CN112311687A

CN112311687A - Traffic scheduling method and device

Info

Publication number: CN112311687A
Application number: CN202011158645.XA
Authority: CN
Inventors: 蒋文栋
Original assignee: New H3C Big Data Technologies Co Ltd
Current assignee: New H3C Big Data Technologies Co Ltd
Priority date: 2020-10-26
Filing date: 2020-10-26
Publication date: 2021-02-02
Anticipated expiration: 2040-10-26
Also published as: CN112311687B

Abstract

The present specification provides a traffic scheduling method and apparatus, where the method includes: acquiring a flow statistical table of a first node on a candidate path in a maintained segment routing strategy, wherein the flow statistical table comprises flow rates corresponding to different flow characteristic identifications in the segment routing strategy of the first node; determining a first proportion of the load capacity of each candidate path to the total residual bandwidth according to the residual bandwidth of each candidate path included in the segment routing strategy; determining a second proportion of the traffic rate in the total traffic rate of the head node according to the traffic rate corresponding to each traffic characteristic identifier; and determining a candidate path corresponding to each flow characteristic identifier according to the first proportion and the second proportion. The method can select a proper candidate path for forwarding the traffic corresponding to each traffic characteristic identifier according to the traffic statistics of the first node, thereby greatly improving the scheduling effectiveness.

Description

Traffic scheduling method and device

Technical Field

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

Background

In recent years, SDN (Software Defined Network) technology is widely applied to various fields and various user actual networks, users have more and more demands on SDN wide area networks, and generally, scheduling encapsulation technology of wide area networks is RSVP (Resource ReSerVation Protocol) MPLS-TE (MPLS Traffic Engineering, multiprotocol label switched path Engineering), SR-TE (Segment Routing of Traffic Engineering), SR-Policy (Segment Routing Policy), Policy of Segment Routing SR based on IPv6 (SRV 6-Policy), and the like, where SR-Policy and SRV6-Policy can support more functions and scenarios, and the technology is relatively advanced.

The SR-Policy structure deployed by the controller comprises candidate path lists, each candidate path has a priority, traffic can be forwarded through the currently available candidate path with the high priority after the SR Policy is determined, one candidate path also comprises a plurality of Segment lists, each Segment list corresponds to a weight, and the traffic can be forwarded in proportion according to the weight of each Segment list Segment-list.

For example, in the networking shown in fig. 1, if the controller selects 4 paths with the minimum Metric from PE1-PE2 according to SRV6-Policy with a residual available bandwidth of a link being 4G for a certain bearer flow, and determines 4 paths according to PRBW (Path Reserved Band Width, residual available bandwidth) of the 4 paths, if the determined paths and their segment lists are respectively:

Segment-list1：PE1->P1->PE2 weight 10；

Segment-list2：PE1->P2->PE2 weight 20；

Segment-list3：PE1->P3->PE2 weight 10；

Segment-list4：PE1->P4->PE2 weight 60；

wherein weight represents the weight.

The controller will deploy the Segment-list to the corresponding network device. On the other hand, the network device performs hash calculation according to the type of the received message to obtain a hash operation result of the message. After the hash operation result of the message is obtained, the network device solidifies the hash value corresponding to each Segment-list distributed to the network device according to the Segment-list weight deployed by the controller for the network device.

For example, Segment-list 1: PE1- > P1- > PE2 weight 10; the flow with the HASH value of 0-9 enters the path.

Segment-list 2: PE1- > P2- > PE2 weight 20; the HASH value of 10-29 flows enter the path.

Segment-list 3: PE1- > P3- > PE2 weight 10; the HASH values of 30-39 flow enter the path.

Segment-list 4: PE1- > P4- > PE2 weight 60; the HASH value of 40-99 flows enter the path.

The above mechanism can achieve to some extent that the policies of the control plane and the data plane remain consistent if the traffic actually forwarded in the network is very uniform. However, if the traffic in the customer site cannot be proportionally equalized to the corresponding value, the path weights deployed by the controller cannot be correctly implemented to the data forwarding plane, resulting in a difference between the control plane and data plane policies.

