CN120567744A

CN120567744A - A method and device for optimizing SDWAN latency

Info

Publication number: CN120567744A
Application number: CN202510592831.0A
Authority: CN
Inventors: 张洪钏
Original assignee: China Telecom Cloud Technology Co Ltd
Current assignee: China Telecom Cloud Technology Co Ltd
Priority date: 2025-05-08
Filing date: 2025-05-08
Publication date: 2025-08-29

Abstract

The invention relates to the technical field of software-defined wide area networks and discloses a SDWAN time delay tuning method and device, wherein SDWAN comprises a controller, a plurality of POP devices and a plurality of CPE devices, the method comprises the steps that the controller creates a SDWAN backbone network between the POP devices, performs time delay detection on the SDWAN backbone network to obtain backbone network time delay data, creates a plurality of tunnels between the CPE devices and the POP devices, performs access terminal time delay detection on the tunnels respectively through the CPE devices to obtain access terminal time delay data, and determines end-to-end time delay data based on the backbone network time delay data and the access terminal time delay data and determines an optimal transmission link based on the end-to-end time delay data. The invention accurately calculates the end-to-end time delay data, thereby effectively improving the time delay tuning effect.

Description

SDWAN time delay optimization method and SDWAN time delay optimization device

Technical Field

The invention relates to the technical field of software-defined wide area networks, in particular to a SDWAN time delay tuning method and device.

Background

With the development of SDWAN (Software-DEFINED WIDE AREA Network), more and more enterprises choose to build an intranet using SDWAN, SDWAN has become an important tool for pushing enterprise digital transformation and improving business efficiency by providing flexible, intelligent and reliable Network connection.

However, in practical application SDWAN may face a time delay problem, and the relevant time delay tuning method cannot accurately reflect the real time delay from the transmitting end to the receiving end in SDWAN, so that the time delay tuning effect is poor.

Disclosure of Invention

In view of this, the invention provides a method and a device for adjusting and optimizing time delay of SDWAN, so as to solve the problem that the related method for adjusting and optimizing time delay cannot accurately reflect the real time delay from a transmitting end to a receiving end in SDWAN, and further the effect of adjusting and optimizing time delay is poor.

In a first aspect, the present invention provides a delay tuning method SDWAN, where SDWAN includes a controller, a plurality of POP devices, and a plurality of CPE devices, and the method includes:

the controller creates SDWAN backbone networks between the POP devices, and detects the time delay of the SDWAN backbone networks to obtain backbone network time delay data;

the controller creates a plurality of tunnels between the CPE equipment and the POP equipment, and respectively detects the delay of the access terminal of the plurality of tunnels through the CPE equipment to obtain delay data of the access terminal;

the controller determines end-to-end delay data based on backbone delay data and access end delay data, and determines an optimal transmission link based on the end-to-end delay data.

According to the SDWAN delay tuning method, the controller is used for creating the SDWAN backbone network between the POP devices, performing delay detection on the SDWAN backbone network to obtain backbone network delay data, creating a plurality of tunnels between the CPE devices and the POP devices, performing access terminal delay detection on the plurality of tunnels through the CPE devices to obtain access terminal delay data, finally accurately calculating end-to-end delay data based on the backbone network delay data and the access terminal delay data, and utilizing the end-to-end delay data to realize the selection of an optimal transmission link, so that the delay tuning effect is effectively improved.

In an alternative embodiment, the controller creates SDWAN backbone networks between POP devices, and performs delay detection on SDWAN backbone networks to obtain backbone network delay data, including:

The controller sets a source POP device and a destination POP device in the plurality of POP devices, and creates SDWAN backbone networks based on the source POP device and the destination POP device;

The controller randomly generates a plurality of detection source ports, and sends the detection source ports and the destination source ports corresponding to the destination POP equipment to the source POP equipment;

The source POP equipment transmits the message to the target source port through the detection source port to obtain a detection result, wherein the detection result is backbone network delay data corresponding to the source POP equipment;

the source POP devices sort the detection results, select a first preset number of source POP devices based on the sorting results, and send the first preset number of source POP devices and backbone network delay data corresponding to the first preset number of source POP devices to the controller.

According to the SDWAN delay tuning method, the backbone network is effectively constructed by setting the source POP equipment and the destination POP equipment, the source POP equipment detects the lowest delay transmitted by the backbone network through a plurality of different detection source ports, accurate detection of backbone network delay data is achieved, detection results are ordered, a first preset number of source POP equipment and backbone network delay data corresponding to the first preset number of source POP equipment are sent to the controller, and the efficiency of backbone network delay detection is improved.

In an alternative embodiment, the controller creates a plurality of tunnels between the CPE device and the POP device, and performs access end delay detection on the plurality of tunnels through the CPE device to obtain access end delay data, including:

The controller calculates the equipment distances between the CPE equipment and the first preset number of source POP equipment respectively, and selects a second preset number of source POP equipment based on the equipment distances, wherein the second preset number is smaller than the first preset number;

The controller creates a plurality of tunnels between the plurality of CPE devices and a second preset number of source POP devices;

CPE equipment respectively detects the time delay of the access terminal of the plurality of tunnels to obtain the time delay data of the access terminal.

