WO2015039616A1 - Method and device for packet processing - Google Patents

Abstract

Disclosed are a method and device for packet processing. A same TCN is preconfigured for packets in a same packet stream. The TCN and a value of the number of service nodes in a hop are utilized for a load-balancing computation. If suitable TCNs can be selected respectively for different packet streams on the basis of the number of packet streams, sharing of packet streams among service nodes can be implemented. When the number of packet streams is stable, regardless of that the topology states of the service nodes change, load balancing for the service nodes in a same hop can be implemented in a statistical sense.

Description

Message processing method and device

The present application claims priority to Chinese Patent Application No. CN201310430814.4, entitled "A Message Processing Method and Apparatus", filed on September 18, 2013, the entire contents of which are incorporated by reference. In this application.

Technical field

The present invention relates to the field of wireless communications, and in particular, to a packet processing method and device.

Background technique

The service node with the network service function can process the network packet, and the service node can be a firewall, a transmission control protocol (English: Transmission Control Protocol, abbreviation: TCP), an acceleration node or a network address translation (English network address translation, abbreviation NAT) nodes, etc. Service nodes include the following types of network service functions: security, filtering, statistical monitoring, billing, and traffic acceleration. The service node with the network service function is deployed on the path through which the packet flows, and the service node performs the corresponding network service function for each packet in the packet flow. The tuple of the message in the same packet stream (English: tuple), such as the quintuple (English: quintuple), has the same value.

It is assumed that the network service function of the service node B1 is called the B network service function, and the service node B2 with the B network service function is also deployed in the network structure of the service node, and the service node B1 and the service node B2 form the service node network. One hop, that is, one hop includes at least one service node: a service node having the same network service function in the network of the service node, any one of which can provide a message in the packet flow The network business function.

If a service node with the same network function is regarded as a node set, a path composed of a series of node sets is called a service path. Assume that there is two node sets in the one-hop service route. The former node set contains the service node A, and the latter node set includes the service node B1 and the service node B2. Taking a certain packet in the packet flow as an example, when the packet is sent by the previous service node A to the corresponding service node B1 and sent to the next service node B1, the packet can be regarded as the service node. A jump in the network topology that makes up the network. In fact, the process of transmitting a message from the service node A to the service node B1 may pass through multiple physical network devices, such as flowing through multiple network switches, but from the network of the service node, the service of the service node A is next. The hop is the service node B1 or the service node B2, and the packet stream is transmitted from the service node A to the service node B1 through one hop. At this time, the service node B1 may be referred to as the service next hop of the service node A.

Since the number of service nodes with the same network service function in one hop is likely to change, for example, if the service node is increased or decreased according to the processing load of the service node, this situation often occurs when the service node utilizes cloud computing. When implemented in a virtual machine. In addition, if a service node fails in a hop, the number of service nodes deployed in the hop is reduced, resulting in a load balancing problem of the service nodes in the same hop.

Summary of the invention

The present invention provides a packet processing method and device for solving the problem that load balancing cannot be achieved between service nodes in the same hop when the number of service nodes changes in one hop.

The first aspect provides a packet processing method, where the method includes:

According to the service route identifier included in the packet, when there are multiple service nodes in the service next hop of the packet, the traffic is classified according to the traffic classification number included in the packet and the number of service nodes in the next hop of the service. Equalization operation

Selecting a service node from the service nodes of the service next hop according to the operation result, and transmitting the message to the selected service node;

The traffic classification numbers of all the packets in the same packet flow are the same.

With reference to the first aspect, in a first possible implementation manner of the first aspect, the load balancing operation is performed according to the traffic classification number included in the packet and the quantity value of the service node in the service next hop, and according to the operation result Selecting a service node from the service node of the next hop of the service, specifically includes:

And modulating the traffic classification number included in the packet and the value of the service node in the next hop of the service;

The corresponding relationship between the identifier of the service node and the operation value is queried, and the modulo result is used as the operation value to determine the identifier of the corresponding service node, and the determined service node identifier is used as the selected service node.

With reference to the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the traffic classification number included in the packet is generated by the central control module and sent to After the flow classification module is added to the packet by the flow classification module.

In conjunction with the second possible implementation of the first aspect, in a third possible implementation manner of the first aspect, the traffic classification number is a value randomly selected by the central control module from the resource pool, or a central control The value obtained by the module according to the tuple of the packet, or a value determined by the central control module for the traffic path through which the packet needs to flow, according to the load state between the multiple service nodes in the hop.

In conjunction with the first aspect, and the first possible implementation of the first aspect, to any one of the third possible implementations of the first aspect, in a fourth possible implementation of the first aspect, the method Also includes:

The service node that receives the packet reports the tuple of the packet to the central control module, and the tuple of the reverse flow corresponding to the packet flow of the packet;

And generating, by the central control module, a traffic classification number for the packet in the reverse flow, generating a traffic classification number that is the same as the packet in the packet flow, so that the packet flow in the reverse flow The transited service node is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

A second aspect provides a packet processing device, where the packet processing device includes:

The operation unit is configured to: according to the service route identifier included in the packet, when the service hop of the packet is determined to have multiple service nodes, according to the traffic classification number included in the packet and the service node in the next hop of the service The number of values is load-balanced, where the traffic classification numbers of all the packets in the same packet flow are the same;

a selecting unit, configured to select a service node from the service nodes of the service next hop according to the operation result of the operation unit;

And a transmitting unit, configured to transmit the message to the service node selected by the selecting unit.

With reference to the second aspect, in a first possible implementation manner of the second aspect, the computing unit is configured to: take the traffic classification number included in the packet, and the value of the service node in the next hop of the service mold;

The selecting unit is specifically configured to determine, according to the correspondence between the identifier of the service node and the calculated value, the modulo result as an operation value, determine the identifier of the corresponding service node, and select the service node corresponding to the determined identifier of the service node as the selection. Business node.

A third aspect provides a message processing device, including: an interface, a memory, and a processor, wherein the interface, the memory, and the processor are connected by a bus, wherein:

The interface is configured to receive a message, and transmit the received message to the processor, and transmit the message to the service node selected by the processor;

The memory is configured to store program code, and store a service path reflected by the service route identifier and information of each service node in each hop of the service path, and transmit the stored program code to the processor;

a processor, configured to obtain the program code stored in the memory, and execute according to the obtained program code: according to the service routing identifier included in the packet, when determining that the service next hop has multiple service nodes, according to the Performing a load balancing operation on the traffic classification number included in the packet and the number of service nodes in the service next hop, and selecting a service node from the service node of the service next hop according to the operation result, and selecting the selected The identifier of the service node is sent to the interface, where the traffic classification numbers of all the packets in the same packet flow are the same.

In conjunction with the third aspect, in a first possible implementation of the third aspect,

The memory is further configured to store a correspondence between an identifier of the service node and an operation value, and transmit the relationship to the processor;

The processor is specifically configured to perform the following load balancing operation and select a service node: modulo the traffic classification number included in the packet and the quantity value of the service node in the service next hop, and according to the identifier of the service node Corresponding relationship with the operation value, the modulo result is used as the operation value, the identifier of the corresponding service node is determined, and the service node corresponding to the identified service node identifier is used as the selected service node.

In a fourth aspect, a central control device is provided, and the central control device includes:

a generating unit, configured to generate a traffic classification number and a service routing identifier for the packet;

And a sending unit, configured to send the traffic classification number and the service routing identifier to the traffic classification device, where the traffic classification device adds the traffic classification number and the service routing identifier to the packet.

With reference to the fourth aspect, in a first possible implementation manner of the fourth aspect,

The generating unit is specifically configured to use a value randomly selected from the resource pool as the generated traffic classification number, or the value obtained according to the tuple of the packet as the generated traffic classification number, or the flow of the packet One hop in the service path, a value determined according to the load status between the plurality of service nodes in the hop as the generated traffic classification number.

With reference to the fourth aspect, or the first possible implementation manner of the fourth aspect, in the second possible implementation manner of the fourth aspect, the central control device further includes:

a receiving unit, configured to receive a tuple of the packet reported by the service node, and a tuple of the reverse flow corresponding to the packet flow of the packet;

The generating unit is configured to: when generating a traffic classification number for the packet in the reverse flow, generate a traffic classification number that is the same as the packet in the packet flow, so that the report in the reverse flow The service node through which the text flows is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

In a fifth aspect, a central control device is provided, including an interface, a memory, and a processor, wherein the interface, the memory, and the processor are connected by a bus;

The memory is configured to store program code and transmit the stored program code to a processor;

The processor is configured to generate a traffic classification number and a service route identifier for the packet, and transmit the traffic classification number and the service route identifier to the interface;

The interface is configured to send the traffic classification number and the service route identifier to the traffic classification device, and the traffic classification device adds the traffic classification number and the service route identifier to the packet.

With reference to the fifth aspect, in a first possible implementation manner of the fifth aspect,

The memory is further configured to store a value in the resource pool and transmit the value to the processor;

The processor is specifically configured to use a value randomly selected from the resource pool as the generated traffic classification number, or a value obtained according to the tuple of the packet as the generated traffic classification number, or A hop in the service path through which the packet needs to pass, and a value determined according to the load status between the multiple service nodes in the hop as the generated traffic classification number.

With reference to the fifth aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner of the fifth aspect,

The interface is further configured to receive a tuple of the packet reported by the service node, and a tuple of the reverse flow corresponding to the packet flow of the packet, and transmit the tuple to the processor;

The processor is configured to: when generating a traffic classification number for the packet in the reverse flow, generate a traffic classification number that is the same as the packet in the packet flow, so that the report in the reverse flow The service node through which the text flows is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

The present invention pre-configures the same traffic classification number for each packet in the same packet flow. When the packet has multiple service nodes in the service next hop of the service node network, the traffic classification number is used for load balancing. After the operation, a service node is selected from the plurality of service nodes, so that the packet flows through the selected service node, and when the number of the packet flows is stable, the packets between multiple service nodes in one hop can be implemented. The average load of the flow can realize the load balancing between the service nodes in a statistical sense even if the number of service nodes in one hop changes.

DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present invention, Those skilled in the art can also obtain other drawings based on these drawings without paying for inventive labor.

1 is a schematic diagram of steps of a message processing method according to Embodiment 1 of the present invention;

2 is a schematic diagram of a service node network according to Embodiment 1 and Embodiment 2 of the present invention;

3 is a schematic diagram of steps of a packet processing method according to Embodiment 2 of the present invention;

4 is a schematic diagram of a service node network in Embodiment 3 of the present invention;

FIG. 5 is a schematic diagram of steps of a packet processing method according to Embodiment 3 of the present invention; FIG.

6(a) and 6(b) are schematic diagrams showing the structure of a message processing device according to Embodiment 4 of the present invention;

7(a) and 7(b) are schematic diagrams showing the structure of a central control device according to a fifth embodiment of the present invention.

detailed description

The present invention will be further described in detail with reference to the accompanying drawings, in which FIG. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

In order to implement the load balancing of the service node in the case where the number of the service nodes in the one-hop of the service path is dynamically changed, the embodiment of the present invention provides a packet processing scheme, which is pre-determined for each packet in the same packet flow. The traffic classifier number (English: traffic classify number, abbreviation: TCN) is configured to have multiple service nodes in one hop of any packet in the packet flow (that is, multiple hops have the same network service function). The service node may select a suitable service node from the plurality of service nodes according to the TCN configured for the packet, so that the packet flows through the selected service node. When the number of packet flows is large and stable, the total load of the packet flows between multiple service nodes in one hop can be realized. Even if the number of service nodes changes in one hop, multiple statistics can be realized. Load balancing between business nodes.

In the solution of the embodiment of the present invention, in the process of transmitting the packet flow from the source node to the destination node, if the packet in the packet flow needs to perform several kinds of network service functions, the flow needs to be formed through multiple service nodes. Business node network. After the packet stream starts from the source node, it can reach the portal of the service node network through multiple network switches. When the packet stream is transmitted in the service node network, it can also be sent from the previous service node to the next service node after one hop. Pass through multiple network switches.

The process of transmitting the packet flow from the source node to the destination node may be, for example, a process in which the video server that is the source node generates a packet flow and then transmits the packet flow to the terminal that is the destination node.

Taking the service node network shown in FIG. 2 as an example, when the packet flow arrives at the entry of the service node network (the flow classification module in FIG. 2), the flow classification module sequentially forwards each packet in the packet flow to the service first hop. After the service node A transmits, the service node A performs the corresponding network service function on the received message. Then, it is transmitted to one of the service node B1, the service node B2, and the service node B3 in the second hop of the service.

In the service node network shown in Figure 2, each service node is attached to a load balancing module. The service node is attached to the load balancing module. The load balancing module transmits the packet to the service node and receives the packet. After the service node performs the network service function and performs the next hop transmission on the packet, the service node does not need to care about where the packet comes from and where to go. As long as the packet transmitted by the attached load balancing module is received, The corresponding network service function can be performed on the message.

The load balancing module needs to record information such as media access control (MAC) address of the service node attached to the load balancing module, and can communicate with the service node, if a load balancing module is attached For multiple service nodes, the interface number of the communication interface that communicates with each service node is also recorded. At the same time, the service node also needs to record information such as the MAC address of the attached load balancing module that can communicate with the load balancing module.

The load balancing module records information of each service node in the node set to which the service node attached to the load balancing module belongs, and records the location of the node set to which the service node attached to the load balancing module belongs in the service path. Record information of the service node in the next hop node set of the node set. The information of the service node includes: an identifier of the service node, an operation value corresponding to the service node for performing load balancing operation, and an address of the load balancing module to which the service node is attached.

Since the number of service nodes in a hop changes, the number of service nodes in each hop of the service node network can be monitored by the central control module in FIG. 2, and if the number of service nodes in a hop changes, Then, the central control module determines the node set to which the added or decreased service node belongs, and sends the information of the added or reduced service node to the SRF to which the service node is attached in the node set; and the central control module further performs the service according to the node set. The location in the path determines a set of upper hop nodes of the set of nodes, and sends information of the added or decreased service node to the load balancing module attached to each service node in the set of previous hop nodes through an update message.

The load balancing module that receives the update message sent by the central control module needs to update the information of the locally recorded service node according to the content included in the received update message.

The load balancing module may be a module independent of the service node attached thereto. For example, the load balancing module may be a service routing function (SRF); the load balancing module may also be A module in which the business nodes attached to it are integrated, and the load balancing module appears as a virtual machine. For the convenience of description, the load balancing module is an SRF as an example.

The process of transmitting a message from one service node to another service node in the embodiments of the present invention refers to the transmission of the message on the service path.

The present invention will be specifically described below in conjunction with specific embodiments. Of course, the solution of the present invention is not limited to the following embodiments.

Embodiment 1:

FIG. 1 is a schematic diagram of steps of a packet processing method according to Embodiment 1 of the present invention, where the method includes the following steps:

Step 101: Perform load balancing according to the TCN of the packet and the number of service nodes in the service next hop according to the service route identifier in the packet, when the service next hop has multiple service nodes. Operation.

In the solution of the first embodiment, the TCNs added to all the packets in the same packet flow are the same.

Step 102: Select a service node from the service nodes of the service next hop according to the operation result, and transmit the message to the selected service node.

With the packet processing method described in the first embodiment of the present invention, when the number of packet flows is large and stable, the total load of the packet flows between multiple service nodes in one hop can be realized, even in a hop service. The number of nodes changes, and load balancing between multiple service nodes can also be achieved statistically.

The following describes the solution of the first embodiment of the present invention in detail by using the service node network shown in FIG. 2 as an example.

Embodiment 2:

FIG. 3 is a schematic diagram of steps of a packet processing method according to Embodiment 2 of the present invention, where the method includes the following steps:

Step 201: The central control module generates a TCN and a service route identifier for the packet in the packet flow.

In the solution of the first embodiment of the present invention, when the packet flow arrives at the entry of the service node, that is, the flow classification module in FIG. 2, the flow classification module first sends the first packet in the packet flow to the central control module. The central control module generates a TCN and a service route identifier for the received packet.

The TCN generated by the central control module may be a value randomly selected by the central control module from the resource pool, or may be based on the value obtained by the tuple of the packet, or may be one of the service paths for the packet to flow through. Hop, a value determined based on the load state between multiple service nodes in the hop.

In the case that the TCN generated by the central control module is a value randomly selected by the central control module from the resource pool, a plurality of values that can be used as TCNs are pre-cached in the resource pool. The value that can be pre-cached in the resource pool as a TCN is a positive integer. The largest value of all the values cached in the resource pool meets the following conditions: greater than or equal to M times the maximum value of the service node with the same network service function in each hop through which the packet flows, where M is an integer greater than 1. . Optionally, M is equal to 100. For example, there is one service node A in the first hop of the service to which the packet needs to flow, and three service nodes in the second hop of the service (the service node B1, the service node B2, and the service node B3, respectively), and the service third hop There are two service nodes (the service node C1 and the service node C2). In this case, the maximum value of the service node with the same network service function is 3 in one hop. If M is equal to 100, all in the resource pool. The largest value in the value is an integer greater than or equal to 300.

The tuple of the packet is a combination of one or more of the following: a source Internet protocol (English: Internet Protocol, abbreviated IP) address; a destination IP address; a protocol number; and the packet is a transmission control protocol. : Transmission Control Protocol (abbreviation: TCP), the TCP source port number of the packet, the TCP destination port number of the packet when the packet is a TCP packet, and the user data in the packet. The UDP source port number of the packet in the case of the User Datagram Protocol (UDP) packet. The UDP destination port number of the packet when the packet is a UDP packet.

In the case where the TCN generated by the central control module is a value obtained according to a tuple of a message, specifically, the TCN is calculated according to the tuple. The above calculated algorithm can be a hash (English: hash) algorithm. The tuple of the packet is the destination IP address of the packet, and the algorithm is calculated for the purpose. The IP address is added as an example of the four decimal integers in dotted decimal notation (English: dot-decimal notation). If the destination IP address of the packet is 200.1.1.154, the value in the destination IP address is 200. The sum of 1, 1, and 154, 356, is used as the TCN.

In a case where the TCN generated by the central control module is a value determined according to load states of a plurality of service nodes in one hop, the central control module determines a service node with a lower load among the plurality of service nodes in the hop, so that the SRF After the load balancing operation in step 209 is performed by using the determined TCN, the service node with the lower load can be selected from the plurality of service nodes according to the operation result, so as to achieve load balancing between the service nodes.

The central control module may generate a service route identifier according to the service path of the packet in the network of the service node, where the service route identifier may reflect a one-hop service path. In other words, the service route identifier may reflect the packet requirement. The service node having the network service function performs the corresponding network service function on it, but if there are multiple service nodes having the same network service function, the service route identifier only reflects the need to be among the multiple service nodes. A service node performs a corresponding network service function on the message, and does not reflect which of the plurality of service nodes performs a corresponding network service function on the message.

Taking the network of the service node shown in FIG. 2 as an example, the central control module determines that the packet needs to be executed by the service node A to perform the corresponding network service function, and is executed by one of the service node B1, the service node B2, or the service node B3. The network service function, the generated service route identifier can reflect that the packet needs to flow through the service node A in the network shown in FIG. 2, and also needs to flow through one of the service node B1, the service node B2, or the service node B3, but It is impossible to determine which one of the service node B1, the service node B2, or the service node B3 flows through.

The step 201 may be a preferred step for implementing the object of the present invention. The embodiment of the present invention may also generate a TCN and a service route identifier for the packet in the packet flow by other network devices, and the method for generating the TCN is not limited to the above three. Ways.

Step 202: The central control module transmits the generated TCN and the service route identifier to the flow classification module.

The central control module can control one or more flow classification modules, and send the report to the flow classification module. After the TCN and the service route identifier are generated, the packet, the TCN and the service route identifier generated by the packet are transmitted to the traffic classification module that sends the packet.

Step 203: The flow classification module adds the TCN and the service route identifier to each packet in the packet flow.

The TCN and service route identifier added in the packet will always be accompanied by the transmission of the packet in the service node network, and will not change.

Preferably, the embodiment of the present invention extends the packet header of the packet, and the traffic classification module adds a service routing packet header to the packet, and the traffic classification module adds the TCN and the service routing identifier to the newly added Service routing message header. The advantage of this is that when some service nodes perform the corresponding network service function on the packet, the content in the IP header is modified, and the TCN and the service route identifier are added to the service routing packet header without being modified. In addition, when the TCN is used for load balancing, it does not depend on the packet format. It supports a new packet format. The TCN and the service route identifier are added to the service routing header. The SRF can be identified from the packet regardless of the format of the packet. The TCN and service route identifier contained in the packet.

The functions of the central control module and the flow classification module may be implemented by two network devices, or may be implemented by one network device, that is, the network device has the functions of the central control module and the flow classification module.

The foregoing steps 202 and 203 are preferred steps of the embodiment of the present invention. The embodiment of the present invention may add a TCN to the packet by using other methods or by other network devices.

Step 204: The traffic classification module transmits each packet in the packet flow to the SRF1 attached to the service node A according to the service route identifier included in the packet.

In the network shown in FIG. 2, there is only one service node A in the first hop of the service. Therefore, the traffic classification module can directly transmit the packet to the service node A. Assume that there are multiple service nodes with the same network service function in the first hop of the service, such as the service node A1 and the service node A2, and the flow classification module also follows the schemes of the following steps 208 to 211, from the service node A1 and the service. A service node is selected in the node A2, and the message is transmitted to the SRF to which the selected service node is attached.

This step 204 is a preferred step of the embodiment of the present invention, and the solution of the embodiment of the present invention may be His network device transmits a message to SRF1.

Step 205: The SRF1 attached to the service node A receives the packet.

Step 206: SRF1 transmits the message to service node A.

Step 207: After the service node A performs the corresponding network service function on the packet, the service node A returns the packet to the SRF1.

The service node records the MAC address of the attached SRF. After the service node A performs the corresponding network service function on the packet, the service node returns the packet to SRF1 according to the MAC address of the SRF1.

The foregoing step 201 to step 207 are performed before the step 101 in the first embodiment. For any SRF, the SRF needs to transmit the received message to the attached before the packet is transmitted to the next hop. The service node on the SRF performs the operation of step 101 in the first embodiment when receiving the packet that has been sent by the service node and has performed the network service function.

Step 208: The SRF1 determines the service node of the service next hop according to the service route identifier in the packet. If there are multiple service nodes in the next hop of the service, step 209 is performed; if there is only one service node in the next hop of the service, The message is directly transmitted to the SRF to which the service node of the next hop of the service is attached.

The SRF can store the information of each service node in the service path in the service path in the form of a load balancing table, as shown in Table 1.

Table 1

SRF1 queries the load balancing table stored in the local device according to the service route identifier included in the packet, and determines that the next hop of the service includes the service node B1 and the service node B2 according to the content in Table 1. Table 1 also records the operation value of the service node B1 and the address of the attached SRF2, and records the operation value of the service node B2 and the address of the attached SRF3.

The operation value is used to compare the obtained operation result with the obtained operation result after the load balancing operation in the subsequent step 209 to select the purpose of the service node. The operation value may be a service node B1 and a service section. The sequence values of the SRF2 and SRF3 addresses respectively attached to point B2 are sorted.

Assume that in the network shown in FIG. 2, a service node B3 having a B network service function (a portion indicated by a broken line in FIG. 1) is added, and the service node B3 is attached to the SRF4, and the central control module can identify the service node B3. The address of SRF4 and the calculated value corresponding to service node B3 are notified to SRF1, and SRF1 updates the contents of Table 1 to Table 2.

Table 2

In addition to updating Table 1 in SRF1, SRF2, SRF3, and SRF4 also update the local load balancing table, where SRF2 records the identity of service node B2, the operational value of service node B2, and the address of SRF3 to which service node B2 is attached. Recording the identity of the service node B3, the operation value of the service node B3, and the address of the SRF4 to which the service node B3 is attached; the SRF3 records the identity of the service node B1, the operation value of the service node B1, and the address of the SRF2 to which the service node B1 is attached, Recording the identity of the service node B3, the operation value of the service node B3, and the address of the SRF4 to which the service node B3 is attached; the SRF4 records the identity of the service node B1, the operation value of the service node B1, and the address of the SRF2 to which the service node B1 is attached. The identification of the service node B2, the operation value of the service node B2, and the address of the SRF3 to which the service node B2 is attached are recorded.

It should be noted that FIG. 2 is an example in which a service node is attached to each SRF. When multiple service nodes are attached to one SRF, each service node can communicate with the SRF through an independent port. It is assumed that in the network diagram shown in FIG. 2, a service node B3 having a B network service function is added, the service node B3 is attached to the SRF3, and the operation value of the service node B3 is 2, and the central control module can set the service node. The identifier of B3, the operation value, and the address of the attached SRF4 are notified to SRF1, and SRF1 updates the contents of Table 1 to Table 3.

table 3

In addition to updating Table 1 in SRF1, both SRF2 and SRF3 also update the local load balancing table, where SRF2 records the identity of service node B2, the operational value of service node B2, and the address of SRF3 to which service node B2 is attached, and also records the service. The identifier of the node B3, the operation value of the service node B3, and the address of the SRF3 to which the service node B3 is attached; the SRF3 records the identifier of the service node B1, the operation value of the service node B1, and the address of the SRF2 to which the service node B1 is attached, and records the service. The interface number of the communication interface of node B2 and SRF3 (assumed to be interface number 1), the interface number of the communication interface of service node B3 and SRF3 (assumed to be interface number 2).

Step 209: SRF1 performs load balancing operation according to the TCN in the packet and the value of the service node in the next hop of the service.

The above step 209 is a detailed description of the step 101 in the first embodiment.

Step 210: SRF1 selects one service node from multiple service nodes of the next hop of the service according to the operation result.

In this step 210, after obtaining the operation result by the load balancing operation, SRF1 queries Table 1, Table 2, or Table 3, and the value range of the operation result is the same as the set of the operation values in Table 1, Table 2, or Table 3, According to the obtained operation result, the same operation value can be queried from Table 1, Table 2 or Table 3, and then the service node corresponding to the calculated operation value is determined. For example, the above load balancing operation can be a hash calculation.

Step 211: SRF1 transmits the message to the selected service node.

It is assumed that the service node selected by the SFR1 through the load balancing operation and the query table 1 in step 209 and step 210 is the service node B2. In this step 211, the SRF1 is further reported according to the address of the SRF3 attached to the service node B2 in Table 1. The message is transmitted to the SRF3, and then the message is transmitted by the SRF3 to the service node B2, and the process of transmitting the message from the service node A to the service node B2 is realized.

It is assumed that the service node selected by SRF1 through the load balancing operation and the query table 3 in step 209 and step 210 is the service node B2. In this step 211, the SRF1 is further directed to the SRF3 according to the address of the SRF3 attached to the service node B2 in Table 3. After the message is received, the SRF3 determines that both the service node B2 and the service node B3 are attached to the SRF3. Therefore, the SRF3 performs the TCN according to the received message and the recorded value of the current hop in the current hop. The same load balancing operation as that of SRF1 in step 209, the obtained operation result is the same as the operation result of SRF1 in step 209, and it can be determined that the message needs to be transmitted to the service node B2, and the SRF3 is based on the interface of the communication interface between the service node B2 and the SRF3. The number (assumed to be interface number 1) transmits a message to the service node B2.

After the service node B2 performs the corresponding network service function on the packet, the service packet is returned to the SRF3. In the case that the transmission process of the packet in the service node network shown in FIG. 2 is ended, if the SRF3 has the function of forwarding the packet according to the IP address, the SRF3 forwards the packet according to the destination IP address of the packet; if the SRF3 does not have According to the function of forwarding packets according to the IP address, the SRF3 transmits the packet to the designated router according to the preset forwarding mode. The router continues to forward the packet according to the destination IP address of the packet.

The above steps 210 and 211 are specific descriptions of the step 102 in the first embodiment.

The solution of the first embodiment of the present invention is described by taking the network diagram shown in FIG. 2 as an example. For other forms of service node networks, the transmission process of the message in the network is similar to the transmission process of step 211 of the step 208. , will not repeat them here.

With the solution of the first embodiment of the present invention, the number of service nodes in the next hop can be used to select the service node in the service next hop by using the TCN included in the packet. The load balancing of each service node in one hop is ensured; even if the TCN is allocated for the packets in each packet flow in a random manner, because the number of service nodes in one hop is limited, a large number of packet flows can be basically All are divided into various business nodes.

Embodiment 2:

The specific description of the steps 209 to 211 in the first embodiment of the present invention is as follows:

For example, the service node network diagram shown in FIG. 2 is taken as an example. Assume that during the execution of the first embodiment, the number of service nodes having the B network service function is 2, which are the service node B1 and the service node respectively. Point B2. In step 209, the SRF1 needs to select a service node from the service node B1 and the service node B2. Specifically, the SRF1 can perform a load balancing operation by using a hash algorithm, and then select a service node according to the operation result.

The load balancing operation mode includes, but is not limited to, modulating the value of the service node in the TCN and the service next hop in the packet, and selecting a service node according to the modulo result. The specific implementation process is as follows:

First, SRF1 modulates the value of the TCN in the packet and the number of service nodes in the next hop of the service (in this case, the value is 2). The modulo result is 0 or 1 regardless of the value of the TCN.

Then, the SRF1 queries the correspondence between the identifier of the service node and the operation value, and uses the modulo result as the operation value, and the identifier of the corresponding service node, that is, the query table 1, determines the service node corresponding to the operation value with the same modulo result. Logo.

Finally, SRF1 takes the determined service node corresponding to the identity of the service node as the selected service node.

Taking the modulo result as 0 as an example, it can be seen from the query table 1 that the service node corresponding to the operation value of 0 is the service node B1, and in step 109, the SRF1 selects the service node B1 from the service node B1 and the service node B2. After that, SRF1 transmits the packet to SRF2 according to the address of SRF2 attached to service node B1 recorded in Table 1, and then transmits the packet to service node B1 by SRF2, and completes the packet from service node A to service node B1. The transfer process.

The foregoing is an example in which the service node in the second hop of the service is the service node B1 and the service node B2, and the number of service nodes in the second hop of the service is changed, for example, the service node attached to the SRF3 is added. B3. At this time, the number of service nodes having the B network service function in the second hop of the service is 3, and both the service node B2 and the service node B3 are attached to the SRF3, and the load balancing table stored in the SRF1 is Table 3.

The process of selecting a service node from service node B1, service node B2, and service node B3 by SRF1 is:

First, SRF1 modulates the value of the TCN in the packet and the number of service nodes in the next hop of the service (in this case, the value is 3), and the modulo result is 0, 1, or 2.

Then, SRF1 queries Table 3 to determine the same operational value as the modulo result.

Finally, SRF1 uses the determined service node corresponding to the calculated value as the selected service node.

Taking the modulo result as 1 as an example, when the operation parameter is 1 and the corresponding service node is the service node B2, the service node B2 is taken as the service node selected in step 109. Thereafter, the SRF1 lookup table 3 determines the address of the SRF3 to which the service node B2 is attached, and transmits the message to the SRF3 according to the address of the SRF3. After receiving the packet, the SRF3 modulo the value of the TCN in the packet and the number of the currently hopped service node (the value is 3), and the obtained modulo result is also 1, and the corresponding operation value is determined to be 1. The service node is the service node B2, and the SRF3 transmits the message to the service node B2 according to the interface number of the interface with the service node B2.

The solution of the second embodiment of the present invention is a preferred solution of the first embodiment. However, the first embodiment of the present invention is not limited to selecting a service node from the next hop of the service by using another load balancing algorithm, and the packet is in the next hop of the service. Business node.

Preferably, on the basis of the solution described in the first embodiment or the second embodiment, if the packet flow transmitted from the source node to the destination node is regarded as a forward flow, after the packet flow reaches the destination node, the destination node receives the packet. After the received packet stream is processed, if the packet stream sent back to the source node is generated, the packet stream can be regarded as a reverse stream. If the reverse flow is transmitted back from the destination node to the source node, the service node that flows through is the same as the service node through which the forward flow flows, but the flow sequence is reversed. And the reverse flow is common. In this case, if the transmission process of the forward flow in the service node network implements load balancing between the service nodes, the reverse flow can also achieve the above effect. The bidirectional transmission process of the forward flow and the reverse flow will be described below through the third embodiment.

Embodiment 3:

The third embodiment of the present invention may further increase the reverse flow processing process based on the first embodiment and the second embodiment, and take the service node network diagram shown in FIG. 4 as an example, and assume a forward flow (assuming forward flow) The order of the flow is the flow through the service node in the network shown in Figure 4: service node A - service node B1 - service node C, in order to make the reverse flow (assuming the reverse flow is identified as flow-r) For the forward flow common path, the order of the reverse flow flowing through the service node in the network shown in FIG. 4 should be: service node C-service node B1-service node A.

Suppose the central control module generates a TCN value of 24 for the packets in the forward flow, and generates the generated service. The IP address of the SRF1 to which the service node A is attached is 10.10.1.1, the IP address of the SRF2 to which the service node B1 is attached is 10.10.1.2, and the IP address of the SRF3 to which the service node B2 is attached is 10.10.1.3. The IP address of the SRF4 to which the service node C is attached is 10.10.1.4, and the load balancing table stored in the SRF1 is as shown in Table 4:

Table 4

The load balancing table stored in SRF2 is shown in Table 5(a) and Table 5(b):

Table 5 (a)

Table 5(b)

The load balancing tables stored in SRF3 are shown in Table 6(a) and Table 6(b):

Table 6 (a)

Table 6(b)

The load balancing table stored in SRF4 is shown in Table 7:

Table 7

Based on the network shown in FIG. 4, a schematic diagram of the steps of the packet processing method in the forward flow and the reverse flow is shown in FIG. 5, and mainly includes the following steps:

Step 301: SRF1 sequentially transmits the packet of the forward flow to the SRF2 attached to the service node B1.

The traffic classification module firstly transmits the packet of the forward flow to the SRF1 to which the service node A is attached, and the SRF1 transmits the packet to the service node A. When the SRF1 receives the packet returned by the service node A to perform the corresponding network service function, the packet is returned. If the service route identifier (in this case, the service route identifier is 1001) is queried in Table 4, it is determined that there are two service nodes in the service next hop; then the TCN included in the packet (the value of the TCN at this time) 24) and the service next hop service node number value 2 modulo, the modulo result is 0, through the query table 4, it is determined that the message needs to flow through the service node B1 in the service next hop, and finally from the table 4 The address of the SRF2 to which the service node B1 is attached transmits the message to the SRF2.

Step 302: SRF2 transmits the message to the service node B1.

Step 303: The service node B1 reports to the central control module a tuple of the forward flow in which the packet is located, and a tuple of the reverse flow corresponding to the forward flow.

Since the tuple can uniquely represent the packet flow, after the service node B1 reports the tuple of the forward flow and the reverse flow to the central control module, the central control module can identify the forward direction of the common path according to the received tuple. Stream and reverse flow. The tuples involved in the embodiments of the present invention may be a five-tuple or a seven-tuple.

In fact, in the transmission process of the service node network, the forward flow may flow through multiple service nodes, and each service node that flows through it does not need to report the forward flow and the reverse flow to the central control module. A quintuple, but a service node that has a common demand for the forward flow and the reverse flow, needs to report the quintuple of the forward flow and the reverse flow to the central control module. For example: NAT service node needs to know the positive direction The source IP address and destination IP address of the packets in the flow, and the source IP address and destination IP address of the corresponding reverse flow packets. Therefore, the NAT service node has a forward flow and a reverse flow common path. Demand.

Step 304: The central control module locally records the association relationship between the received tuple of the forward stream and the tuple of the reverse stream, and determines that the TCN generated for the packet in the forward stream is 24.

Step 305: The forward flow continues to be transmitted from the service node B1 to the service node C, and then transmitted to the destination node to complete the forward flow transmission process.

In this step 305, after performing the corresponding network service function on the packet, the service node B1 returns the packet to the SRF2, and after the SRF2 queries the table 5(a), the packet is transmitted to the SRF4, and the SRF4 transmits the packet. After the service node C performs the corresponding network service function on the packet, the service node C returns to the SRF4, and then the SRF4 continues to perform network transmission on the packet until the packet reaches the destination node.

Step 306: The central control module generates a TCN and a service route identifier for the reverse flow, where the TCN generated for the reverse flow is the same as the TCN of the forward flow.

After the forward flow reaches the destination node, the destination node processes the forward flow to generate a reverse flow, and after the multi-hop through the network, the reverse flow reaches the network entry shown in FIG. 3. The flow classification module in FIG. 3 sends the first packet of the reverse flow to the central control module, and the central control module identifies the tuple in the packet, and the tuple and reverse flow of the locally recorded forward flow. If the tuple in the received packet is the same as the tuple in the reverse relationship in the association relationship, it is determined that the received packet flow is reversed. The flow is forwarded, and the reverse flow needs to be shared with the forward flow. Therefore, the central control module generates a service route identifier for the reverse flow (the service route identifier is 1002 at this time), and generates the same reverse flow for the reverse flow. TCN (at this time TCN is 24).

In the solution of the third embodiment, when the forward flow enters the network shown in FIG. 3, the flow classification module of the ingress and the flow classification module of the ingress when the reverse flow enters the network may be the same network device, or may be independent. The two network devices of the present invention do not limit the form of the flow classification module.

Step 307: The traffic classification module adds the TCN and the service route identifier to the reverse flow packet, and transmits the reverse flow packet to the SRF4 to which the service node C is attached.

Step 308: SRF4 transmits the message to the service node C, and receives the return from the service node C. When the message is received, the message is transmitted to the SRF2 to which the service node B1 is attached.

In this step 308, SRF4 queries Table 7, and determines that there are two service nodes in the next hop of the service, and the modulo result of the TCN and 2 modulo is 0, and the service node B1 in the next hop of the service is selected, and The message is transmitted to the SRF2 to which the service node B1 is attached.

By comparing step 301 and step 308, when the forward stream and the reverse stream are flowing through one of the service node B1 and the service node B2, the TCN and the reverse stream message in the packet of the forward flow The TCN in the same is the same. Therefore, after the same modulo operation, SRF1 selects the service node B1 for the message in the forward stream, and SRF4 selects the service node B1 for the message in the reverse stream to ensure the forward flow and The same way of reverse flow.

Step 309: The reverse flow continues to be transmitted from the service node B1 to the service node A, and then transmitted to the source node to complete the reverse flow transmission process.

The source node in this step 309 is the source node of the forward stream and is the destination node of the reverse stream.

According to the solution of the third embodiment of the present invention, when the TCNs of the packets in the forward flow and the reverse flow are the same, the same path of the forward flow and the reverse flow can be realized, which is not only implemented in the forward flow transmission process. Load balancing between service nodes also implements load balancing during reverse flow transmission. At the same time, the same path of forward flow and reverse flow can well meet the service requirements of NAT service nodes.

It should be noted that, in the embodiment of the present invention, a fixed value TCN and a service route identifier are directly added to the newly added service routing packet header. Of course, if the TCN is not directly added to the packet, it is When the service node needs to be selected from the next hop of the service, the value obtained by performing the hash operation on some fields of the network packet header may also perform the load balancing operation according to the solution of the embodiment of the present invention. However, the solution of the embodiment of the present invention has the following advantage:

1) After the message is executed by the service node, the content of some fields in the network packet header changes, for example, after the NAT service node performs the corresponding network service function, the network packet header is The destination IP address will change. If the value obtained by hashing the content in the changed field is used for load balancing operation, the final load balancing operation result may interfere with the equalization of the packet flow between the service nodes. The purpose of load balancing between business nodes. With the solution of the embodiment of the present invention, the content in the service routing packet header is not modified, and the TCN is ensured. The value does not change so that the message flow is evenly divided among the business nodes.

2) If the packet in the packet flow adopts a new packet format, the SRF that is attached to the currently deployed service node cannot identify the packet in the new packet format. Therefore, it is difficult to implement load balancing operation. In the solution of the embodiment of the present invention, the TCN and the service route identifier are added to the newly added service routing packet header, and the format of the service routing packet header does not change, and the service node can identify the service routing packet header. TCN and service route identifier.

Embodiment 4:

A fourth embodiment of the present invention describes a message processing device. As shown in FIG. 6(a), the message processing device includes an operation unit 11, a selection unit 12, and a transmission unit 13, wherein:

The operation unit 11 is configured to: according to the service route identifier included in the packet, when there are multiple service nodes in the service next hop of the packet, according to the TCN included in the packet and the number of service nodes in the next hop of the service The value is load-balanced, where the TCNs of all the packets in the same packet flow are the same;

The selecting unit 12 is configured to select one service node from the service nodes of the service next hop according to the operation result of the operation unit 11;

The transmission unit 13 is configured to transmit the message to the service node selected by the selection unit 12.

Optionally, the operation unit 11 is configured to: modulate a value of a TCN in the packet and a value of a service node in a service next hop;

The selecting unit 12 is specifically configured to determine, according to the correspondence between the identifier of the service node and the calculated value, the modulo result as an operation value, determine the identifier of the corresponding service node, and select the determined service node as the service node. Business node

The transmission unit 13 is specifically configured to: according to the correspondence between the service node and the load balancing module to which the service node is attached, transmit a packet to the load balancing module attached to the selected service node, and then the load balancing module sends the packet to the load balancing module. The selected service node attached to the load balancing module transmits a message.

The packet processing device involved in the fourth embodiment may be the load balancing module involved in the first embodiment to the third embodiment, the correspondence between the identifier of the service node and the calculated value, and the service node and the service. The correspondence between the load balancing modules attached to the service nodes may be recorded in the load balancing table in the first embodiment, and the selecting unit 12 and the transmitting unit 13 respectively determine the two correspondences by querying the load balancing table.

Another embodiment of the present invention further describes another message processing device. As shown in FIG. 6(b), the message processing device includes an interface 21, a memory 22, and a processor 23, wherein the interface 21, the memory 22, and The processors 23 are connected by a bus.

The interface 21 is configured to receive a message, and transmit the received message to the processor 23, and transmit the message to the service node selected by the processor 23;

The memory 22 is configured to store the program code, and store the service path reflected by the service route identifier and the information of each service node in each hop on the service path, and transmit the stored program code to the processor 23;

The processor 23 is configured to obtain the program code stored in the memory 22, and execute according to the obtained program code: according to the service routing identifier included in the packet, when determining that the service next hop has multiple service nodes, according to Performing a load balancing operation on the number of service nodes included in the TCN and the service next hop in the packet, and selecting a service node from the service node of the service next hop according to the operation result, and selecting the selected The identifier of the service node is sent to the interface 21, where the TCNs of all the packets in the same packet flow are the same.

The memory 22 is further configured to store a correspondence between the identifier of the service node and the operation value, and transmit the corresponding relationship to the processor 23;

The processor 23 is specifically configured to perform a load balancing operation and select a service node: modulo the value of the TCN and the service node in the service next hop included in the packet, and according to the identifier of the service node Corresponding relationship between the calculated values, the modulo result is used as the operation value, the identifier of the corresponding service node is determined, and the service node corresponding to the determined identity of the service node is taken as the selected service node.

The memory 22 is further configured to store a correspondence between the service node and the load balancing module to which the service node is attached, and transmit the result to the processor 23. Specifically, the memory 22 may be stored in the load balancing table in the first embodiment. And transmitted to the processor 23.

The processor 23 is specifically configured to determine a load balancing module to which the selected service node is attached, and transmit the address of the load balancing module to the interface 21;

The interface 21 is specifically configured to transmit the packet to the load balancing module according to the address of the load balancing module.

The interface 21 may be one or more of the following: a network interface controller (English: network interface controller, NIC), such as an Ethernet NIC, which can provide a copper interface and a fiber interface. Or provide both a copper interface and a fiber interface; a NIC that provides a wireless interface, such as a wireless local area network (WLAN) NIC.

The memory 22 may be a volatile memory, such as a random access memory (English: random-access memory, abbreviation: RAM), or a non-volatile memory (English: non-volatile memory). For example, flash memory (English: flash memory), hard disk (English: hard disk drive, abbreviated: HDD) or solid state drive (English: solid-state drive, abbreviation: SSD); or a combination of the above types of memory.

The processor 23 may be a central processing unit (English: central processing unit, abbreviated: CPU), or a combination of a CPU and a hardware chip.

The above hardware chip may be a combination of one or more of the following: an application-specific integrated circuit (ASIC: ASIC), a field-programmable gate array (English: field-programmable gate array, abbreviation: FPGA) , Complex programmable logic device (English: complex programmable logic device, abbreviation: CPLD), network processor (English: network processor, abbreviation: NP).

Embodiment 5:

A fifth embodiment of the present invention also describes a central control device. As shown in FIG. 7(a), the central control device includes a generating unit 31 and a transmitting unit 32, where:

The generating unit 31 is configured to generate a TCN and a service routing identifier for the packet.

The sending unit 32 is configured to send the TCN and the service routing identifier to the traffic classification device, and instruct the traffic classification device to add the TCN and the service route identifier in the packet.

The generating unit 31 is specifically configured to use a value randomly selected from the resource pool as the generated TCN, or the value obtained according to the tuple of the packet as the generated TCN, or in the service path through which the packet needs to flow. One hop, a value determined based on the load state between the plurality of service nodes in the hop as the generated TCN.

The central control device further includes a receiving unit 33, configured to receive a tuple of the packet reported by the service node, and a tuple of the reverse flow corresponding to the packet flow of the packet;

The generating unit 31 is configured to: when generating a TCN for the packet in the reverse flow, generate a TCN that is the same as the packet in the packet flow, so that the packet in the reverse flow flows through The service node is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

The central control device involved in the fifth embodiment may be the central control module involved in the second embodiment and the third embodiment. The flow classification device may be the flow classification module involved in the second embodiment and the third embodiment.

Another fifth central control device is also described in the fifth embodiment of the present invention. As shown in FIG. 7(b), the central control device includes an interface 41, a memory 42 and a processor 43, wherein the interface 41, the memory 42 and the processor 43 is connected by bus.

The memory 42, for storing program code, and transmitting the stored program code to the processor 43;

The processor 43 is configured to generate a TCN and a service route identifier for the packet, and transmit the TCN and the service route identifier to the interface 41;

The interface 41 is configured to send the TCN and the service route identifier to the traffic classification device, and instruct the traffic classification device to add the TCN and the service route identifier to the packet.

The memory 42, is also used to store the value in the resource pool and transmitted to the processor 43;

The processor 43 is specifically configured to use a value randomly selected from the resource pool as the generated TCN, or the value obtained according to the tuple of the packet as the generated TCN, or the service path that the packet needs to flow through. In the first hop, a value determined based on the load status between the plurality of service nodes in the hop is used as the generated TCN.

The interface 41 is further configured to receive a tuple of the packet reported by the service node, and the packet a tuple of the reverse stream corresponding to the packet stream, and transmitted to the processor 43;

The processor 43 is configured to: when generating a TCN for the packet in the reverse flow, generate a TCN that is the same as the packet in the packet flow, so that the packet in the reverse flow flows through The service node is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

The interface 41 may be one or more of the following: a network interface controller (English: network interface controller, NIC), such as an Ethernet NIC, which can provide a copper interface, a fiber interface. Or provide both a copper interface and a fiber interface; a NIC that provides a wireless interface, such as a wireless local area network (WLAN) NIC.

The memory 42 may be a volatile memory, such as a random access memory (RAM), or a non-volatile memory. For example, flash memory (English: flash memory), hard disk (English: hard disk drive, abbreviated: HDD) or solid state drive (English: solid-state drive, abbreviation: SSD); or a combination of the above types of memory.

The processor 43 may be a central processing unit (English: central processing unit, abbreviated: CPU), or a combination of a CPU and a hardware chip.

The above hardware chip may be a combination of one or more of the following: an application-specific integrated circuit (ASIC: ASIC), a field-programmable gate array (English: field-programmable gate array, abbreviation: FPGA) , Complex programmable logic device (English: complex programmable logic device, abbreviation: CPLD), network processor (English: network processor, abbreviation: NP).

Those skilled in the art will appreciate that embodiments of the present application can be a method, system, or computer program product. Moreover, the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the present application. It should be understood that the flow chart can be implemented by computer program instructions And/or a combination of the processes and/or blocks in the block diagrams, and the flowcharts and/or blocks in the flowcharts. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

In a typical configuration, the computer device includes one or more central processing units (CPUs), input/output interfaces, network interfaces, and memory. The memory may include random access memory (RAM) or read only memory (ROM) in a computer readable medium. As defined herein, computer readable media does not include non-persistent computer readable media (Transitory Media) such as modulated data signals and carrier waves.

While the preferred embodiment of the present application has been described, those skilled in the art can make further changes and modifications to these embodiments once they are aware of the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and the modifications and

It will be apparent to those skilled in the art that various modifications and changes can be made in the present application without departing from the scope of the application. Thus, it is intended that the present invention cover the modifications and variations of the present invention.

Claims (10)

1. A packet processing method, characterized in that the method comprises:

   According to the service route identifier included in the packet, when there are multiple service nodes in the next hop of the service, the traffic classification number TCN and the number of service nodes in the next hop of the service are performed according to the value of the service node. Load balancing operation;

   Selecting a service node from the service nodes of the service next hop according to the operation result, and transmitting the message to the selected service node;

   The TCNs of all the packets in the same packet flow are the same.
2. The packet processing method according to claim 1, wherein the load balancing operation is performed according to the TCN included in the packet and the number of service nodes in the next hop of the service, and the operation result is from the service. Select one service node from the one-hop service node, including:

   And modulating the value of the number of service nodes in the TCN and the service next hop included in the packet;

   The corresponding relationship between the identifier of the service node and the operation value is queried, and the modulo result is used as the operation value to determine the identifier of the corresponding service node, and the determined service node identifier is used as the selected service node.
3. The packet processing method according to claim 1 or 2, wherein the TCN included in the message is generated by the central control module and sent to the stream classification module, and then added by the stream classification module. In the message.
4. The packet processing method according to claim 3, wherein the TCN is a value randomly selected by the central control module from the resource pool, or a value obtained by the central control module according to the tuple of the message, or The central control module is configured to determine a value according to a load state between multiple service nodes in the hop according to a hop in the service path through which the packet needs to flow.
5. The packet processing method according to any one of claims 1 to 4, wherein the method further comprises:

   The service node that receives the packet reports the tuple of the packet to the central control module, and the tuple of the reverse flow corresponding to the packet flow of the packet;

   When the central control module generates a TCN for the packet in the reverse flow, the same TCN is generated as the packet in the packet flow, so that the service node in the reverse flow flows through the service node. The same as the service node through which the packet in the packet flow flows, but in the reverse order.
6. A message processing device, wherein the message processing device includes:

   The operation unit is configured to: according to the service routing identifier included in the packet, when the service hop of the packet is determined to have multiple service nodes, according to the traffic classification number TCN and the service in the next hop of the service The load balancing operation is performed on the number of nodes, where the TCNs of all the packets in the same packet flow are the same;

   a selecting unit, configured to select a service node from the service nodes of the service next hop according to the operation result of the operation unit;

   And a transmitting unit, configured to transmit the message to the service node selected by the selecting unit.
7. A message processing device according to claim 6, wherein:

   The operation unit is specifically configured to: modulate a value of a TCN and a service node in a service next hop included in the packet;

   The selecting unit is specifically configured to determine, according to the correspondence between the identifier of the service node and the calculated value, the modulo result as an operation value, determine the identifier of the corresponding service node, and select the service node corresponding to the determined identifier of the service node as the selection. Business node.
8. A central control device, wherein the central control device comprises:

   a generating unit, configured to generate a traffic classification number TCN and a service routing identifier for the packet;

   And a sending unit, configured to send the TCN and the service routing identifier to the traffic classification device, and instruct the traffic classification device to add the TCN and the service routing identifier to the packet.
9. A center control device according to claim 8 wherein:

   The generating unit is specifically configured to use a value randomly selected from the resource pool as the generated TCN, or the value obtained according to the tuple of the packet as the generated TCN, or in the service path through which the packet needs to flow. One hop, a value determined based on the load state between the plurality of service nodes in the hop as the generated TCN.
10. A central control device according to claim 8 or 9, wherein said central control Equipment also includes:

    a receiving unit, configured to receive a tuple of the packet reported by the service node, and a tuple of the reverse flow corresponding to the packet flow of the packet;

    The generating unit is configured to: when generating a TCN for the packet in the reverse flow, generate a TCN that is the same as the packet in the packet flow, so that the packet in the reverse flow flows. The service node is the same as the service node through which the packet in the packet flow flows, but in the reverse order.

Images