Disclosure of Invention

In order to overcome the problems in the related art, the present specification provides a traffic scheduling method and apparatus.

According to a first aspect of embodiments herein, there is provided a traffic scheduling method, the method including:

acquiring a flow statistical table of a head node on a candidate path in a maintained segment routing strategy, wherein the flow statistical table comprises flow rates corresponding to different flow characteristic identifications in the segment routing strategy of the head node;

determining a first proportion of the load capacity of each candidate path to the total residual bandwidth according to the residual bandwidth of each candidate path included in the segment routing strategy; determining a second proportion of the traffic rate in the total traffic rate of the head node according to the traffic rate corresponding to each traffic characteristic identifier; determining candidate paths corresponding to the flow characteristic identifications according to the first proportion and the second proportion;

and issuing the corresponding relation between the candidate path and the flow characteristic identifier to the head node.

In an optional implementation manner, the traffic feature identifier is a digest value calculated according to the feature of the traffic, and the traffic feature identifier is used for performing load sharing in the candidate path according to the traffic feature identifier.

In an optional implementation manner, if the controller cannot obtain the traffic statistic table, determining, according to remaining bandwidths of the candidate paths included in the segment routing policy, a first ratio of the bearer traffic of each candidate path to the total remaining bandwidth, and sending the first ratios of the candidate paths and the candidate paths to the head node, so that the head node forwards the traffic according to the first ratios corresponding to the candidate paths.

In an optional implementation manner, if the candidate path corresponding to each traffic characteristic identifier cannot be determined according to the first ratio and the second ratio, the candidate path with the largest residual bandwidth is determined, and a correspondence between all traffic characteristic identifiers and the candidate path with the largest residual bandwidth is generated, so that traffic corresponding to all traffic characteristic identifiers is forwarded through the candidate path with the largest residual bandwidth.

In an optional implementation manner, when the controller detects that there is any one or more of a path break, a link blockage, or a preset traffic statistic table acquisition period, acquiring the traffic statistic table of the head node on the candidate path included in the maintained segment routing policy is performed.

A second aspect of the present disclosure provides a traffic scheduling apparatus, including: the device comprises an acquisition module, a path distribution module and a sending module;

an obtaining module, configured to obtain a traffic statistics table of a head node on a candidate path included in a maintained segment routing policy, where the traffic statistics table includes traffic rates corresponding to different traffic feature identifiers in the segment routing policy of the head node;

the path allocation module is used for determining a first proportion of the load-bearing flow of each candidate path in the total residual bandwidth according to the residual bandwidth of each candidate path in the segment routing strategy; determining a second proportion of the traffic rate in the total traffic rate of the head node according to the traffic rate corresponding to each traffic characteristic identifier; determining candidate paths corresponding to the flow characteristic identifications according to the first proportion and the second proportion;

and the sending module is used for issuing the corresponding relation between the candidate path and the flow characteristic identifier to the first node.

Optionally, the path allocation module uses a summary value calculated by the traffic characteristic as a traffic characteristic identifier, where the traffic characteristic identifier is used to perform load sharing in the candidate path according to the traffic characteristic identifier.

Optionally, the path allocation module is further configured to determine, according to the remaining bandwidth of each candidate path included in the segment routing policy, a first ratio of the bearer traffic of each candidate path to the total remaining bandwidth if the controller cannot obtain the traffic statistic table, and send the first ratio of each candidate path and the corresponding candidate path to the head node, so that the head node forwards the traffic according to the first ratio of each candidate path.

Optionally, the path allocation module is further configured to determine a candidate path with the largest remaining bandwidth if the candidate path corresponding to each traffic feature identifier cannot be determined according to the first ratio and the second ratio, and generate a corresponding relationship between all traffic feature identifiers and the candidate path with the largest remaining bandwidth, so as to forward traffic corresponding to all traffic feature identifiers through the candidate path with the largest remaining bandwidth.