According to the SDWAN delay optimization method, the device distances between the CPE devices and the source POP devices are calculated through the controller, the second preset number of source POP devices are selected by the device distances, the detection efficiency of the access terminal delay is improved, and a plurality of tunnels are established between the CPE devices and the second preset number of source POP devices and comprise the combination of the CPE devices and the POP devices, so that the detection process can fully consider deployment scenes of multiple transmission links and the POP devices of users, access terminal delay detection is carried out on the tunnels, and the accuracy of access terminal delay data is improved.

In an alternative embodiment, the controller determines end-to-end delay data based on backbone delay data and access end delay data, and determines an optimal transmission link based on the end-to-end delay data, comprising:

the controller determines end-to-end delay data based on backbone network delay data and access end delay data;

the controller scores the POP devices by using the end-to-end time delay data to obtain scores of the POP devices;

the controller selects an optimal transmission link between the CPE device and the POP device based on the plurality of POP device scores.

According to the SDWAN delay tuning method, the end-to-end delay data is accurately calculated based on the backbone network delay data and the access end delay data, the accuracy of the end-to-end delay data is improved, in the process of scoring a plurality of POP devices, the advantages and disadvantages of combination between the CPE devices are dynamically evaluated, the optimal transmission link between the CPE device and the POP device is selected according to the POP device score, and the delay tuning effect is improved.

In an alternative embodiment, the controller selects an optimal transmission link between the CPE device and the POP device based on the plurality of POP device scores, comprising:

The controller sorts the scores of the POP devices, and selects a main POP device and a standby POP device based on the sorting result;

the controller selects an optimal transmission link based on the priority of the connection links between the CPE device and the primary and backup POP devices.

According to the SDWAN time delay optimization method, the main POP equipment and the standby POP equipment are determined by sequencing the scores of the POP equipment, and the regularization and automation of the link selection process are realized and the consistency of the optimal transmission link is ensured by setting the priorities of different connection links between the CPE equipment and the main POP equipment and between the CPE equipment and the standby POP equipment.

In an alternative embodiment, the method further comprises:

when the optimal transmission link has a communication fault, the controller switches the access POP equipment in the optimal transmission link based on the priority of the connection link to obtain an updated optimal transmission link.

According to the SDWAN time delay optimization method, when the optimal transmission link has a communication fault, the access POP equipment in the optimal transmission link is switched by using the priority of the connection link, so that the automatic convergence of the optimal transmission link is realized, the continuity of the service is ensured, and the user experience is further improved.

In a second aspect, the present invention provides a time delay optimizing apparatus of SDWAN, the apparatus comprising an SDWAN, the SDWAN comprising a controller, a plurality of POP devices, and a plurality of CPE devices, the controller, the plurality of POP devices, and the plurality of CPE devices being interconnected;

The controller is used for creating SDWAN backbone networks between the POP devices, and carrying out time delay detection on the SDWAN backbone networks to obtain backbone network time delay data;

The controller is used for creating a plurality of tunnels between the CPE equipment and the POP equipment, and respectively detecting the delay of the access terminal of the plurality of tunnels through the CPE equipment to obtain delay data of the access terminal;

And the controller is used for determining end-to-end delay data based on the backbone network delay data and the access end delay data and determining an optimal transmission link based on the end-to-end delay data.

In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory and the processor are communicatively connected to each other, and the memory stores computer instructions, and the processor executes the computer instructions, thereby executing the time delay optimization method of SDWAN in the first aspect or any one of the corresponding embodiments.

In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method of delay tuning SDWAN of the first aspect or any of its corresponding embodiments.

In a fifth aspect, the present invention provides a computer program product comprising computer instructions for causing a computer to perform the method of delay tuning SDWAN of the first aspect or any of its corresponding embodiments.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a diagram of SDWAN networking architecture according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of a SDWAN delay tuning method according to an embodiment of the present invention;

FIG. 3 is a flow chart of another SDWAN delay tuning method according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a backbone latency probing flow according to an embodiment of the present invention;

FIG. 5 is a flow chart of a delay tuning method of yet another SDWAN according to an embodiment of the present invention;

Fig. 6 is a schematic diagram of an access-side delay detection flow according to an embodiment of the present invention;

FIG. 7 is a flow chart of a delay tuning method of yet another SDWAN according to an embodiment of the present invention;

FIG. 8 is a SDWAN routing policy diagram according to an embodiment of the invention;

fig. 9 is a schematic diagram of an optimal transmission link according to an embodiment of the present invention;

fig. 10 is a schematic diagram of an optimal transmission link after a failure of a main POP according to an embodiment of the present invention;

fig. 11 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

SDWAN applies the concept of SDN (Software-Defined Network) to a wide area Network scenario, and centrally manages multiple wide area Network connections by means of Software, providing intelligence and flexibility of Network services.

