control board or the independent network board. When processing service message forwarding, the switching chip can forward the service message to the destination line card according to the descriptor sent by the source line card.

As shown in fig. 1, fig. 1 is a schematic flow chart of forwarding service messages by using a conventional line card and a switch network board. In fig. 1, after a service packet enters a Line Card (LC) 0, a corresponding table look-up process is performed to generate a descriptor. LC0 sends descriptors and service messages to the switch board (english: switch Fabric Card, abbreviated as SFC). Because the ports connected with the SFCs and the LCs are subjected to speed limiting processing, the ports are generally aggregated, and the service messages are shared among member ports of an aggregation group, so that the service messages are prevented from being sent to one SFC in a concentrated manner, and the corresponding ports are prevented from being congested.

In the process that the LC sends the descriptor and the service message to the SFC, the LC selects one member port from a plurality of member ports to execute forwarding according to a hash routing rule. Under some conditions, hash routing is performed according to five-tuple information of the service message, and when the service flow characteristics are single, the situation that the hash routing is seriously uneven, namely congestion and packet loss occur in a certain member port, can occur; while the other member ports are idle.

Aiming at the problems, a dynamic load balancing (English: dynamic Load Balancing, abbreviated as DLB) technology is generally adopted at present to overcome the problems. The LC dynamically selects and records the member ports with the lightest load by monitoring the bandwidth occupation condition of each member port, fully utilizes the bandwidth of the member ports, and avoids the phenomenon that a certain member port is congested.

However, existing DLB technology also exposes the following drawbacks: 1) After the DLB flow table is aged, load balancing cannot be performed again, so that the phenomenon of flow imbalance occurs again; 2) In the process of realizing load balancing, the uplink LC only locally performs load balancing, but cannot know the bandwidth occupation condition of a member port in the aggregation group between the SFC and the downlink LC, so that congestion and packet loss occur in a member port in the aggregation group between the SFC and the downlink LC, and other member ports are idle.

Disclosure of Invention

In view of this, the present application provides a communication method and apparatus, which are used to solve the problem that in the existing DLB technology, after the DLB flow table is aged, the phenomenon of flow imbalance occurs again, and the uplink LCs implement load balancing respectively, so that congestion and packet loss occur at member interfaces between the SFC and the downlink LCs.

In a first aspect, the present application provides a communication method, where the method is applied to a first LC, the first LC has transmitted a first service packet to a first SFC through a member port in a first aggregation group, and the first SFC forwards the first service packet to a second LC, and the method includes:

Receiving a second service message, wherein the second service message and the first service message belong to different service flows;

when the line card receiving the second service message is determined to be the second LC, selecting a second SFC connected with the second LC by using a global SFC DLB table, wherein the second SFC is the SFC with the minimum current bearing service flow in all SFCs connected with the second LC;

acquiring a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in a flow setting table;

selecting a first member port from the member ports according to the first current load state and the second current load state;

and sending the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

In a second aspect, the present application provides a communication apparatus, the apparatus being applied to a first LC, the first LC having transmitted a first service packet to a first SFC through a member port in a first aggregation group, the first SFC forwarding the first service packet to a second LC, the apparatus comprising:

the receiving unit is used for receiving a second service message, and the second service message and the first service message belong to different service flows;

The first selecting unit is configured to select, when it is determined that the line card receiving the second service packet is the second LC, a second SFC connected to the second LC by using a global SFC DLB table, where the second SFC is an SFC with a minimum current bearer service flow among all SFCs connected to the second LC;

an obtaining unit, configured to obtain a first current load state of each member port in the first aggregation group and a second current load state of each flow table entry in the flow setting table;

a second selecting unit, configured to select a first member port from the member ports according to the first current load state and the second current load state;

and the sending unit is used for sending the second service message to the second SFC through the first member port so that the second SFC forwards the second service message to the second LC.

In a third aspect, the application provides a network device comprising a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor to cause the processor to perform the method provided by the first aspect of the application.

Therefore, by applying the communication method and the device provided by the application, the first LC receives the second service message, and the second service message and the first service message belong to different service flows; when the line card receiving the second service message is determined to be the second LC, the first LC selects a second SFC connected with the second LC by using a global SFC DLB table, wherein the second SFC is the SFC with the minimum current bearing service flow in all SFCs connected with the second LC; the first LC acquires a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in the flow setting table; according to the first current load state and the second current load state, the first LC selects a first member port from member ports; and the first LC sends the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

In this way, by setting the multi-stage DLB structure, in the process of forwarding the service message, a primary DLB is utilized to select a proper SFC, a secondary DLB is utilized to select a proper forwarding interface, and finally, the service message is forwarded to the selected SFC through the forwarding interface. The method solves the problems that in the existing DLB technology, after the DLB flow table is aged, the phenomenon of unbalanced flow occurs again, and the uplink LCs respectively realize load balancing, so that congestion and packet loss occur at member interfaces between SFCs and the downlink LCs, and the maximum utilization of the performance of the switching network is realized.