Optionally, the apparatus further comprises:

and the detection module is used for triggering the acquisition module to execute the acquisition of the flow statistic table of the head node on the candidate path included in the maintained segment routing strategy when detecting that any one or more of the path break, the link blockage and the arrival of the preset flow statistic table acquisition period exist.

The technical scheme provided by the embodiment of the specification can have the following beneficial effects: according to the traffic scheduling method and device provided by the disclosure, the proportion of the traffic rate corresponding to each hash value to the total traffic rate borne by the first node is determined according to the hash value of the application flow flowing through the first node, and the controller can continuously perform dynamic load sharing on the candidate path corresponding to the segment routing strategy according to the traffic corresponding to the hash value according to the traffic change, so that a proper candidate path is selected, and the scheduling effectiveness is greatly improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.

FIG. 1 is a prior art networking schematic provided by the present disclosure;

fig. 2 is a flow chart diagram of a traffic scheduling method provided by the present disclosure;

fig. 3 is a schematic structural diagram of a traffic scheduling apparatus provided in the present disclosure;

fig. 4 is a schematic structural diagram of a controller according to another embodiment of the present disclosure.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent 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 specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.

In the method provided by the present disclosure, the controller performs dynamic load sharing on the candidate path corresponding to the segment routing policy according to the change of the traffic of the device (that is, the traffic characteristic value also changes with the change of the traffic), so as to ensure that even if there is only one traffic (that is, the traffic is not uniform) within a certain time, the traffic can be allocated to a proper path for forwarding.

The present disclosure provides a traffic scheduling method, and fig. 2 shows a flow diagram of the traffic scheduling method provided by the present disclosure, and as shown in fig. 2, the method includes:

step 201, obtaining a flow statistics table of a head node on a candidate path included in a maintained segment routing policy, where the flow statistics table includes flow rates corresponding to different flow feature identifiers in the segment routing policy of the head node.

The controller maintains SR-Policy segment routing strategies of network devices in the whole network, and each segment routing strategy can comprise one or more candidate paths to guide traffic to be forwarded. The controller issues the segment routing strategy to the relevant network equipment, and the network equipment conducts the message diversion to the corresponding segment routing strategy through the priority or the diversion strategy, and then conducts the message forwarding according to the candidate paths included in the segment routing strategy.

The network device may perform statistics on traffic data via the network device, for example, may perform statistics on the rate of traffic under different segment routing policies via the network device.

For load sharing of traffic, traffic signatures may be calculated for different traffic, and the traffic signatures calculated for different traffic may be the same. And forwarding the traffic with the same traffic characteristic identifier according to the same strategy.

An implementation manner of calculating the traffic feature identifier is to use any one or more hash values of the calculated traffic in the quintuple information of the packet as the traffic feature identifier.

For convenience of description, the following embodiments take the traffic characteristic identifier as a hash value as an example for illustration. It should be understood that the flow characteristic identification may be determined in other manners.

In general, the head node of the candidate path may be used to maintain and count the traffic rates corresponding to different hash values of the device. Which may have one or more segment routing policies configured thereon. Specifically, as shown in table 1, table 1 shows an illustration of a traffic statistic table of the network device, where traffic rates corresponding to different hash values in each segment routing policy are maintained in the traffic statistic table. And then the sum of the flow rates corresponding to a certain hash value can be determined according to different segment routing strategies. For example, the traffic rate with the hash value of 0 is the sum of the traffic rates of different SR policies, that is, 5kb/s +6kb/s +4kb/s + … …, and the following description will take the example that the sum of the traffic rates with the hash value of 0 is 10M. The flow rate of different SR strategies corresponding to the hash value of 1 is 10kb/s +0kb/s +2kb/s +3kb/s + … ….

In one implementation, the controller may periodically obtain a traffic statistic table on a head node corresponding to a candidate path included in the segment routing policy. In another implementation manner, if the controller detects that there is a candidate path with an open circuit, a link congestion, and an SLA compliance that also does not meet the traffic demand, etc., any one or more of the above conditions are met, and at this time, the controller acquires a traffic statistical table of a first node on the candidate path included in the maintained segment routing policy.