Along with the development of SDWAN, more and more enterprises select SDWAN to construct own enterprise intranet, the application scene of SDWAN is very wide and mainly comprises cross-regional branch office interconnection, intelligent networking, home office networking, cloud primary application, data center interconnection and telemedicine, meanwhile, the application scene of SDWAN covers multiple aspects of enterprise network, cloud computing, remote work, data center interconnection and the like, and SDWAN has become an important tool for promoting enterprise digital transformation and improving business efficiency by providing flexible, intelligent and reliable network connection.

VXLAN (Virtual Extensible LocalArea Network, virtual extended local area network) is a network virtualization technology applied in SDWAN, and aims to solve the expansibility and isolation limitations encountered by VLAN (Virtual Local Area Network ) technology in large data centers and cloud computing environments, and implements virtual network communication across physical network boundaries by constructing a virtual two-layer network over a three-layer IP (Internet Protocol ) network, and by encapsulating ethernet frames of the second layer in IP packets of the third layer, using MAC-in-UDP (a network tunneling encapsulation technology) encapsulation technology, so that the extension of the network is no longer limited by the limitations of physical devices.

VPP (Vector Packet Processing ) is a high-performance network packet processing technology, in application SDWAN, which allows network devices or software to process network packets with extremely low delay and extremely high throughput, and is used for accelerating network function virtualization (Network Functions Virtualization, abbreviated as NFV) and packet processing tasks in a data center network, and the core idea of VPP is to utilize parallel processing capability of a multi-core CPU (Central Processing Unit ) to realize efficient packet processing by optimizing various links of packet receiving, processing, forwarding and the like, and generally comprises a set of software libraries and APIs (Application Programming Interface, application programming interfaces) for realizing functions in a user mode, thereby bypassing a kernel-mode processing path and reducing the overhead of context switching and interrupt processing.

SDWAN the delay refers to the time required for data transmission from the source to the destination when SDWAN technology is used, and when there is a delay sensitive service for enterprise users, the delay is regarded as a key index, SDWAN is a novel network architecture, and aims to manage and control the wide area network in a software-defined manner, so as to provide more efficient and reliable network connection, however, SDWAN may still face the problem of overlarge delay in practical application.

The reason for the excessive time delay SDWAN is shown by an unreasonable network topology structure, which may cause that a data packet needs to pass through a plurality of nodes to reach a destination, so that transmission time is increased, time delay is generated, network bandwidth is insufficient, when the network bandwidth is insufficient to meet data transmission requirements, the data packet may need to be queued to be sent, so that time delay is generated, network equipment failure SDWAN may cause that the data packet needs to be resent or routed to be adjusted, time delay SDWAN is increased, network congestion is generated, and in some cases, network links may be congested due to overlarge traffic, so that data packet transmission speed is reduced, and time delay is generated.

Two common SDWAN delay solutions exist for the problems, namely, the first solution is to detect in a multi-outlet scene, detect preset addresses of the Internet according to the multiple outlets by identifying the multiple outlets capable of accessing the Internet in gateway equipment, obtain network quality evaluation values corresponding to the outlets respectively, and obtain target outlets accessing the Internet through a gateway according to the network quality evaluation values; the detection in the multi-outlet scene is a general method, and better link selection is realized by detecting different outlets, however, the detection results in the outlet delay of the Internet, which cannot truly reflect the access delay of SDWAN and further cannot reflect the end-to-end real delay of SDWAN.

The second scheme is to reduce overall time delay by using a specific algorithm, receive messages on multiple sub-stream connections connected by a multi-path transmission control protocol MPTCP (Multipath Transmission Control Protocol, multi-path transmission control protocol) through a network component, and determine that the messages of an MPTCP layer are out of order according to the received messages, the network component determines that no blocking message causing the out of order is received within a tolerance time, the tolerance time should be smaller than the maximum RTO in retransmission timeout time RTO (Retransmission Time Out, timeout retransmission mechanism) of the multiple sub-stream connections, and the network component sends retransmission indication messages of the blocking message on the target sub-stream connection in the multiple sub-stream connections.

In summary, a solution that can reduce all SDWAN traffic transmission delays simultaneously, and does not need to add extra operator-specific transmission links is a premise and guarantee that the SDWAN user experience is improved.

In order to solve the technical problems, the embodiment of the invention provides a SDWAN delay tuning method, which aims at the defect that the SDWAN transmission delay is overlarge, designs a delay tuning method suitable for SDWAN backbone networks through VXLAN transmission protocol based on SDWAN backbone networks, combines access end delay selection to achieve end-to-end optimal delay, can reduce all SDWAN transmission delays without adding extra operator special transmission links, the backbone networks run on the operator private network, different POP devices bear multi-tenant traffic through VXLAN tunnels, the POP devices run VPP processes, the operator private network has a plurality of equivalent transmission paths in two cities for the VXLAN messages, realizes traffic sharing through five-tuple hash, and detects a plurality of source ports with lowest delay by changing the VXLAN message source ports, so that the shortest path in the network can be fully utilized for the subsequent user traffic average allocation to the lowest path, the shortest path in the network can be prevented from appearing, the delay of the end-to-end needs to calculate the access network delay through the VXLAN tunnel, the shortest path is determined, and the priority is adjusted by comprehensively adjusting the shortest delay.