Drawings

FIG. 1 is a schematic flow chart of forwarding service messages by using a conventional line card board and a switch board;

FIG. 2 is a flow chart of a communication method according to an embodiment of the present application;

fig. 3 is a schematic flow chart of forwarding service packets by a line card board and a switch board according to an embodiment of the present application;

fig. 4 is a schematic flow chart of forwarding a service message by another line card board and a switch board according to an embodiment of the present application;

fig. 5 is a block diagram of a communication device according to an embodiment of the present application;

fig. 6 is a hardware structure of a network device according to an embodiment of the present application.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the corresponding listed items.

It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the application. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.

The communication method provided by the embodiment of the application is described in detail below. Referring to fig. 2, fig. 2 is a flowchart of a communication method according to an embodiment of the present application. The method is applied to the first LC, and the communication method provided by the embodiment of the present application may include the following steps.

Step 210, receiving a second service message, where the second service message and the first service message belong to different service flows;

specifically, the first LC may be connected with the first SFC through a plurality of ports. The plurality of ports form a first aggregation group, and the plurality of ports are member ports in the first aggregation group. Similarly, the plurality of ports in the first SFC connected to the first LC also form a second aggregation group, and the plurality of ports are member ports in the second aggregation group.

After the first LC receives the first service packet, the first LC calculates a first hash value corresponding to the first service packet according to quintuple information included in the first service packet, and determines a service flow to which the first service packet belongs, for example, the first hash value is 1, the service flow belongs to flow1, and the flow rate is 1G.

Meanwhile, the first LC performs table lookup forwarding processing through the destination address included in the quintuple information. When the first LC determines that the destination line card for receiving the first service message at the next hop is the second LC, the first LC selects SFC connected with the second LC by using a local global SFC DLB table. For example, the first LC selects a first SFC connected to the second LC, where the first SFC is the SFC with the smallest current bearer traffic among all SFCs connected to the second LC.

In the global SFC DLB table, a current load state, which is also specifically indicated by a flow rate, when each SFC connected to the first LC is connected to the destination LC (i.e., the next hop LC of the first LC) is stored.

The first LC locally establishes a first DLB flow entry and stores the first DLB flow entry into the DLB flow table. The first DLB flow entry includes a flow identification, an SFC identification, and a first LC identification.

It will be appreciated that the first DLB flow entry also includes a flow rate; the flow rate may also be embodied as the current load state of the SFC. The first LC may also update the current load status included in the corresponding global SFC DLB table according to the flow rate included in the first DLB flow entry. Of course, the current load status included in the global SFC DLB table may also be updated by a notification message periodically sent by the SFC, and this updating manner is described in the following embodiments.

The first LC obtains a first current load status of each member port in the first aggregation group, and a second current load status of each flow entry in the local flow setting table. In the embodiment of the present application, the first service packet is a packet of the first service flow received by the first LC, so that the first current load state of each member port in the first aggregation group is 0, the flow setting table also does not include a flow table entry, and the second current load state is 0. The first LC may randomly select one member port from the plurality of member ports as an egress port for forwarding the first service packet. For example, the first LC selects the second member port as the egress port for forwarding the first service packet.

The first LC locally builds a flow entry including a flow identification, a flow rate, and a member port identification, and stores the flow entry in a flow setup table. The flow table entry may also include an LC identification, and the flow rate may also be specifically expressed as the current load status of the member port.

And through the selected second member port, the first LC sends the first service message to the first SFC, so that the first SFC forwards the first service message to the second LC.

It may be appreciated that the first LC sends the first service message to the first SFC and simultaneously sends a descriptor for forwarding the first service message to the first SFC. And forwarding the first service message to the second LC by the first SFC according to the descriptor.

It should be noted that, the first SFC is also connected to the second LC through a plurality of member ports included in the aggregation group, and when the first SFC selects a member port to forward the first service message to the second LC, the first SFC may also select, according to the DLB technology, the member port with the smallest current bearer service flow to forward the first service message.

After the first LC finishes the forwarding processing of the first service message, the first LC receives the second service message again. The first LC calculates a second hash value corresponding to the second service message through five-tuple information included in the second service message, and determines a service flow to which the second service message belongs, for example, the second hash value is 2, the service flow is flow2, and the flow rate is 1G.

220, when it is determined that the line card receiving the second service packet is the second LC, selecting a second SFC connected to the second LC by using a global SFC DLB table, where the second SFC is an SFC with the smallest current bearer service flow among all SFCs connected to the second LC;

specifically, according to the description of step 210, after the first LC receives the second service packet, the first LC performs the table lookup forwarding process through the destination address included in the quintuple information. When the first LC determines that the destination line card for receiving the second service message at the next hop is the second LC, the first LC selects SFC connected with the second LC by using a local global SFC DLB table. For example, the first LC selects a second SFC connected to the second LC, where the second SFC is the SFC with the smallest current bearer traffic among all SFCs connected to the second LC.

Optionally, the specific process of selecting the second SFC connected to the second LC by using the global SFC DLB table by the first LC is: from the global SFC DLB table, the first LC selects a member port with the smallest current load service flow of the member ports in the second aggregation group; and the first LC takes the SFC of the member port with the smallest current load service flow as a second SFC.

In the embodiment of the application, the SFC stored in the global SFC DLB table is connected with the current load state of the destination LC (namely, the next hop LC of the first LC). The current load state is the current load state of each member port included in the second aggregation group in the SFC. Thus, through the global SFC DLB table, the first LC can select the member port with the smallest current load service flow, and further select the SFC with the member port with the smallest current load service flow as the switching network board for forwarding to the target LC. Meanwhile, the load balancing of the SFC and the target LC is realized by selecting according to the current load state of the port on the SFC.

Optionally, after the first LC selects the second SFC, establishing a second DLB flow table entry is further performed, where the second DLB flow table entry includes a flow identifier, an SFC identifier, and a first LC identifier. It will be appreciated that the second DLB flow entry also includes a flow rate; the flow rate may also be embodied as the current load state of the SFC.

It will be appreciated that from the descriptions of steps 210 and 220, two DLB flow entries are already included in the DLB flow table. A DLB flow table entry is used to indicate a first SFC selected by the first LC when forwarding the first service packet; the other DLB flow table entry is used to indicate the second SFC selected by the first LC when forwarding the second service packet.

Step 230, obtaining a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in the flow setting table;

specifically, according to the description of step 220, after the first LC selects the second SFC as the switching fabric with the second LC by using the global SFC DLB table, the first LC needs to select a member port from the first aggregation group as a port for forwarding the second service packet to the second SFC.

The first LC obtains a first current load state of each member port in the first aggregation group and a second current load state of each flow entry in the local flow setting table.

Step 240, selecting a first member port from the member ports according to the first current load state and the second current load state;

specifically, as can be seen from the description of step 230, after the first LC acquires the first current load state and the second current load state, the first member port is selected from the member ports of the first aggregation group. The first member port is an output port for forwarding the second service message.

As can be seen from the description of step 210, when forwarding the first service packet, the first LC selects the second member port from the first aggregation group as the output port for forwarding the first service packet, and establishes a flow table entry in the flow setting table. At this time, the first current load state obtained by the first LC includes the traffic currently carried by the second member port, that is, the flow rate of the traffic message carried by the second member port, for example, the flow rate is 1G, while the other member ports do not carry the traffic, that is, the flow rate of the other member ports is 0.

The second current load state obtained by the first LC is the traffic flow currently carried by the flow table entry established by the first LC in step 210, i.e. the flow rate of the traffic message carried by the second member port, for example, the flow rate is 1G.

Optionally, in one implementation, according to the first current load state and the second current load state, the specific process of selecting the first member port from the member ports by the first LC is: if the first LC determines that the idle member port exists in the member ports according to the first current load state and the second current load state, the first LC randomly selects one idle member port as the first member port.

Optionally, in another implementation, according to the first current load state and the second current load state, a specific process of selecting the first member port from the member ports by the first LC is: if the first LC determines that the idle member port does not exist in the member ports according to the first current load state and the second current load state, the first LC calculates the total traffic currently carried by the first LC; calculating the current service shunt volume carried by each member port according to the total service flow and the number of the member ports in the first aggregation group; according to the service shunt quantity, the first LC selects a first member port from the member ports, and updates the flow table items of the first member port and the flow table items of the other member ports except the first member port so as to enable the load sharing among the member ports to be met.

Optionally, after the first LC randomly selects one of the idle member ports as the first member port, or after the first LC selects the first member port from the member ports, the first LC further performs establishing a flow table entry, and stores the flow table entry in the flow setting table, where the flow table entry includes a flow identifier, a flow rate, and a member port identifier.

In the following embodiments, a process of selecting the first member port by the first LC and establishing the flow table entry will be specifically described, and will not be repeated here.

Step 250, sending the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

Specifically, as can be seen from the description of step 240, after the first LC selects the first member port, the second service message is sent to the second SFC through the first member port.

It is understood that the first LC sends the second service message to the second SFC and simultaneously sends the descriptor for forwarding the second service message to the second SFC. And forwarding the second service message to the second LC by the second SFC according to the descriptor.

It should be noted that, the second SFC is also connected to the second LC through a plurality of member ports included in the aggregation group, and when the second SFC selects a member port to forward the second service packet to the second LC, the second SFC may also select, according to the DLB technology, the member port with the smallest current bearer service flow to forward the second service packet.

Therefore, by applying the communication method provided by the application, the first LC receives the second service message, and the second service message and the first service message belong to different service flows; when the line card receiving the second service message is determined to be the second LC, the first LC selects a second SFC connected with the second LC by using a global SFC DLB table, wherein the second SFC is the SFC with the minimum current bearing service flow in all SFCs connected with the second LC; the first LC acquires a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in the flow setting table; according to the first current load state and the second current load state, the first LC selects a first member port from member ports; and the first LC sends the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

Optionally, in the embodiment of the present application, the method further includes a process that the first LC receives the SFC to send a notification message, and establishes a global SFC DLB table according to the notification message.

Specifically, each SFC connected to the first LC periodically generates a notification message, where each notification message includes a current load status of each member port in the second aggregation group used by the SFC to connect to each destination LC and a corresponding LC identification.

Each SFC sends a notification message to the first LC, respectively. After the first LC receives the plurality of notification messages, the current load status of each member port in the second aggregation group used by the SFC for connection with each destination LC and the corresponding LC identification are obtained from each notification message.

The first LC may establish a plurality of global SFC DLB entries, each of which stores a current load status of one member port in the second polymer group and a corresponding LC identification. The first LC stores the established plurality of global SFC DLB entries into a global SFC DLB table.

Since the SFC is periodically sent, the first LC may correspondingly update the previously established global SFC DLB entry according to the currently received notification message.

It can be appreciated that the global SFC DLB entry also includes an SFC identifier and a member port identifier; the current load state may also be expressed in terms of flow rate.

The communication method provided by the embodiment of the application is described in detail below. Referring to fig. 3, fig. 3 is a schematic flow chart of forwarding a service packet by a line card board and a switch board according to an embodiment of the present application.

In fig. 3, a plurality of LCs are included in the switch, and LC1 and LC2 are illustrated as examples. The switch also includes a plurality of SFCs, such as SFC1 and SFC 2. Each LC comprises an aggregation group and a plurality of member ports are included in the aggregation group; each SFC also includes an aggregation group and a plurality of member ports are included within the aggregation group. For example, LC1 comprises a first aggregation group; LC2 includes a second polymeric group; SFC1 includes a third polymerization group; SFC2 includes a fourth aggregation group.

The first aggregation group and the third aggregation group are assumed to comprise a member port 1 and a member port 2, and the member ports with corresponding numbers are in communication connection with each other; the second polymerization group and the third polymerization group comprise a member port 3 and a member port 4, and the member ports with corresponding numbers are in communication connection with each other. Similarly, it is assumed that the first aggregation group and the fourth aggregation group both comprise a member port 5 and a member port 6, and the member ports with corresponding numbers are in communication connection with each other; the second aggregation group and the fourth aggregation group comprise member ports 7 and member ports 8, and the member ports with corresponding numbers are in communication connection with each other.

After the LC1 receives the service packet 1, the LC1 calculates the hash value 1 corresponding to the service packet through five-tuple information included in the service packet 1, and determines a service flow to which the service packet 1 belongs. For example, the hash value 1 is 1, the flow of the service is flow1, and the flow rate is 1G. Meanwhile, LC1 performs table lookup forwarding processing through the destination address included in the quintuple information. When the LC1 determines that the destination line card of the next hop receiving service message 1 is LC2, the LC1 searches the global SFC DLB table. Global SFC DLB table is shown in table 1 below.

TABLE 1 Global SFC DLB Table

SFC identification	Member port identification	Current load status (flow rate)	LC identification
				SFC1	Member mouth 3	0	LC2
SFC1	Member mouth 4	0	LC2
				SFC2	Member mouth 7	0	LC2
SFC2	Member mouth 8	0	LC2

Through the global SFC DLB table, the LC1 determines that all member ports included in a third aggregation group and a fourth aggregation group which are included in SFC1 and SFC2 connected with the LC2 do not bear service traffic currently, namely, all member ports are idle ports, and neither SFC1 nor SFC2 bear service traffic currently, and are idle screen boards.

At this time, LC1 may randomly select one SFC as the next hop for forwarding service packet 1. For example, LC1 selects SFC1 to forward service message 1.

LC1 selects SFC1 and then locally builds a DLB flow table. The DLB flow table is shown in table 2 below.

TABLE 2DLB flow Table

SFC identification	Flow identification	Current load status (flow rate)	LC identification
				SFC1	flow1	1	LC2

Because LC1 does not select SFC2 to forward service message 1, the DLB flow table does not include the DLB flow entry corresponding to SFC 2.

LC1 obtains a first current load status of the member port 1 and the member port 2 in the first aggregation group and a second current load status of each flow entry in the local flow setting table.

In the embodiment of the present application, the service packet 1 is a packet of the first service flow received by the LC1, so the first current load states of the member port 1 and the member port 2 are both 0, the flow setting table also does not include a flow table entry, and the second current load state is also 0.LC1 may randomly select one member port from member ports 1 and 2 as an output port for forwarding service packet 1. For example, LC1 selects member port 2 as the egress interface for forwarding service message 1.

LC1 creates a flow entry locally and stores the flow entry in a flow setup table. The flow setting table is shown in table 3 below.

Table 3 flow set table

LC identification	Flow identification	Current load status (flow rate)	Member port identification
				LC1	flow1	1	Member mouth 2

And sending the service message 1 to the SFC1 through the selected member port 2 by the LC1, so that the SFC1 forwards the service message 1 to the LC 2.

It can be understood that LC1 sends a descriptor for forwarding service packet 1 to SFC1, along with service packet 1 to SFC 1. According to the descriptor, SFC1 forwards service message 1 to LC 2.

When the SFC1 selects the member port to forward the service message 1 to the LC2, the member port with the minimum current bearing service flow can be selected to forward the service message 1 according to the DLB technology.

In the embodiment of the application, both SFC1 and SFC2 generate and send notification messages to LC1 periodically. The notification message includes the current load status of each member port that SFC1, SFC2 is connected to LC2, and the LC2 identification. Before LC1 receives service packet 1, the first notification message sent by SFC1 and SFC2 to LC1 may be used to enable LC1 to build the global SFC DLB table shown in table 1.

SFC1 may also generate and send a second notification message to LC1 after forwarding service message 1, for LC1 to update the global SFC DLB table shown in Table 1 above. The updated global SFC DLB table is shown in Table 4 below.

TABLE 4 Global SFC DLB Table updated

SFC identification	Member port identification	Current load status (flow rate)	LC identification
				SFC1	Member mouth 3	1	LC2
SFC1	Member mouth 4	0	LC2
				SFC2	Member mouth 7	0	LC2
SFC2	Member mouth 8	0	LC2

As can be seen from table 4, SFC1 selects member port 3 to forward service message 1 to LC2, while the other member ports are idle ports.

Similarly, SFC2 may also periodically generate and send a second notification message to LC1 for LC1 to update the global SFC DLB table shown in Table 1 above. The updated global SFC DLB table is shown in Table 4 above. As can be seen from table 4, all the member ports included in SFC2 are idle ports.

After the LC1 completes the forwarding process of the service packet 1, the LC1 receives the service packet 2 again. LC1 calculates hash value 2 corresponding to service message 2 through five-tuple information included in service message 2, and determines service flow to which service message 2 belongs, for example, hash value 2 is 2, the service flow belongs to flow2, and flow rate is 1G. Meanwhile, LC1 performs table lookup forwarding processing through the destination address included in the quintuple information. When the LC1 determines that the destination line card of the next hop receiving service message 2 is LC2, the LC1 searches the global SFC DLB table. Global SFC DLB tables are shown in table 4 above.

Through the global SFC DLB table, the LC1 determines that all member ports except the member port 3 do not currently bear the service flow in the third aggregation group and the fourth aggregation group which are included in the SFC1 and the SFC2 connected with the LC2, namely, all member ports except the member port 3 are idle ports, and the SFC1 currently bears the service flow; SFC2 does not bear traffic currently, is idle screen.

At this time, in order to make each SFC satisfy load balancing, LC1 selects SFC2 as a switching fabric for forwarding service packet 2. LC1 updates the locally established DLB flow table after selecting SFC. The updated DLB flow table is shown in table 5 below.

TABLE 5 updated DLB flow Table

SFC identification	Flow identification	Current load status (flow rate)	LC identification
				SFC1	flow1	1	LC2
SFC2	Flow2	1	LC2

In the embodiment of the present application, when the LC1 forwards the service packet 1, the member port 2 is selected from the first aggregation group as the output port for forwarding the service packet 1, and a flow table entry is established in the flow setting table. At this time, the first current load state acquired by the LC1 includes the current traffic carried by the member port 2, that is, the traffic rate of the traffic message carried by the member port 2 is 1G, while the traffic is not carried by the member port 1, that is, the traffic rate of the member port 1 is 0.

The second current load state obtained by LC1 is the traffic flow currently carried by the flow table entry in table 3, that is, the flow rate of the traffic message carried by the member port 2 is 1G. According to the first current load state and the second current load state, the LC1 determines that the member port 1 in the first aggregation group is an idle member port. LC1 selects member port 1 as the outgoing interface for forwarding service message 2.

LC1 re-establishes the flow entry locally and stores the flow entry in the flow setup table. The updated flow setting table is shown in table 6 below.

Table 6 updated flow set table

LC identification	Flow identification	Current load status (flow rate)	Member port identification
				LC1	flow1	1	Member mouth 2
LC1	Flow2	1	Member's mouth 1

And sending the service message 2 to the SFC2 through the selected member port 1 by the LC1, so that the SFC2 forwards the service message 2 to the LC 2.

It will be appreciated that LC1 sends a descriptor for forwarding service message 2 to SFC2, along with service message 2 to SFC 2. According to the descriptor, SFC2 forwards service message 2 to LC 2.

When the SFC2 selects the member port to forward the service message 2 to the LC2, the member port with the minimum current bearing service flow can be selected to forward the service message 2 according to the DLB technology.

SFC2 may also generate and send a third notification message to LC1 after forwarding service message 2, for LC1 to update the global SFC DLB table shown in Table 4 above. The updated global SFC DLB table is shown in Table 7 below.

Table 7 updated global SFC DLB table

SFC identification	Member port identification	Current load status (flow rate)	LC identification
				SFC1	Member mouth 3	1	LC2
SFC1	Member mouth 4	0	LC2
				SFC2	Member mouth 7	1	LC2
SFC2	Member mouth 8	0	LC2

As can be seen from table 7, SFC2 selects member port 7 to forward service message 2 to LC2, while the other member ports are idle ports.

Similarly, SFC1 may also periodically generate and send a third notification message to LC1 for LC1 to update the global SFC DLB table shown in Table 4 above. The updated global SFC DLB table is shown in Table 7 above. As can be seen from table 7, the member port 3 included in the SFC1 carries traffic, and the member port 4 is an idle port.

Two different scenarios are described below.

In one scenario, in fig. 3, the forwarding description of the service packet is performed by including LC1 and LC2 in the switch, and in practical application, the switch includes a plurality of LCs, for example: LC1, LC2 and LC3. As shown in fig. 4, fig. 4 is a schematic flow chart of forwarding a service packet by another line card board and a switch board according to an embodiment of the present application.

In fig. 4, each LC is connected to the SFC through a member port included in the aggregation group. When the LC1 receives the service packet 1 and determines that the destination line card for receiving the service packet 1 in the next hop is LC3, the LC1 selects SFC1 as the switch board for forwarding the service packet 1. The selection process of the SFC and the selection process of the local outbound interface are similar to the process of the embodiment shown in fig. 3 and will not be repeated here.

While in receiving service message 2, the process is different from that shown in fig. 3: LC2 receives service message 2. In the embodiment of the application, when the LC2 receives the service message 2 and determines that the destination line card of the next hop for receiving the service message 2 is LC3, the LC2 also searches the local global SFC DLB table.

As can be seen from the foregoing embodiment shown in fig. 3, each SFC will send a first notification message to each LC, so that each LC locally builds the global SFC DLB table shown in table 1. At the same time, each SFC will send a second notification message to each LC to cause each LC to update the global SFC DLB table shown in Table 1 above.

In fig. 4, the global SFC DLB table shown in table 1 is also established in LC2, and the global SFC DLB table shown in table 1 is updated to table 4 according to the second notification message sent subsequently to SFC 1.

In this way, when the LC2 selects the SFC, it can be seen from table 4 that the SFC2 does not currently carry traffic, and is an idle mesh plate.

At this time, in order to make each SFC satisfy the load balancing, LC2 selects SFC2 as the next hop for forwarding service packet 2. After LC1 selects SFC, the same steps as those of LC1 in the embodiment shown in fig. 3 described above are performed, and will not be repeated here.

It will be appreciated that each LC builds a DLB flow table locally. If each SFC has carried service flow, at this time, the LC can select proper SFC according to whether the member port carries service flow; if each member port included in each SFC also carries traffic, at this time, the LC may select an appropriate SFC according to the calculated sum of the traffic carried by the member ports.

As can be seen from fig. 4, when the target LCs are the same and different source LCs forward the service message, appropriate SFCs are selected through the global SFC DLB tables established locally, so that each SFC satisfies load balancing.

In another scenario, in fig. 3, LC1 receives service message 3 again after completing the forwarding process of service message 2. LC1 calculates hash value 3 corresponding to service message 3 through five-tuple information included in service message 3, and determines service flow to which service message 3 belongs, for example, hash value 2 is 3, and service flow to which service message 3 belongs is flow3. In this scenario, the flow3 flow rate is 2G.

LC1 may be selected with reference to the embodiment shown in fig. 3 described above in selecting SFC, and will not be repeated here. When the LC1 selects the output interface for forwarding the service packet 3 to the SFC, it can be known from the calculated first current load state and the second current load state that the current load states of the member port 1 and the member port 2 are 1G, and the current load state of each flow table entry is 1G.

If the LC1 forwards the service message 3 through one member port, the load of each member port will be unbalanced. Therefore, in order to satisfy the load balancing of each member port, LC1 calculates the total traffic flow of the current service to be 4G, and the number of member ports to be 2, that is, the traffic flow of each member port to be loaded to be 2G.

On this basis, LC1 updates the member port identifier of flow1 in table 6 as member port 1, i.e., member port 1 carries both flows 1 and 2.LC1 selects member port 2 as the outgoing interface for forwarding service message 3.

LC1 re-establishes the flow entry locally and stores the flow entry in the flow setup table. The updated flow setting table is shown in table 8 below.

Table 8 updated flow set table

LC identification	Flow identification	Current load status (flow rate)	Member port identification
				LC1	flow1	1	Member's mouth 1
LC1	Flow2	1	Member's mouth 1
				LC1	Flow3	2	Member mouth 2

And sending the service message 3 to the SFC2 through the selected member port 2 by the LC1, so that the SFC2 forwards the service message 3 to the LC 2.

Based on the same inventive concept, the embodiment of the application also provides a communication device corresponding to the communication method. Referring to fig. 5, fig. 5 is a communication device provided by an embodiment of the present application, where the device is applied to a first LC, the first LC has transmitted a first service packet to a first SFC through a member port in a first aggregation group, and the first SFC forwards the first service packet to a second LC, and the device includes:

a receiving unit 510, configured to receive a second service packet, where the second service packet and the first service packet belong to different service flows;

a first selecting unit 520, configured to select, when it is determined that the line card receiving the second service packet is the second LC, a second SFC connected to the second LC by using a global SFC DLB table, where the second SFC is an SFC with a minimum current bearer service flow among all SFCs connected to the second LC;

An obtaining unit 530, configured to obtain a first current load state of each member port in the first aggregation group and a second current load state of each flow table entry in the flow setting table;

a second selecting unit 540, configured to select a first member port from the member ports according to the first current load state and the second current load state;

and a sending unit 550, configured to send the second service packet to the second SFC through the first member port, so that the second SFC forwards the second service packet to the second LC.

Optionally, the receiving unit 510 is further configured to receive a notification message sent by each SFC, where the notification message includes a current load status of each member port in the second aggregation group used by the SFC to connect with each destination LC and the corresponding LC identifier;

the apparatus further comprises: a first establishing unit (not shown in the figure) is configured to establish the global SFC DLB table, where the global SFC DLB table includes a current load status of each member port in the second aggregation group and the corresponding LC identifier.

Optionally, the first selecting unit 520 is specifically configured to select, from the global SFC DLB table, a member port with the smallest current load traffic flow of the member ports in the second aggregation group;

And taking the SFC of the member port with the minimum current load service flow as the second SFC.

Optionally, the apparatus further comprises:

a second establishing unit (not shown in the figure) is configured to establish a DLB flow table, where the DLB flow table includes a flow identifier, an SFC identifier, and the first LC identifier.

Optionally, the second selecting unit 540 is specifically configured to, if it is determined that an idle member port exists in the member ports according to the first current load state and the second current load state, randomly select one idle member port as the first member port.

Optionally, the second selecting unit 540 is specifically configured to calculate the total traffic currently carried by the first LC if it is determined that no idle member port exists in the member ports according to the first current load state and the second current load state;

calculating the current service split flow carried by each member port according to the total service flow and the number of the member ports in the first aggregation group;

and selecting the first member port from the member ports according to the service shunt quantity, and updating the flow table item where the first member port is located and the flow table items where other member ports except the first member port are located so that each member port meets load sharing.

Optionally, the apparatus further comprises:

a third establishing unit (not shown) for establishing the flow table item including flow identifier, flow rate and member port identifier

Therefore, by applying the communication device provided by the application, the first LC receives the second service message, and the second service message and the first service message belong to different service flows; when the line card receiving the second service message is determined to be the second LC, the first LC selects a second SFC connected with the second LC by using a global SFC DLB table, wherein the second SFC is the SFC with the minimum current bearing service flow in all SFCs connected with the second LC; the first LC acquires a first current load state of each member port in the first aggregation group and a second current load state of each flow table item in the flow setting table; according to the first current load state and the second current load state, the first LC selects a first member port from member ports; and the first LC sends the second service message to the second SFC through the first member port, so that the second SFC forwards the second service message to the second LC.

Based on the same inventive concept, the embodiment of the present application further provides a network device, as shown in fig. 6, including a processor 610, a transceiver 620, and a machine-readable storage medium 630, where the machine-readable storage medium 630 stores machine-executable instructions capable of being executed by the processor 610, and the processor 610 is caused to perform the communication method provided by the embodiment of the present application. The communication device shown in fig. 5 may be implemented by using a hardware structure of a network device as shown in fig. 6.

The computer readable storage medium 630 may include a random access Memory (in english: random Access Memory, abbreviated as RAM) or a nonvolatile Memory (in english: non-volatile Memory, abbreviated as NVM), such as at least one magnetic disk Memory. Optionally, the computer readable storage medium 630 may also be at least one storage device located remotely from the aforementioned processor 610.

The processor 610 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; it may also be a digital signal processor (English: digital Signal Processor; DSP; for short), an application specific integrated circuit (English: application Specific Integrated Circuit; ASIC; for short), a Field programmable gate array (English: field-Programmable Gate Array; FPGA; for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

In an embodiment of the present application, processor 610, by reading machine-executable instructions stored in machine-readable storage medium 630, is caused by the machine-executable instructions to implement processor 610 itself and invoke transceiver 620 to perform the communication methods described in the previous embodiments of the present application.

Additionally, embodiments of the present application provide a machine-readable storage medium 630, the machine-readable storage medium 630 storing machine-executable instructions that, when invoked and executed by the processor 610, cause the processor 610 itself and the invoking transceiver 620 to perform the communication methods described in the foregoing embodiments of the present application.

The implementation process of the functions and roles of each unit in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.

For the device embodiments, reference is made to the description of the method embodiments for the relevant points, since they essentially correspond to the method embodiments. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purposes of the present application. Those of ordinary skill in the art will understand and implement the present application without undue burden.

For the communication device and the machine-readable storage medium embodiments, since the method content involved is substantially similar to the method embodiments described above, the description is relatively simple, and reference will only be made to part of the description of the method embodiments.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather to enable any modification, equivalent replacement, improvement or the like to be made within the spirit and principles of the application.

Claims

1. A communication method, wherein the method is applied to a first LC, the first LC having transmitted a first service message to a first SFC through a member port in a first aggregation group, the first SFC forwarding the first service message to a second LC, the method comprising:

2. The method of claim 1, wherein prior to receiving the second service message, the method further comprises:

receiving a notification message sent by each SFC, wherein the notification message comprises the current load state of each member port in a second aggregation group connected with each target LC by the SFC and the corresponding LC identifier;

and establishing the global SFC DLB table, wherein the global SFC DLB table comprises the current load state of each member port in the second aggregation group and the corresponding LC identification.

3. The method of claim 2, wherein selecting a second SFC for connection with the second LC using a global SFC DLB table, in particular comprises:

Selecting a member port with the smallest current load service flow of the member ports in the second aggregation group from the global SFC DLB table;

4. The method of claim 1, wherein after the selecting the second SFC coupled to the second LC, the method further comprises:

and establishing a DLB flow table, wherein the DLB flow table comprises a flow identifier, an SFC identifier and the first LC identifier.

5. The method according to claim 1, wherein selecting a first member port from the member ports according to the first current load state and the second current load state, comprises:

and if the idle member port exists in the member ports according to the first current load state and the second current load state, randomly selecting one idle member port as the first member port.

6. The method according to claim 1, wherein selecting a first member port from the member ports according to the first current load state and the second current load state, comprises:

if the fact that the idle member ports do not exist in the member ports is determined according to the first current load state and the second current load state, calculating the total traffic currently carried by the first LC;

7. The method according to claim 5 or 6, wherein after said randomly selecting one free member port as said first member port or after said selecting said first member port from member ports, the method further comprises:

and establishing the flow table item, wherein the flow table item comprises a flow identifier, a flow rate and a member port identifier.

8. A communication apparatus, the apparatus being applied to a first LC that has transmitted a first service message to a first SFC through a first intra-aggregation-group member port, the first SFC forwarding the first service message to a second LC, the apparatus comprising:

9. The apparatus of claim 8, wherein the receiving unit is further configured to receive a notification message sent by each SFC, the notification message including a current load status of each member port in the second aggregation group used by the SFC for connection with each destination LC and the corresponding LC identification;

The apparatus further comprises: a first establishing unit, configured to establish the global SFC DLB table, where the global SFC DLB table includes a current load status of each member port in the second aggregation group and the corresponding LC identifier.

10. The apparatus of claim 9, wherein the first selecting unit is specifically configured to select, from the global SFC DLB table, a member port with a smallest current load traffic flow for the member port in the second aggregation group;

11. The apparatus of claim 8, wherein the apparatus further comprises:

and the second establishing unit is used for establishing a DLB flow table, wherein the DLB flow table comprises a flow identifier, an SFC identifier and the first LC identifier.

12. The apparatus of claim 8, wherein the second selecting unit is specifically configured to randomly select one of the idle member ports as the first member port if it is determined that the idle member port exists in the member ports according to the first current load state and the second current load state.

13. The apparatus of claim 8, wherein the second selecting unit is specifically configured to calculate a total traffic currently carried by the first LC if it is determined that no idle member port exists in the member ports according to the first current load state and the second current load state;

14. The apparatus according to claim 12 or 13, characterized in that the apparatus further comprises:

and the third establishing unit is used for establishing the flow table item, wherein the flow table item comprises a flow identifier, a flow rate and a member port identifier.