TABLE 1

Step 203, determining a first proportion of the load-bearing flow of each candidate path in the total residual bandwidth according to the residual bandwidth of each candidate path included in the segment routing strategy; determining a second proportion of the traffic rate in the total traffic rate corresponding to the segment routing strategy of the head node according to the traffic rate corresponding to each traffic characteristic identifier; and determining a candidate path corresponding to each flow characteristic identifier according to the first proportion and the second proportion.

Taking the networking shown in fig. 1 as an example, if the candidate paths included in the segment routing Policy SR Policy1 are:

Segment-list1：PE1->P1->PE2；

Segment-list2：PE1->P2->PE2；

Segment-list3：PE1->P3->PE2；

Segment-list4：PE1->P4->PE2。

the remaining available bandwidth of the entire candidate path may be determined according to the remaining available bandwidth of the link corresponding to each candidate path. For example, the remaining available bandwidth of the link between PE1 and P1 is 20M, the remaining available bandwidth between P1 and PE2 is 30M, and then the remaining available bandwidth of the candidate path PE1- > P1- > PE2 is the minimum of the remaining available bandwidths of the individual links that make up the candidate link.

How to calculate the remaining available bandwidth of the path may adopt a manner in the prior art, which is not described in this embodiment again.

And determining the proportion of the flow capacity of each candidate path according to the residual available bandwidth of the candidate path included in the routing strategy. For example, if the remaining available bandwidth of the candidate path corresponding to Segment-list1 is 10M, the remaining available bandwidth of the candidate path corresponding to Segment-list2 is 20M, the remaining available bandwidth of the candidate path corresponding to Segment-list2 is 50M, and the remaining available bandwidth of the candidate path corresponding to Segment-list12 is 20M. Then it can be determined that the four candidate paths can share the traffic in a ratio of 1:2:5: 2. The proportions of the traffic that can be carried by the candidate paths corresponding to Segment-list 1-Segment-list 4 to the total remaining traffic are 1/10, 1/5, 1/2 and 1/5, respectively. For convenience of description, the ratio of the traffic that can be carried by the candidate path to the total remaining traffic is referred to as a first ratio hereinafter in this embodiment. The total remaining traffic is the sum of the remaining bandwidths of the candidate paths.

Further, the controller calculates the proportion of each hash value in each segment routing strategy in the current segment routing strategy in the total flow rate according to the acquired flow statistical table. For example, taking the example of table 2 as an illustration, the hash values of the traffic corresponding to the first node PE1 include 0-99, and the traffic rate corresponding to each hash value can be determined. The flow rate can be determined according to the sum of the flow rates of the hash values under different SR strategies.

Taking the flow rate corresponding to the hash value of 0 as 10M and the flow rate corresponding to the hash value of 1 as 20M as an example, if the total flow rate of the first node corresponding to the hash values of 0-99 is 100000M, the ratio of the flow rate corresponding to the hash value of 0 to the total flow rate is 10/100000, the ratio of the flow rate corresponding to the hash value of 1 to the sum of the total flow rates is 20/100000, and so on, the description of the ratio of the flow rate corresponding to the hash value of 2-99 to the total flow rate is omitted. Wherein the total traffic rate is determined according to the sum of the traffic rates of the respective hash values.

This was determined to be Segment-list 1: PE1- > P1- > PE2 can carry the proportion of the total traffic of 10%, and then the traffic accounting for 10% of the total rate in the hash values 0-99 can be allocated to the candidate path PE1- > P1- > PE 2; and allocating the traffic accounting for 20% of the total traffic rate in the hash values 0-99 to the candidate path PE1- > P2- > PE 2.

Traffic of 10% of the total rate may be composed of a plurality of small traffic, for example, a plurality of relatively small traffic allocations having a hash value of 0, a hash value of 1, a hash value of 3, a hash value of 4, etc. may be carried by PE1- > P1- > PE 2.

For a traffic rate of 10000 for a hash value of 2, accounting for 10% of the total traffic rate, since PE1- > P1- > PE2 can only carry 10% of the total traffic, and thus PE1- > P1- > PE2 cannot carry traffic with a hash value of 3, the traffic with a hash value of 3 can be assigned to PE1- > P3- > PE 2.