As shown in fig. 1, in the embodiment of the present invention, SDWAN includes a controller 101, a plurality of POPs (Point of Presence, access point) devices 102 and a plurality of CPE (Customer Premises Equipment, customer premise equipment) devices 103, where a switch is a device for customer access SDWAN, a user may directly access the switch to the CPE devices 103 at a relatively small site, a user may access the CPE devices 103 through a two-layer switch at a relatively large site, a user headquarters network may be interconnected with the CPE devices 103 through three-layer switches, and a LAN (LocalArea Network ) side may interface with more than one switch.

CPE device 103 is a common access device of SDWAN, usually located at a customer end, and can provide multiple access modes, including wired and wireless access, support multiple network protocols and transmission technologies, and ensure stable and reliable network connection for branch offices of an enterprise, and in SDWAN network, POP device 102 refers to devices disposed at various physical locations in the network, POP device 102 is mainly responsible for connecting different geographic locations, and provides network connection and transmission functions for implementing connection between branch offices.

By deploying POP devices 102, the branch office can access through the internet, thereby greatly reducing the cost, providing better flexibility and performance, connecting POP devices 102 through the private network of the operator, providing stable and reliable core networking capability, and the controller 101 establishes management connection with both CPE device 103 and POP device 102 and issues configuration.

In accordance with an embodiment of the present invention, there is provided a latency tuning method embodiment of SDWAN, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order other than that shown or described herein.

In this embodiment, a method for optimizing delay SDWAN is provided, which may be used in SDWAN described above, and fig. 2 is a flowchart of a method for optimizing delay SDWAN according to an embodiment of the present invention, as shown in fig. 2, where the flowchart includes the following steps:

in step S201, the controller creates SDWAN backbone networks between POP devices, and performs delay detection on the SDWAN backbone networks to obtain backbone network delay data.

Specifically, when the SDWAN backbone network is created, whether backbone network delay data exist locally or not needs to be checked, if backbone network delay data exist, the existing backbone network delay data and a source port corresponding to the backbone network delay data can be directly used without acquiring the backbone network delay data again, if backbone network delay data do not exist locally, delay detection is needed to be carried out on the SDWAN backbone network, and therefore backbone network delay data are obtained, real-time detection is not needed to be carried out on the backbone network delay data, and updating is carried out once a week.

In step S202, the controller creates multiple tunnels between the CPE device and the POP device, and performs access end delay detection on the multiple tunnels through the CPE device, so as to obtain access end delay data.

In step S203, the controller determines end-to-end delay data based on the backbone delay data and the access end delay data, and determines an optimal transmission link based on the end-to-end delay data.

According to the SDWAN delay tuning method, the controller is used for creating SDWAN backbone networks between the POP devices, performing delay detection on the SDWAN backbone networks to obtain backbone network delay data, creating a plurality of tunnels between the CPE devices and the POP devices, performing access terminal delay detection on the tunnels through the CPE devices to obtain access terminal delay data, and finally accurately calculating end-to-end delay data based on the backbone network delay data and the access terminal delay data, so that the accuracy of the end-to-end delay data is improved, an optimal transmission link is determined by utilizing the end-to-end delay data, and further the delay tuning effect is effectively improved.

In this embodiment, a method for optimizing delay SDWAN is provided, which may be used in SDWAN described above, and fig. 3 is a flowchart of a method for optimizing delay SDWAN according to an embodiment of the present invention, as shown in fig. 3, where the flowchart includes the following steps:

in step S301, the controller creates SDWAN backbone networks between POP devices, and performs delay detection on the SDWAN backbone networks to obtain backbone network delay data.

Specifically, the step S301 includes:

In step S3011, the controller sets a source POP device and a destination POP device among the plurality of POP devices, and creates SDWAN a backbone network based on the source POP device and the destination POP device.

Specifically, the enterprise interconnection is created by planning the CPE device to the POP device and simultaneously creating interconnection services (i.e. SDWAN backbone network) between different POP devices, wherein different POP devices are subordinate to the source end and the destination end, the delay from the source end to the destination end is determined by the transmission distance and the transmission device, and meanwhile, the delay from the source end to the destination end is generally a fixed value because of the fixed equipment and links deployed by the operator, but different flows can be converted into different equivalent links through five-tuple Hash (a technology for converting any length of input data into fixed length of output data through an algorithm) due to the load sharing behavior of multiple links when deployed by the operator, so that the situation of inconsistent delay corresponding to different flows can be generated.

In step S3012, the controller randomly generates a plurality of detection source ports, and sends the detection source ports and destination source ports corresponding to the destination POP device to the source POP device.

Specifically, as shown in fig. 4, the controller generates 100 random VXLAN source ports (i.e., probing source ports) and issues probing source ports and destination source ports to the source POP device.

And step S3013, the source POP device transmits the message to the destination source port through the detection source port to obtain a detection result, wherein the detection result is backbone network delay data corresponding to the source POP device.

Specifically, the source POP device generates 100 detection results by transmitting the message to the destination source port through the detection source port, and because the detection source port of the source POP device is continuously changed, the destination source port of the return path remains unchanged, and the detection results corresponding to different detection source ports can be obtained.

Further, although the source address and the destination address of the corresponding outer layer are consistent after the traffic in the backbone network is encapsulated by the VXLAN, even if the destination ports are consistent, the source port is inconsistent, which results in different backbone network time delays, so as to obtain different backbone network time delay data, because the VXLAN uses UDP (User Datagram Protocol, user data protocol) messages.

Step S3014, the source POP devices sort the detection results, select a first preset number of source POP devices based on the sorting results, and send the first preset number of source POP devices and backbone network delay data corresponding to the first preset number of source POP devices to the controller.

Specifically, after waiting for a period of time, the controller can actively acquire backbone network delay data from the source POP devices, the POP devices sort one hundred detection results, and return the ten source POP devices with the minimum delay and backbone network delay data corresponding to the ten source POP devices with the minimum delay to the controller.

Furthermore, in order to achieve a better backbone network delay detection effect, the target POP device can reversely detect the source POP device once to obtain backbone network delay data from the target POP device to the source POP device, and the controller can record bidirectional backbone network delay data between the source POP device and the target POP device, calculate an average value of the bidirectional backbone network delay data and take the average value as real delay data of the backbone network.

In step S302, the controller creates a plurality of tunnels between the CPE device and the POP device, and performs access end delay detection on the plurality of tunnels through the CPE device, so as to obtain access end delay data. Please refer to step S202 in the embodiment shown in fig. 2, which is not described herein.

In step S303, the controller determines end-to-end delay data based on the backbone delay data and the access end delay data, and determines an optimal transmission link based on the end-to-end delay data. Please refer to step S203 in the embodiment shown in fig. 2 in detail, which is not described herein.

In this embodiment, a method for optimizing delay SDWAN is provided, which may be used in SDWAN described above, and fig. 5 is a flowchart of a method for optimizing delay SDWAN according to an embodiment of the present invention, as shown in fig. 5, where the flowchart includes the following steps:

in step S501, the controller creates SDWAN backbone networks between POP devices, and performs delay detection on the SDWAN backbone networks to obtain backbone network delay data. Please refer to step S301 in the embodiment shown in fig. 3 in detail, which is not described herein.

In step S502, the controller creates multiple tunnels between the CPE device and the POP device, and performs access end delay detection on the multiple tunnels through the CPE device, so as to obtain access end delay data.

Specifically, the step S602 includes:

In step S5021, the controller calculates device distances between the CPE devices and a first preset number of source POP devices, and selects a second preset number of source POP devices based on the device distances, where the second preset number is smaller than the first preset number.

Specifically, as shown in fig. 6, before activation, the CPE devices in SDWAN need to send the hardware devices to the customer site, if the CPE devices are deployed in a software form, the virtual machine needs to be started first, then the corresponding version is deployed, when the CPE devices are online, the location information of the specific CPE devices can be obtained, and the controller calculates the device distances between the multiple CPE devices and the POP devices currently deployed, and selects four POP devices closest to each CPE device based on the device distances.

In step S5022, the controller creates a plurality of tunnels between the plurality of CPE devices and the second preset number of source POP devices.

Specifically, there are many POP devices and public network devices that disable PING (a network diagnostic tool) probing service, and CPE devices have multiple outlets, so that it is not possible to simply determine access delay with a PING probing service of a public network address, use a real tunnel to probe access delay, establish tunnels with four POP devices for each outlet interface of the CPE device, and establish eight tunnels with four POP devices assuming that the CPE device has two outlets.

In step S5023, the CPE device performs access end delay detection on the multiple tunnels, to obtain access end delay data.

In the interconnection scene, the delay of the access end of the source end is required to be focused, the delay of the access end of the destination end is also required to be focused, namely, the CPE equipment of the destination end also needs to perform access end delay detection on a plurality of POP equipment.

In step S503, the controller determines end-to-end delay data based on the backbone delay data and the access end delay data, and determines an optimal transmission link based on the end-to-end delay data. Please refer to step S303 in the embodiment shown in fig. 3 in detail, which is not described herein.

According to the SDWAN delay optimization method, the device distances between the CPE device and the source POP devices are calculated through the controller, the source POP devices are selected according to the device distances, the detection efficiency of the access terminal delay is improved, a plurality of tunnels are established between the CPE device and the source POP devices, the tunnels comprise the combination of the CPE devices and the POP devices, the detection process can fully consider deployment scenes of multiple transmission links and the POP devices of a user, the access terminal delay detection is carried out on the tunnels, and the accuracy of access terminal delay data is improved.

In this embodiment, a method for optimizing delay SDWAN is provided, which may be used in SDWAN described above, and fig. 7 is a flowchart of a method for optimizing delay SDWAN according to an embodiment of the present invention, as shown in fig. 7, where the flowchart includes the following steps:

In step S701, the controller creates SDWAN backbone networks between POP devices, and performs delay detection on the SDWAN backbone networks to obtain backbone network delay data. Please refer to step S501 in the embodiment shown in fig. 5 in detail, which is not described herein.

In step S702, the controller creates multiple tunnels between the CPE device and the POP device, and performs access-end delay detection on the multiple tunnels through the CPE device, so as to obtain access-end delay data. Please refer to step S502 in the embodiment shown in fig. 5 in detail, which is not described herein.

In step S703, the controller determines end-to-end delay data based on the backbone delay data and the access end delay data, and determines an optimal transmission link based on the end-to-end delay data.

Specifically, the step S703 includes:

In step S7031, the controller determines end-to-end delay data based on backbone delay data and access end delay data.

Specifically, the controller adds the delay data of the access end of the source end and the delay data of the destination end to the delay data of the backbone network to obtain the delay data of the end to end, meanwhile, the controller needs to calculate the delay data of the end to end under all conditions, and calculates the delay data of the end to end under all conditions in a traversing way, namely, the delay data of the access end of the source tunnel is selected, the delay data of the backbone network is added, and the delay data of the access end of the destination tunnel is added, so that the delay data of the end to end under all conditions can be calculated, and the more tunnels between the CPE equipment and the POP equipment are, and the more conditions are obtained.

In step S7032, the controller scores the POP devices by using the end-to-end delay data, so as to obtain scores of the POP devices.

Specifically, ten end-to-end delay data with the minimum delay are selected from the end-to-end delay data under all conditions, and POP devices at two ends are scored according to the ten end-to-end delay data.

In step S7033, the controller selects an optimal transmission link between the CPE device and the POP device based on the plurality of POP device scores.

In some alternative embodiments, step S7033 above includes:

And a step a1, the controller sorts the scores of the POP devices, and selects a main POP device and a standby POP device based on the sorting result.

Specifically, the smaller the end-to-end delay data is, the higher the score of the POP equipment is, the controller calculates the score condition of each POP equipment at two ends, each end selects two POP equipment with highest score, the POP equipment with higher score is used as main POP equipment, and the POP equipment with lower score is used as standby POP equipment.

In step a2, the controller selects an optimal transmission link based on the priority of the connection links between the CPE device and the primary and backup POP devices.

Specifically, AS shown in fig. 8, after the primary POP device and the secondary POP device are selected, an appropriate routing policy needs to be used to ensure that the user's traffic flows preferentially on the optimal transmission link, and the specific routing policy and SDWAN networking are inseparable, and in SDWAN, the backbone network operates under the same AS (Autonomous System, which is used to identify the unique number of the autonomous system in the internet), AS shown in fig. 8 AS100, AS200 or AS300, so AS to facilitate the primary and secondary selection of POP devices in different regions.

Further, the remote disaster recovery is used AS a data protection and service continuity policy, so that POP equipment can be ensured to quickly recover operation when facing sudden events such AS natural disasters, human errors and system faults, RR (Route Reflector) equipment is introduced in the same AS for transmitting routes and reducing the number of neighbors, the RR equipment receives Route updates from CPE equipment and reflects the Route updates to other CPE equipment instead of all BGP (Border Gateway Protocol ) peers, complexity of BGP networks and the size of routing tables are reduced, the reduction of the number of BGP sessions is facilitated, particularly, in a large-scale network, a backbone network supports routing of a plurality of tenants, the RR equipment can reflect Route information of all tenants, and the number of neighbors of the backbone network is further reduced.

Further, the POP and CPE device establish IBGP (BGP, an operation mode for exchanging routing information between routers in the same AS) neighbors, and routes of different transmission links between the POP and CPE devices are distinguished by using a WEIGHT value, where the higher the WEIGHT value is, the higher the priority corresponding to the transmission link is, and the transmission link with the highest score is used AS a main transmission link, and the main transmission link has the highest WEIGHT value.

Further, multiple transmission links of the same POP device are accessed, priority is distinguished according to scores, priorities from the CPE device to the standby POP device are smaller than those of the main POP, traffic forwarding is conducted by using the main POP preferentially, routing priorities are controlled by using MED values among backbone networks, MEDs are opposite to WEIGHT attributes, the smaller the MED values are, the higher the priorities of the corresponding transmission links are, and for the main POP device and the standby POP device accessed by the same CPE device, the MED values of the main POP device are smaller than those of the standby POP device, and the MED values are required to be set through a routing strategy.

Further, in the normal traffic forwarding path, the CPE device first performs packet receiving processing, where packet receiving is performed on the LAN side, and if a single-arm deployment mode is adopted, a service packet of a user may come in from the WAN (Wide Area Network ) side, and the received service packet directly queries the route. Because of the access mode of multiple interfaces and multiple POPs, a plurality of next hops exist at the same place, BGP only prefers the next hop with the lowest time delay to send to VPP through WEIGHT configuration, CPE equipment communicates with POP equipment through an IPSec (Internet Protocol Security ) tunnel, wherein the IPSec tunnel is a security communication tunnel based on the IPsec protocol, and security in the data transmission process is ensured by encrypting and verifying the content of a data packet.

Further, as shown in fig. 9, after the data traffic arrives at POP1, the POP device carries the multi-tenant traffic, each tenant has its own independent routing table, and has multiple next hops to reach the destination POP device, and depending on BGP preferred policy, the transmission link to POP3 is selected as the main transmission link, the traffic between POP1 and POP3 is forwarded away from the VXLAN tunnel, and a specific source port is needed to be selected when the VXLAN tunnel is encapsulated, where the source port is issued by the controller, allocated by the tenant, and each tenant uses the same source port, after the traffic arrives at POP3, the traffic is forwarded to CPE2 using the IPSEC tunnel, so as to reach the other end device of the user, and an optimal transmission link is obtained, and the transmission link of the return path is consistent with the optimal transmission link, that is, by setting a proper routing priority, the consistency of the bidirectional transmission link is ensured, where the dotted line indicates the optimal transmission link, and the solid line indicates the standby transmission link.

Further, when the optimal transmission link has a communication fault, the controller switches the access POP equipment in the optimal transmission link based on the priority of the connection link, so as to obtain an updated optimal transmission link.

Further, as shown in fig. 10, the solid line indicates the optimal transmission link with failure, the broken line indicates the updated optimal transmission link, and when the transmission link from the CPE1 to the POP1 fails, the traffic path is automatically switched, and the failure is that the outgoing interface link of the POP1 is disconnected or the POP1 device is dead, and the outgoing path from the switch 1 to the switch 2 is that of the switch 1, the CPE1, the POP2, the POP3, the CPE2 and the switch 2.

Further, when the traffic arrives at the CPE1, POP2 is selected according to the routing priority of BGP, the subsequent flow is not different from the optimal transmission link, when the backhaul packet arrives at the POP3, POP2 can only be used as the next hop, after BGP neighbors between the CPE1 and the POP1 are disconnected, the corresponding network segment routes can be automatically cancelled, that is, the consistency of the back-and-forth transmission links can be ensured, in order to accelerate the convergence speed of BGP, a BFD (Bidirectional Forwarding Detection ) detection function can be configured on a tunnel, the BGP neighbors are disconnected when the BGP is abnormal, and meanwhile, the multi-interface redundant paths on the WAN side are switched through BGP, if the corresponding links have faults, the BGP can automatically switch links.

The embodiment also provides a SDWAN delay tuning device, which is used for implementing the foregoing embodiments and preferred embodiments, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

The embodiment provides a SDWAN delay optimizing device, as shown in fig. 1, which comprises an SDWAN, wherein the SDWAN comprises a controller 101, a plurality of POP devices 102 and a plurality of CPE devices 103, and the controller 101, the POP devices 102 and the CPE devices 103 are connected with each other;

The controller 101 is configured to create SDWAN a backbone network between POP devices 102, and perform delay detection on the SDWAN backbone network to obtain backbone network delay data.

The controller 101 is further configured to create a plurality of tunnels between the CPE device 103 and the POP device 102, and perform access-end delay detection on the plurality of tunnels through the CPE device 103, so as to obtain access-end delay data.

The controller 101 is further configured to determine end-to-end delay data based on the backbone delay data and the access end delay data, and determine an optimal transmission link based on the end-to-end delay data.

Further functional descriptions of the above respective modules and units are the same as those of the above corresponding embodiments, and are not repeated here.

The time delay optimizing device SDWAN in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application SPECIFIC INTEGRATED Circuit) Circuit, a processor and a memory that execute one or more software or firmware programs, and/or other devices that can provide the above functions.

The embodiment of the invention also provides computer equipment, which is provided with the time delay optimizing device SDWAN shown in the figure 1.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a computer device according to an alternative embodiment of the present invention, and as shown in fig. 11, the computer device includes one or more processors 10, a memory 20, and interfaces for connecting components, including a high-speed interface and a low-speed interface. The various components are communicatively coupled to each other using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions executing within the computer device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In some alternative embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories. Also, multiple computer devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 10 is illustrated in fig. 11.

The processor 10 may be a central processor, a network processor, or a combination thereof. The processor 10 may further include a hardware chip, among others. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general-purpose array logic, or any combination thereof.

Wherein the memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform a method for implementing the embodiments described above.

The memory 20 may include a storage program area that may store an operating system, application programs required for at least one function, and a storage data area that may store data created according to the use of the computer device, etc. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, memory 20 may optionally include memory located remotely from processor 10, which may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The memory 20 may comprise volatile memory, such as random access memory, or nonvolatile memory, such as flash memory, hard disk or solid state disk, or the memory 20 may comprise a combination of the above types of memory.

The computer device further comprises input means 30 and output means 40. The processor 10, memory 20, input device 30, and output device 40 may be connected by a bus or other means, for example in fig. 11.

The input device 30 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the computer apparatus, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointer stick, one or more mouse buttons, a trackball, a joystick, and the like. The output means 40 may include a display device, auxiliary lighting means (e.g., LEDs), tactile feedback means (e.g., vibration motors), and the like. Such display devices include, but are not limited to, liquid crystal displays, light emitting diodes, displays and plasma displays. In some alternative implementations, the display device may be a touch screen.

The embodiments of the present invention also provide a computer readable storage medium, and the method according to the embodiments of the present invention described above may be implemented in hardware, firmware, or as a computer code which may be recorded on a storage medium, or as original stored in a remote storage medium or a non-transitory machine readable storage medium downloaded through a network and to be stored in a local storage medium, so that the method described herein may be stored on such software process on a storage medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. The storage medium may be a magnetic disk, an optical disk, a read-only memory, a random-access memory, a flash memory, a hard disk, a solid state disk, or the like, and further, the storage medium may further include a combination of the above types of memories. It will be appreciated that a computer, processor, microprocessor controller or programmable hardware includes a storage element that can store or receive software or computer code that, when accessed and executed by the computer, processor or hardware, implements the methods illustrated by the above embodiments.

Portions of the present invention may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or aspects in accordance with the present invention by way of operation of the computer. Those skilled in the art will appreciate that the existence of computer program instructions in a computer-readable medium includes, but is not limited to, source files, executable files, installation package files, and the like, and accordingly, the manner in which computer program instructions are executed by a computer includes, but is not limited to, the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled programs, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed programs. Herein, a computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by a computer.

Although embodiments of the present invention have been described in connection with the accompanying drawings, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope of the invention as defined by the appended claims.

Claims

1. A method for optimizing latency of an SDWAN, wherein the SDWAN includes a controller, multiple Point of Presence (POPs) and multiple Customer Premises Equipment (CPEs); the method comprises:

The controller creates an SDWAN backbone network between the POP devices, performs delay detection on the SDWAN backbone network, and obtains backbone network delay data;

The controller creates a plurality of tunnels between the CPE device and the POP device, and performs access end delay detection on each of the plurality of tunnels through the CPE device to obtain access end delay data;

The controller determines end-to-end delay data based on the backbone network delay data and the access end delay data, and determines an optimal transmission link based on the end-to-end delay data.

2. The method according to claim 1, wherein the controller creates an SDWAN backbone network between the POP devices, performs latency detection on the SDWAN backbone network, and obtains backbone network latency data, comprising:

The controller sets a source POP device and a destination POP device in the plurality of POP devices, and creates an SDWAN backbone network based on the source POP device and the destination POP device;

The controller randomly generates a plurality of detection source ports, and sends the detection source ports and the destination source ports corresponding to the destination POP device to the source POP device;

The source POP device transmits the message to the destination source port through the detection source port to obtain a detection result; wherein the detection result is the backbone network delay data corresponding to the source POP device;

The source POP device sorts the detection results, selects a first preset number of source POP devices based on the sorting results, and sends the first preset number of source POP devices and the backbone network delay data corresponding to the first preset number of source POP devices to the controller.

3. The method according to claim 1, wherein the controller creates multiple tunnels between the CPE device and the POP device, and performs access end delay detection on each of the multiple tunnels through the CPE device to obtain access end delay data, comprising:

The controller calculates the device distances between the plurality of CPE devices and the first preset number of source POP devices, respectively, and selects a second preset number of source POP devices based on the device distances; wherein the second preset number is smaller than the first preset number;

The controller creates a plurality of tunnels between the plurality of CPE devices and the second preset number of source POP devices;

The CPE device performs access end delay detection on the multiple tunnels respectively to obtain the access end delay data.

4. The method according to claim 1, wherein the controller determines end-to-end delay data based on the backbone network delay data and the access terminal delay data, and determines the optimal transmission link based on the end-to-end delay data, comprising:

The controller determines end-to-end delay data based on the backbone network delay data and the access terminal delay data;

The controller scores the plurality of POP devices using the end-to-end delay data to obtain a plurality of POP device scores;

The controller selects the optimal transmission link between the CPE device and the POP device based on the scores of the multiple POP devices.

5. The method according to claim 4, wherein the controller selects the optimal transmission link between the CPE device and the POP device based on the scores of the plurality of POP devices, comprising:

The controller sorts the scores of the plurality of POP devices, and selects a primary POP device and a backup POP device based on the sorting result;

The controller selects the optimal transmission link based on the connection link priority between the CPE device and the primary POP device and the backup POP device.

6. The method according to claim 5, further comprising:

When a connectivity failure occurs in the optimal transmission link, the controller switches the access POP device in the optimal transmission link based on the connection link priority to obtain an updated optimal transmission link.

7. A SDWAN latency optimization device, characterized in that the device includes the SDWAN, the SDWAN includes: a controller, multiple POP devices, and multiple CPE devices, the controller, the multiple POP devices, and the multiple CPE devices are interconnected;

The controller is configured to create an SDWAN backbone network between the POP devices, perform latency detection on the SDWAN backbone network, and obtain backbone network latency data;

The controller is configured to create a plurality of tunnels between the CPE device and the POP device, and perform access end delay detection on each of the plurality of tunnels through the CPE device to obtain access end delay data;

The controller is configured to determine end-to-end delay data based on the backbone network delay data and the access end delay data, and determine an optimal transmission link based on the end-to-end delay data.

8. A computer device, comprising:

A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the SDWAN delay tuning method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to enable a computer to execute the SDWAN delay tuning method according to any one of claims 1 to 6.

10. A computer program product, characterized in that it comprises computer instructions, wherein the computer instructions are used to enable a computer to execute the SDWAN delay optimization method according to any one of claims 1 to 6.