Therefore, the candidate paths matched with the flow corresponding to the hash values can be selected and converted into a knapsack problem in dynamic planning. Specifically, how to select the candidate path with the appropriate hash value is not described in this embodiment, and a dynamic programming algorithm in the prior art may be adopted.

In one possible result, the final candidate paths and the hash value may form a corresponding relationship as shown in table 2 below:

TABLE 2

Candidate path	First proportion of bearable bandwidth	Hash value
			PE1->P1->PE2	1/10	0、1、3、4……
PE1->P2->PE2	2/10	2、30
			PE1->P3->PE2	1/2	69、70、71……
PE1->P4->PE2	1/5	6、7、8、9……

Step 205, issuing the corresponding relation between the candidate path and the traffic characteristic identifier to the head node.

After the controller determines the hash value corresponding to the traffic that can be carried by each candidate path under each SR policy, the controller may issue the correspondence in table 2 to the head node, so as to instruct the head node to forward the traffic.

It can be seen that, in the traffic scheduling method provided by the present disclosure, the proportion of the traffic rate corresponding to each hash value to the total traffic rate borne by the first node is determined according to the hash value of the application flow flowing through the first node, so as to select a suitable candidate path. According to the prior art, the traffic corresponding to the hash value 2 can only be forwarded through the fixed PE1- > P1- > PE2, but in reality, the PE1- > P1- > PE2 cannot carry the large traffic corresponding to the hash value 2 at all. According to the method, a suitable candidate path can be selected for the traffic corresponding to each hash value according to the traffic statistics of the first node, and the scheduling effectiveness is greatly improved.

In the above-described technique of the embodiment, if a dynamic planning manner is adopted, it cannot be determined that a suitable candidate path is determined for each hash value, that is, the dynamic planning algorithm cannot output an effective solution, and a situation that may occur at this time is that end-to-end traffic of the first node and the last node may only have one flow, that is, traffic of hash values corresponding to the flow is very concentrated. At this time, all the traffic may be allocated to a candidate path, and the remaining allocable bandwidth of the candidate path may be the largest.

On the basis of the above embodiment, if the controller cannot obtain the traffic statistic table, determining, according to the remaining bandwidth of each candidate path included in the segment routing policy, a first proportion of the bearer traffic of each candidate path to the total remaining bandwidth, and sending each candidate path and the first proportion corresponding to each candidate path to the head node, so that the head node forwards the traffic according to the first proportion corresponding to each candidate path. I.e. at this time, the correspondence between the candidate paths and the hash values does not need to be calculated. The traffic is distributed according to the existing scheme in the prior art.

In addition, each candidate path of the segment routing strategy can be recalculated at this time, and the candidate path is determined according to the new residual bandwidth condition.

Fig. 3 is a schematic structural diagram of the traffic scheduling apparatus provided in the present disclosure, and as shown in fig. 3, the apparatus includes: an acquisition module 301, a path allocation module 302 and a sending module 303;

the obtaining module 301 is configured to obtain a traffic statistics table of a head node on a candidate path included in a maintained segment routing policy, where the traffic statistics table includes traffic rates corresponding to different traffic feature identifiers in the segment routing policy of the head node;

a path allocation module 302, configured to determine, according to the remaining bandwidth of each candidate path included in the segment routing policy, a first ratio of the bearer traffic of each candidate path to the total remaining bandwidth; determining a second proportion of the traffic rate in the total traffic rate of the head node according to the traffic rate corresponding to each traffic characteristic identifier; determining candidate paths corresponding to the flow characteristic identifications according to the first proportion and the second proportion;

the sending module 303 is configured to issue a correspondence between the candidate path and the traffic feature identifier to the head node.

In an optional implementation manner, the path allocation module 302 uses a summary value calculated by a feature of a traffic as a traffic feature identifier, where the traffic feature identifier is used to perform load sharing in a candidate path according to the traffic feature identifier.

In an optional implementation manner, the path allocating module 302 is further configured to determine, according to the remaining bandwidth of each candidate path included in the segment routing policy, a first ratio of the bearer traffic of each candidate path to the total remaining bandwidth if the controller cannot obtain the traffic statistical table, and send the first ratio of each candidate path to the head node, so that the head node forwards the traffic according to the first ratio corresponding to each candidate path.

In an optional implementation manner, the path allocating module 302 is further configured to determine a candidate path with the largest remaining bandwidth if the candidate path corresponding to each traffic feature identifier cannot be determined according to the first ratio and the second ratio, and generate a corresponding relationship between all the traffic feature identifiers and the candidate path with the largest remaining bandwidth, so that traffic corresponding to all the traffic feature identifiers is forwarded through the candidate path with the largest remaining bandwidth.

The device further comprises: a detecting module (not shown in the figure), configured to trigger the obtaining module to perform obtaining of the traffic statistic table of the head node on the candidate path included in the maintained segment routing policy when detecting that there is any one or more of a broken path, a blocked link, or a preset traffic statistic table obtaining period.

The present disclosure further provides a controller 40, and fig. 4 is a schematic structural diagram of a controller according to another embodiment of the present disclosure, as shown in fig. 4, the controller 40 includes a processor 401 and a memory 402, where the memory 402 is used to store program instructions, the processor 401 is used to call the program instructions stored in the memory, and when the processor 401 executes the program instructions stored in the memory 402, the controller is used to execute the method executed by the controller according to the above embodiment.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a readable storage medium, which includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned readable storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It will be understood that the present description is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.

The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. A traffic scheduling method is applied to a controller, and the method comprises the following steps:

2. The method according to claim 1, wherein the traffic characteristic identifier is a digest value calculated according to a characteristic of traffic, and the traffic characteristic identifier is used for performing load sharing in the candidate path according to the traffic characteristic identifier.

3. The method according to claim 1, wherein if the controller does not obtain the traffic statistic table, determining a first ratio of the bearable traffic of each candidate path to the total remaining bandwidth according to the remaining bandwidth of each candidate path included in the segment routing policy, and sending the first ratio of each candidate path to the head node, so that the head node forwards the traffic according to the first ratio of each candidate path.

4. The method according to claim 1, wherein if a candidate path corresponding to each traffic feature identifier cannot be determined according to the first ratio and the second ratio, a candidate path with a maximum residual bandwidth is determined, and a correspondence between all traffic feature identifiers and the candidate path with the maximum residual bandwidth is generated, so that traffic corresponding to all traffic feature identifiers is forwarded through the candidate path with the maximum residual bandwidth.

5. The method according to any one of claims 1-4, further comprising:

and when any one or more of the existence of path disconnection, the existence of link blockage or the arrival of a preset flow statistic table acquisition period is detected, acquiring the flow statistic table of the first node on the candidate path included in the maintained segment routing strategy.

6. An apparatus for traffic scheduling, the apparatus comprising:

7. The apparatus according to claim 6, wherein the path allocation module uses a summary value calculated from the characteristics of the traffic as a traffic characteristic identifier, and the traffic characteristic identifier is used for performing load sharing in the candidate path according to the traffic characteristic identifier.

8. The apparatus according to claim 6, wherein the path allocation module is further configured to, if the controller does not obtain the traffic statistic table, determine, according to remaining bandwidths of the candidate paths included in the segment routing policy, a first ratio of the bearer traffic of each candidate path to the total remaining bandwidth, and send the first ratio of each candidate path to the head node, so that the head node forwards the traffic according to the first ratio corresponding to each candidate path.

9. The apparatus according to claim 6, wherein the path allocation module is further configured to determine a candidate path with the largest remaining bandwidth if the candidate path corresponding to each traffic feature identifier cannot be determined according to the first ratio and the second ratio, and generate a corresponding relationship between all traffic feature identifiers and the candidate path with the largest remaining bandwidth, so as to forward traffic corresponding to all traffic feature identifiers through the candidate path with the largest remaining bandwidth.

10. The apparatus according to any one of claims 6-9, further comprising: