WO2019011338A1

WO2019011338A1 - Method for determining shortest path and controller

Info

Publication number: WO2019011338A1
Application number: PCT/CN2018/095702
Authority: WO
Inventors: 黄荣锋; 盛镇醴; 李翰林
Original assignee: 华为技术有限公司
Priority date: 2017-07-13
Filing date: 2018-07-13
Publication date: 2019-01-17
Also published as: CN109257287B; CN109257287A

Abstract

A method for determining a shortest path and a controller, the method comprising: determining a multiplier corresponding to a shortest path that is determined on the basis of a Lagrange Relaxation based Aggregatd Cost (LARAC) algorithm from a source node to a sink node within a network topology; determining the weight of each side in the network topology according to the multiplier, and according to the weight of each side, determining K paths, the K paths being K paths which are in sequence from smallest path weight to largest path weight among the paths from the source node to the sink node in the network topology, and K being a positive integer greater than 1, wherein the weight of a path is the sum of the weights of all sides in the path; the sum of first attributes of the paths among the K paths is made to be less than or equal to a preset threshold, a path having the smallest sum of path second attributes is determined as a first path, and the first path serves as the shortest path from the source node to the sink node in the network topology.

Description

Shortest path determination method and controller

This application claims the priority of the Chinese Patent Application filed on Dec. 13, 2017, the National Patent Office, Application No. 201710571119.8, entitled "A Shortest Path Determination Method and Controller", the entire contents of which are incorporated herein by reference. In the application.

Technical field

The present application relates to the field of communications technologies, and in particular, to a shortest path determining method and a controller.

Background technique

With the development of communication technology, Software Defined Networking (SDN), as a new type of network architecture, has received more and more attention. The core idea of SDN is to separate the control plane of the network from the forwarding plane. The SDN controller on the control plane can uniformly control the forwarding of data in a network on the forwarding plane to achieve flexible deployment of network traffic. When the SDN controller controls the forwarding of data in a certain network, for the network where the source node and the sink node of a certain service are located, the path set of each node in the network may be determined first, and the path set of each node includes The path from the source node to the node is determined based on the path set of each node, and the service path from the source node to the sink node is determined, and then the service data between the source node and the sink node is forwarded according to the service path. The source node refers to a node that serves as a source to send service data, and the sink node refers to a node that acts as a sink to receive service data.

When the SDN controller performs the route calculation between the source node and the sink node, the factors generally considered are relatively simple. For example, only factors such as cost or delay are considered to obtain a quality of service (Quality of Service, QoS) path, so that the calculated shortest path may not be the optimal path. For example, in the scenario where there are multiple heavy edges between two nodes in the network, the advantage of the determined shortest path is not High, can not meet the actual application, so how to obtain a better shortest path is an urgent problem to be solved.

Summary of the invention

The present application provides a shortest path determination method and controller for solving the problem of how to determine a shortest path between a source node and a sink node.

An embodiment of the present application provides a shortest path determining method, where the method includes:

Determining a multiplier corresponding to the shortest path determined by the Lagrangian relaxation algorithm from the source node to the sink node in the network topology; the multiplier corresponding to the shortest path determined by the Lagrangian relaxation algorithm is a fixed value, which is pulling In the last iteration of the Grande relaxation algorithm, the multiplier corresponding to the shortest path is determined.

Determining the weight of each edge in the network topology according to the multiplier, and determining K paths according to the weight of each edge. The K paths are K paths whose weights in the path from the source node to the sink node in the network topology are small to large, and K is a positive integer greater than 1; the weight of the path The sum of the weights of all the edges in the path.

Determining, as the first path, a path in which the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the sum of the second attributes of the paths is the first path The shortest path from the source node to the sink node. The threshold is a user equipment that requests the SDN controller to determine the shortest path from the source node to the sink node, and is set according to its actual network requirement. Of course, the preset threshold may also be a default value, or the SDN controller may be based on the actual situation of the network. Settings.

The first attribute is an attribute of a link corresponding to the edge in the network topology, and the second attribute is another attribute of the link corresponding to the edge in the network topology different from the first attribute. The sum of the first attribute of the path is the sum of the first attributes of all edges in the path, and the sum of the second attributes of the path is the sum of the second attributes of all sides in the path.

According to the method provided by the embodiment of the present application, after determining the multiplier corresponding to the shortest path determined by the Lagrangian relaxation algorithm, determining the weight of each edge in the network topology according to the multiplier, and according to each The weight of the edge determines the path with the least weight of the K paths. Finally, the first shortest path is determined from the K paths. In the embodiment of the present application, the determined first path is based on the constraint that the first attribute of the path is less than or equal to the preset threshold, and the minimum constraint condition of the second attribute of the path is considered, so The first path is better than the path determined only according to the constraint of the first attribute of the path and less than or equal to the preset threshold, that is, the determined second attribute of the first path is determined by the LARAC algorithm. The sum of the second attributes of the shortest path is smaller, thereby achieving a better shortest path.

In some embodiments, before determining the weight of each edge in the network topology according to the multiplier, the method further includes:

Make sure the following preset conditions are met:

The node density of the network topology is greater than or equal to a preset node density, and the average difference coefficient of the shortest path determined based on the LALAC algorithm is less than or equal to a preset difference coefficient.

And determining, by using the foregoing method, that the node density is greater than or equal to a preset node density, and determining that the average difference coefficient of the shortest path determined by the LALAC algorithm is less than or equal to a preset difference coefficient, determining a currently obtained basis The shortest path determined by the LALAC algorithm is not an optimal path, so that the first shortest path can be determined according to the multiplier corresponding to the shortest path determined by the LALAC algorithm, thereby improving the accuracy of determining the shortest path between the source node and the sink node. Sex.

In some embodiments, if it is determined that the shortest path determined by the LALAC algorithm does not satisfy the preset condition, determining the shortest path determined based on the LALAC algorithm as a source node in the network topology The shortest path to the sink node.

In some embodiments, the average difference coefficient of the shortest path determined based on the LALAC algorithm satisfies the following formula:

Av(D)=|E|×2/|V|

Where av(D) is the node density of the network topology, E is the number of edges included in the network topology, and V is the number of nodes included in the network topology.

The preset node density is specifically determined, which is not limited by the embodiment of the present application. The preset node density may be used to determine whether there are multiple heavy edges between nodes in the network topology, thereby determining the minimum determined based on the LARAC algorithm. Whether the path is optimal.

In some embodiments, the average difference coefficient of the second path satisfies the following formula:

Where av(CV) is the average difference coefficient of the shortest path determined by the LALAC algorithm,

The CV _i is a difference coefficient of the weight of the heavy edge between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, where S _i is the shortest path determined by the LARAC algorithm The standard deviation of the weight of the heavy edge between the adjacent two nodes corresponding to the i-edge,

An average value of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, n is a positive integer greater than or equal to 1, and n is the based on the LARAC The shortest path determined by the algorithm includes the number of edges.

The preset difference coefficient is specifically determined. The embodiment of the present application is not limited to this. The preset node density and the preset difference coefficient may be used to determine whether there are multiple heavy edges between nodes in the network topology, thereby determining the basis. Whether the shortest path determined by the LALAC algorithm is optimal.

In some embodiments, determining a weight of each edge in the network topology based on the multiplier includes:

For any one of the network topologies, the sum of the product of the first attribute of the edge and the multiplier and the second attribute of the edge is determined as the weight of the edge.

In some embodiments, after the first path is used as the shortest path from the source node to the sink node in the network topology, the method further includes:

Generating a routing table from the source node to the sink node according to the first path.

In some embodiments, the first attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate;

The second attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate.

In some embodiments, the cost is a fee or distance or energy consumption.

In the embodiment of the present application, the cost refers to the overhead required when the data packet arrives from the sending node in the network topology to the receiving node in the network topology. For example, it may refer to the cost or energy consumption of the data packet from the sending node in the network topology to the receiving node in the network topology, and may also indicate that the data packet arrives at the receiving node in the network topology from the sending node in the network topology. The distance of the link that passes through.

Delay is the time it takes for a packet to travel through a channel and then reach a receiving node in the network topology after it is sent from the transmitting node in the network topology. Delay jitter refers to the change in delay and can reflect the degree of change in delay. The packet loss rate refers to the ratio of the number of lost data packets to the total number of transmitted data packets in the data packets sent from the sending node in the network topology to the receiving node in the network topology.

An embodiment of the present application provides a controller, including:

a receiving module, configured to receive source node information and sink node information; the source node information indicates a source node, and the sink node information indicates a sink node;

a storage module, configured to use the storage network topology information, where the network topology information includes a network topology and a first attribute and a second attribute of each edge in the network topology, where the first attribute is the network topology An attribute of the link corresponding to the middle side, where the second attribute is another attribute of the link corresponding to the edge in the network topology that is different from the first attribute;

a processing module, configured to determine a multiplier corresponding to a shortest path determined by a Lagrangian relaxed LALAC algorithm from a source node to a sink node in a network topology; determining, according to the multiplier, a weight of each edge in the network topology And determining K paths according to the weight of each edge, where the K paths are K paths whose weights in the path from the source node to the sink node in the network topology are small to large, K is a positive integer greater than 1; a weight of the path is a sum of weights of all edges in the path; a sum of first attributes of the paths in the K paths is less than or equal to a preset threshold, and a second attribute of the path And the smallest path is determined as a first path, the first path being a shortest path from the source node to the sink node in the network topology, wherein a sum of the first attributes of the path is a first attribute of all edges in the path And the sum of the second attribute of the path and the sum of the second attributes of all edges in the path.

In some embodiments, the processing module is further configured to:

Make sure the following preset conditions are met:

In some embodiments, the node density of the network topology satisfies the following formula:

Av(D)=|E|×2/|V|

Where av(CV) is the average difference coefficient of the second path,

CV _i is a difference coefficient of the weight of the heavy edge between the adjacent two nodes corresponding to the i-th edge of the second path, and S _i is the weight between the adjacent two nodes corresponding to the i-th edge in the second path The standard deviation of the edge weights,

An average value of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge of the second path, where n is a positive integer greater than or equal to 1, and n is the number of edges included in the second path.

In some embodiments, the processing module is specifically configured to:

In some embodiments, the processing module is further configured to:

In some embodiments, the storage module is further configured to store a routing table;

The processing module is further configured to instruct the storage module to update the routing table, where the routing table is used to route a message between the source node and the sink node.

The present application also provides a computer readable storage medium storing computer software instructions for use in any of the designed functions of any of the shortest path determination methods described above, when the instructions are run on a computer Computer software instructions for performing any of the functions of any of the shortest path determination methods described above.

The embodiment of the present application also provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the shortest path determination method described in the above aspects.

DRAWINGS

FIG. 1 is a schematic diagram of a scenario applicable to an embodiment of the present application;

2 is a schematic flowchart of a method for determining a shortest path according to an embodiment of the present application;

FIG. 3 is a schematic diagram of path calculation according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a network topology structure according to an embodiment of the present application;

FIG. 5 is a schematic flowchart of a method for determining a shortest path according to an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a controller according to an embodiment of the present application;

FIG. 7 is a schematic structural diagram of a controller according to an embodiment of the present application.

Detailed ways

The present application will be further described in detail below with reference to the accompanying drawings.

The embodiments of the present application can be applied to various mobile communication systems, such as: Global System of Mobile communication (GSM) system, Code Division Multiple Access (CDMA) system, and Wideband Code Division Multiple Access (Wideband). Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, Advanced Long Term Evolution (LTE-A) system, general purpose Other mobile communication systems such as the Universal Mobile Telecommunication System (UMTS), the evolved Long Term Evolution (eLTE) system, and the 5G system (for example, New Radio (NR) system).

The execution subject of the shortest path determining method provided by the embodiment of the present application may be a device such as an SDN controller. As shown in FIG. 1 , it is a schematic diagram of a scenario applicable to an embodiment of the present application. The SDN controller collects network topology information of the underlying network topology in real time through the southbound interface. The network topology information includes but is not limited to the number of nodes included in the network, the number of edges, the cost and delay of each edge, and the nodes adjacent to each node. And other information. When the northbound interface of the SDN controller receives the downlink service request sent by the user equipment or the like, the SDN controller calculates the shortest path that satisfies the constraint condition for the source node and the sink node of the service. The SDN controller generates a forwarding routing table according to the shortest path and sends it to the underlying transponder through the southbound interface. The underlying device can send the service packet from the source node to the sink node.

As shown in FIG. 2, a schematic flowchart of a shortest path determining method provided by an embodiment of the present application is shown. In FIG. 2, the execution subject may be a device such as an SDN controller. For convenience of description, the SDN controller is described as an execution subject in the following description.

Referring to Figure 2, the method includes:

Step 201: Determine a multiplier corresponding to the shortest path determined by the Lagrangian relaxation algorithm from the source node to the sink node in the network topology;

Step 202: Determine weights of each edge in the network topology according to the multiplier, and determine K paths according to weights of each edge.

The K paths are K paths whose weights in the path from the source node to the sink node in the network topology are small to large, and K is a positive integer greater than 1; the weight of the path The sum of the weights of all the edges in the path.

Step 203: Determine a path in which the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the path of the second attribute of the path is the smallest path, where the first path is the network. The shortest path from the source node to the sink node in the topology.

Before step 201, the SDN controller can obtain network topology information of the underlying network topology from the southbound interface, thereby determining the entire network topology. The network topology information includes information such as the connection relationship between each node in the network topology. The southbound interface is an interface that provides access and management to other vendors or operators, that is, a downward interface.

After receiving the service request message sent by the user equipment, the SDN controller may determine the source node according to the source node information in the service request message, determine the sink node and other information according to the information of the sink node, and the service request message may further include a constraint condition. The SDN controller can determine the shortest path that satisfies the constraint based on the constraints.

In step 201, the SDN controller determines, according to a Lagrangian Relaxation Based Aggregated Cost (LARAC) algorithm, that the shortest path from the source node to the sink node is a path that satisfies the constraint condition, and the constraint condition may be the first path. The sum of the attributes is less than or equal to the preset threshold. In the embodiment of the present application, the process of determining the shortest path based on the Lagrangian relaxation algorithm may be as follows:

Step 1: Using the Dijkstra algorithm to calculate the second attribute and the minimum path P1 of the path in the path from the source node to the sink node in the network topology, if the sum d(P1) of the first attribute of the path P1 is less than or equal to a preset threshold, Then it is determined that the path P1 is the shortest path determined based on the LALAC algorithm, and the result is output and the entire algorithm is ended, otherwise it is transferred to Step 2.

The first attribute may be any one of a cost, a delay, a delay jitter, and a packet loss rate, and the second attribute may be any one of a cost, a delay, a delay jitter, and a packet loss rate. For example, the first attribute can be a time delay and the second attribute can be a cost.

Delay is the time it takes for a packet to travel through a channel and then reach a receiving node in the network topology after it is sent from the transmitting node in the network topology. The delay can also be referred to as delay or delay (transmission) time and the like. Delay jitter refers to the change in delay and can reflect the degree of change in delay.

The packet loss rate refers to the ratio of the number of lost data packets to the total number of transmitted data packets in the data packets sent from the sending node in the network topology to the receiving node in the network topology.

Step 2: Calculate the first attribute and the minimum path P2 of the path in the path from the source node to the sink node in the network topology by using the Dijkstra algorithm, and determine if the sum d(P2) of the first attribute of the path P2 is greater than a preset threshold. There is no shortest path that meets the first attribute that satisfies the path and is less than or equal to the preset threshold, otherwise it goes to Step 3.

Step 3: Calculate the multiplier according to the path P1 and the path P2, and update the weight of each edge in the entire network topology by using the formula (1) according to the calculated multiplier.

Where formula (1) is:

c _λi =c _i +λ×d _i ········(1)

In the formula (1), c _λi is the weight of the ith side, c _i is the first attribute of the ith side, d _i is the second attribute of the ith side, and λ is the multiplier.

Wherein, the calculated multiplier λ satisfies the formula (2):

In the formula (2), c(P2) is the sum of the second attribute of the path P2, and c(P1) is the sum of the second attribute of the path P1.

Step 4: Calculate the path P3 of the minimum weight of the source node s to the sink node t using the Dijkstra algorithm.

Step 5: If the weight d _λ (P3) of the path P3 is equal to the weight d _λ (P2) of the path P2, determine that the path P2 is the shortest path from the source node to the sink node, output the result and end the entire algorithm, otherwise use the path P3 Replace path P2 and go to Step 3.

It should be noted that the shortest path determined by the LALAC algorithm may not be the shortest path from the source node to the sink node. In order to determine the shortest path from the source node to the sink node, the embodiment of the present application will be based on LALAC. The shortest path determined by the algorithm is optimized to obtain a substantial shortest path from the source node to the sink node. The process of optimization is described later and will not be described here.

In conjunction with the foregoing process, in the embodiment of the present application, the multiplier is an iterative value in the iterative process of the algorithm, which is a variable, but after determining the shortest path based on the LALAC algorithm, the multiplier corresponding to the shortest path determined by the LALAC algorithm is a The fixed value is the multiplier corresponding to the shortest path during the last iteration of the LARAC algorithm.

In the above method, the two constraint parameters (the first attribute and the second attribute) in the network topology are aggregated into a constraint parameter by using the multiplier λ: weight, so that the Dijkstra algorithm can be used to solve the network topology in which multiple constraint parameters exist. The shortest path problem. However, in the above method, only one constraint parameter is considered, that is, the sum of the first attributes of the path is less than or equal to a preset threshold, so the calculated shortest path may not be optimal.

In some embodiments, after the SDN controller determines the shortest path based on the LALAC algorithm, it may be determined whether the preset condition is met: determining whether the node density of the network topology is greater than or equal to a preset node density, and determining that the LALAC is based on Whether the average difference coefficient of the shortest path determined by the algorithm is less than or equal to the preset difference coefficient.

If it is determined that the node density of the network topology is greater than or equal to the preset node density, and determining that the average difference coefficient of the shortest path determined by the LALAC algorithm is less than or equal to the preset difference coefficient, proceeding to the steps subsequent to step 201, otherwise The shortest path determined based on the LALAC algorithm can be directly determined as the shortest path from the source node to the sink node in the network topology.

Specifically, the SDN controller may perform the following steps to determine whether the preset condition is met:

Step 301: The SDN controller determines a node density of the network topology.

The SDN controller can determine the node density of the network topology according to the following formula:

Av(D)=|E|×2/|V|·······(3)

Step 302: The SDN controller determines whether the node density is greater than or equal to the preset node density, and if so, then proceeds to step 303, otherwise proceeds to step 306;

In the embodiment of the present application, the value of the preset node density is not limited, and the specific value may be close to whether there are multiple critical values of the heavy edge between the nodes in the network topology, so that the preset node density may be used to determine Whether there are multiple heavy edges between nodes in the network topology. Specifically, if the node density of the network topology is less than the preset node density, it may be determined that there are no multiple heavy edges between the nodes in the network topology, so that the shortest path determined by the LALAC algorithm may be determined to be the optimal path.

If the node density of the network topology is greater than or equal to the preset node density, it may be determined that there may be multiple heavy edges between the nodes in the network topology (whether or not there are certain heavy edges that may be determined by the average difference coefficient of the combined paths), It can be determined that the shortest path determined based on the LALAC algorithm is not necessarily optimal, and the shortest path determined based on the LALAC algorithm may need to be optimized.

Step 303: The SDN controller determines an average difference coefficient of the shortest path determined by the LALAC algorithm.

The SDN controller may determine an average difference coefficient of the shortest path determined by the LALAC algorithm according to the following formula:

The CV _i is a difference coefficient of the weight of the heavy edge between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, and S _i is the i-th edge corresponding to the shortest path determined by the LALAC algorithm The standard deviation of the weight of the heavy edge between two adjacent nodes,

The average value of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, n is a positive integer greater than or equal to 1, and n is the shortest determined by the LALAC algorithm The number of edges included in the path.

It should be noted that each edge corresponds to two adjacent nodes, but there may be multiple edges between the adjacent two nodes. These edges may be called heavy edges, and each heavy edge corresponds to a first attribute and a second edge. Attributes. In the embodiment of the present application, the method for determining the weight of each heavy edge between adjacent nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, and the method for determining the weight of each edge in the network topology Similarly, specifically, for any one of the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, the product of the first attribute of the heavy side and the multiplier may be The sum of the second attributes of the heavy edges is determined as the weight of the heavy edge. Finally, the standard deviation of the weights of the heavy edges and the average weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge can be determined, thereby determining the difference of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge. coefficient.

For example, the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm include four heavy edges, and the corresponding first attribute and second attribute are: (2, 2), (1 , 6), (2, 1), (5, 1). The first parameter in parentheses is the first attribute, and the second parameter is the second attribute. If the multiplier is 3, the weight of each heavy side is: 8, 9, 7, 16 respectively. The standard deviation of the weights of the heavy edges and the average weights of the heavy edge weights between the adjacent two nodes corresponding to the i-th edge are: 3.53, 10, and finally the adjacent two nodes corresponding to the i-th edge can be determined. The coefficient of variation between the weights of the heavy edges is 0.353.

Step 304: The SDN controller determines whether the average difference coefficient of the shortest path determined by the LALAC algorithm is less than or equal to the preset difference coefficient, and if yes, proceeds to step 305, otherwise proceeds to step 306;

The average difference coefficient is the average difference between the weight of each heavy edge and the average of the weight of the heavy edge. The larger the mean difference coefficient, the greater the difference between the weight of each heavy edge and the average of the heavy edge weights; the smaller the average difference coefficient, the difference between the weight of each heavy edge and the average of the heavy edge weights. The difference is less.

When the average difference coefficient exceeds the preset difference coefficient, it can be considered that the degree of difference between each heavy side is very large, and each heavy side can be regarded as an independent side, so there are multiple heavy-edge network topologies between nodes, essentially It can be thought of as a network topology where there are no multiple heavy edges between nodes. The specific value of the preset difference coefficient may be close to a critical value of a network topology in which a plurality of heavy edges exist substantially as a network topology in which no multiple heavy edges exist.

In the embodiment of the present application, the preset difference coefficient is specifically determined, which is not limited by the embodiment of the present application. The preset difference coefficient can determine whether there are substantially multiple heavy edges between nodes in the network topology by combining the preset node density and the preset difference coefficient. Specifically, when the node density of the network topology is greater than or equal to the preset node density, if the average difference coefficient of the path is less than or equal to the preset difference coefficient, it may be determined that there are substantially multiple heavy edges between the nodes in the network topology. Therefore, it can be determined that the shortest path determined based on the LALAC algorithm is not necessarily optimal, and the shortest path determined based on the LALAC algorithm needs to be optimized.

When the node density of the network topology is greater than or equal to the preset node density, if the average difference coefficient of the path is greater than the preset difference coefficient, the network topology can be regarded as a network topology without multiple heavy edges, so the LARAC algorithm can be determined. The shortest path determined is the optimal path.

Step 305: According to the judgment result of step 304, it is determined that there are multiple heavy edges between the nodes in the current network topology, so the shortest path determined by the LALAC algorithm determined by the LALAC algorithm is not optimal, and needs to be determined based on the LALAC algorithm. The shortest path is optimized.

Step 306: Generate a routing table from the source node to the sink node according to the shortest path determined based on the LALAC algorithm.

Through the above process, when it is determined that the node density is less than or equal to a preset node density, and the average difference coefficient of the shortest path determined by the LALAC algorithm is greater than a preset difference coefficient, the nodes in the current network topology may be determined. There is no multiple heavy edges, so the shortest path determined based on the LALAC algorithm is the optimal path, so that the routing table from the source node to the sink node can be directly generated according to the shortest path determined by the LALAC algorithm, thereby improving the determination of the source node to The efficiency of the shortest path between sink nodes. Correspondingly, when it is determined that the preset condition is not met, it may be determined that there are multiple heavy edges between the two nodes in the current network topology, so the goodness of the shortest path determined by the LALAC algorithm is not high, and the actual application cannot be satisfied. Need to be optimized.

In step 202, for any one of the network topologies, the sum of the product of the first attribute of the edge and the multiplier and the second attribute of the edge may be determined as the weight of the edge, that is, in the network topology. The weight of the i-th edge satisfies the formula (1).

For example, as shown in FIG. 4, it is a schematic diagram of a network topology provided by an embodiment of the present application. In FIG. 4, the network topology includes nodes A, B, C, D, E, and F; the first attribute and the second attribute corresponding to the edge included in the network topology are: (2, 2), (1, 6), ( 3, 3), (1, 2), (2, 1), (1, 1), (5, 1), (4, 3). The first parameter in parentheses is the first attribute, and the second parameter is the second attribute. If the multiplier is 3, the weights of each side are: 8, 9, 12, 5, 7, 4, 16, and 15, respectively.

Then, the SDN controller may determine the K paths according to the shortest path algorithm, and the shortest path algorithm may be a Dijkstra algorithm, a Yen algorithm, an Eppstein algorithm, etc., which is not limited by the embodiment of the present application.

In step 203, the SDN controller may determine that the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the path with the smallest sum of the second attributes of the paths is determined as the first path, thereby The first path acts as the shortest path from the source node to the sink node in the network topology. That is, unlike the shortest path determined by the LARAC algorithm in the prior art as the shortest path from the source node to the sink node, in the embodiment of the present invention, the first path that satisfies the condition is used as the shortest path from the source node to the sink node, SDN. The controller may generate a routing table from the source node to the sink node according to the first path, where the routing table is used to route a message between the source node and the sink node.

The preset threshold is a user equipment that requests the SDN controller to determine the shortest path from the source node to the sink node, and is set according to its actual network requirement. For example, the first attribute is a delay, and the user equipment A needs a shortest path with a low delay. The user equipment A can set the preset threshold to a smaller value, for example, set to 1, and the user equipment A can The preset threshold is set to notify the SDN controller, and the SDN controller can thus determine the shortest path for the user equipment A that the delay is less than or equal to 1. Of course, the preset threshold can also be a default value, or the SDN controller can be set according to the actual situation of the network.

The value of the preset threshold may further restrict the K paths, and whether the sum of the first attributes of the path is less than or equal to a preset threshold, and the K paths are filtered, thereby further according to the sum of the second attributes of the path. Determine the first path. The larger the preset threshold is, the larger the sum of the first attributes of the paths selected from the K paths is, so that the sum of the first attributes of the determined first path is larger; otherwise, the smaller the preset threshold is. The smaller the sum of the first attributes of the paths filtered from the K paths, the smaller the sum of the first attributes of the determined first paths.

It should be noted that, in the embodiment of the present application, the preset threshold may be determined according to actual conditions, and details are not described herein again.

According to the method provided by the embodiment of the present application, after determining the multiplier from the source node to the sink node in the network topology, determining the weight of each edge in the network topology according to the multiplier, and according to the weight of each edge Determine the path with the least weight of the K paths. Finally, the first path is determined from the K paths. In the embodiment of the present application, the determined first path is based on the constraint that the first attribute of the path is less than or equal to the preset threshold, and the minimum constraint of the second attribute of the path is considered, Determining the first path is better than the path determined according to the constraint of the first attribute of the path and less than or equal to the preset threshold, that is, the determined sum of the second attributes of the first path is determined according to the LARAC algorithm. The sum of the second attributes of the shortest path is smaller, thereby achieving a better shortest path.

The foregoing process is described below by a specific embodiment.

As shown in FIG. 5, it is a schematic flowchart of a method for determining a shortest path according to an embodiment of the present application.

In the flow shown in Figure 5, the first attribute is a delay and the second attribute is a cost.

Step 501: The SDN controller acquires information such as a source node, a sink node, and a delay constraint. The delay constraint condition is that the sum of the delays of the path from the source node to the sink node in the network topology is less than or equal to a preset threshold.

Step 502: The SDN controller determines, according to the LALAC algorithm, a path that satisfies the delay constraint condition from all paths of the source node to the sink node in the network topology, that is, the shortest path determined based on the LALAC algorithm, and determines the shortest path determined by the LALAC algorithm. The corresponding multiplier.

For a specific iterative process of the LARAC algorithm, reference may be made to the foregoing description, and details are not described herein again.

Step 503: The SDN controller determines a node density of the network topology and an average difference coefficient of the second path.

Step 504: The SDN controller determines whether the node density is greater than or equal to a preset node density, and whether the average difference coefficient of the second path is less than or equal to a preset difference coefficient. If yes, go to step 505, otherwise, The shortest path determined based on the LALAC algorithm is determined to be the shortest path from the source node to the sink node in the network topology, and proceeds to step 508.

Step 505: The SDN controller determines a weight of each edge in the network topology according to the multiplier, and determines K paths according to the weight of each edge.

Step 506: The SDN controller determines a first path by using a sum of delays of the paths in the K paths that are less than or equal to a preset threshold, and a minimum of the cost of the path, and determines the first path as the The shortest path from the source node to the sink node in the network topology.

Step 507: The SDN controller generates a routing table from the source node to the sink node according to the first path.

Step 508: The SDN controller generates a routing table from the source node to the sink node according to the shortest path determined based on the LALAC algorithm.

Based on the same technical concept, the embodiment of the present application further provides a controller.

As shown in FIG. 6, the embodiment of the present application provides a schematic structural diagram of a controller. The controller can execute the flow shown in FIG. 2.

Referring to FIG. 6, the controller 600 includes a receiving module 601, a processing module 602, and a storage module 603.

The receiving module 601 is configured to receive source node information and sink node information; the source node information indicates a source node, and the sink node information indicates a sink node;

The storage module 603 is configured to use the storage network topology information, where the network topology information includes a network topology and a first attribute and a second attribute of each edge in the network topology, where the first attribute is the network An attribute of the link corresponding to the edge in the topology, where the second attribute is another attribute of the link corresponding to the edge in the network topology that is different from the first attribute;

The processing module 602 is configured to determine a multiplier corresponding to the shortest path determined by the Lagrangian relaxed LALAC algorithm from the source node to the sink node in the network topology; and determine, according to the multiplier, each side of the network topology Weighting, and determining K paths according to the weight of each edge, the K paths being K paths with small to large weights of paths in the path from the source node to the sink node in the network topology, K a positive integer greater than 1; the weight of the path is the sum of the weights of all edges in the path; the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the second attribute of the path And a minimum path determined as a first path, the first path being a shortest path from the source node to the sink node in the network topology, wherein a sum of the first attributes of the path is the first of all sides in the path The sum of the attributes, the sum of the second attributes of the path, and the sum of the second attributes of all edges in the path.

It should be understood that the division of each module above is only a division of logic functions, and the actual implementation may be integrated into one physical entity in whole or in part, or may be physically separated.

The receiving module 601, the processing module 602, and the storage module 603 can also perform other content. For details, refer to the descriptions in steps 201 to 203, and details are not described herein.

As shown in FIG. 7, the embodiment of the present application provides a schematic structural diagram of a controller. The controller can execute the flow shown in FIG. 2.

Referring to FIG. 7, the controller 700 includes a processor 701, a memory 702, and an input/output interface 703.

The processor 701 can be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or the like. Programming logic devices, discrete gates or transistor logic devices, discrete hardware components.

Memory 702 can include read only memory and random access memory and can provide instructions and data to processor 701. A portion of the memory 702 may also include a Non-Volatile Random Access Memory (NVRAM).

The various components of controller 700 are coupled together by a bus system 704, which may include, in addition to the data bus, a power bus, a control bus, a status signal bus, and the like. However, for clarity of description, various buses are labeled as bus system 704 in the figure.

An input/output interface 703, configured to receive source node information and sink node information; the source node information indicates a source node, and the sink node information indicates a sink node;

a memory 702, configured to: store the network topology information, where the network topology information includes a network topology and a first attribute and a second attribute of each edge in the network topology, where the first attribute is the network topology An attribute of the link corresponding to the middle side, where the second attribute is another attribute of the link corresponding to the edge in the network topology that is different from the first attribute;

The processor 701 is configured to determine a multiplier corresponding to the shortest path determined by the Lagrangian relaxed LALAC algorithm from the source node to the sink node in the network topology, and determine, according to the multiplier, each side of the network topology Weighting, and determining K paths according to the weight of each edge, the K paths being K paths with small to large weights of paths in the path from the source node to the sink node in the network topology, K a positive integer greater than 1; the weight of the path is the sum of the weights of all edges in the path; the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the second attribute of the path And a minimum path determined as a first path, the first path being a shortest path from the source node to the sink node in the network topology, wherein a sum of the first attributes of the path is the first of all sides in the path The sum of the attributes, the sum of the second attributes of the path, and the sum of the second attributes of all edges in the path.

In some embodiments, the processor 701 is further configured to:

Make sure the following preset conditions are met:

Av(D)=|E|×2/|V|

Where av(CV) is the average difference coefficient of the second path,

In some embodiments, the processor 701 is specifically configured to:

In some embodiments, the processor 701 is further configured to:

In some embodiments, the memory 702 is further configured to store a routing table;

The processor 701 is further configured to instruct the memory 702 to update the routing table, where the routing table is used to route a message between the source node and the sink node.

The embodiment of the present application further provides a computer readable storage medium for storing computer software instructions required to execute the foregoing processor, which includes a program for executing the above-mentioned processor.

Those skilled in the art will appreciate that embodiments of the present application can be provided as a method, system, or computer program product. Thus, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment in combination of software and hardware. Moreover, the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) including computer usable program code.

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

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

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

The above is only a part of the embodiments of the present application, and the scope of protection of the present application is not limited thereto.

Claims

A shortest path determining method, the method comprising:

Determining a multiplier corresponding to the shortest path determined by the Lagrangian relaxed LARAC algorithm from the source node to the sink node in the network topology;

Determining, according to the multiplier, a weight of each edge in the network topology, and determining K paths according to weights of each edge, where the K paths are from the source node to the sink in the network topology The path of the node in the path is weighted from small to large K paths, K is a positive integer greater than 1; the weight of the path is the sum of the weights of all edges in the path;

Determining, as the first path, a path in which the sum of the first attributes of the paths in the K paths is less than or equal to a preset threshold, and the sum of the second attributes of the paths is the first path a shortest path from the source node to the sink node, where the first attribute is an attribute of a link corresponding to an edge in the network topology, and the second attribute is an edge of the network topology Another attribute of the corresponding link that is different from the first attribute; the sum of the first attribute of the path is the sum of the first attributes of all edges in the path, and the sum of the second attributes of the path is all sides of the path The sum of the second attribute.
The method according to claim 1, wherein before determining the weight of each edge in the network topology according to the multiplier, the method further comprises:

Make sure the following preset conditions are met:

The node density of the network topology is greater than or equal to a preset node density, and the average difference coefficient of the shortest path determined based on the LALAC algorithm is less than or equal to a preset difference coefficient.
The method of claim 2 wherein said node density of said network topology satisfies the following formula:

Av(D)=|E|×2/|V|

Where av(D) is the node density of the network topology, E is the number of edges included in the network topology, and V is the number of nodes included in the network topology.
The method according to claim 2 or 3, wherein the average difference coefficient of the shortest path determined based on the LALAC algorithm satisfies the following formula:

Where av(CV) is the average difference coefficient of the shortest path determined by the LALAC algorithm,
The CV i is a difference coefficient of the weight of the heavy edge between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, where S i is the shortest path determined by the LARAC algorithm The standard deviation of the weight of the heavy edge between the adjacent two nodes corresponding to the i-edge,
An average value of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge of the shortest path determined by the LALAC algorithm, n is a positive integer greater than or equal to 1, and n is the based on the LARAC The shortest path determined by the algorithm includes the number of edges.
The method according to any one of claims 1 to 4, wherein determining the weight of each edge in the network topology according to the multiplier comprises:

For any one of the network topologies, the sum of the product of the first attribute of the edge and the multiplier and the second attribute of the edge is determined as the weight of the edge.
The method according to any one of claims 1 to 5, further comprising:

Generating a routing table from the source node to the sink node according to the first path.
The method according to any one of claims 1 to 6, wherein the first attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate;

The second attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate and is different from the first attribute.
A controller, comprising:

a receiving module, configured to receive source node information and sink node information; the source node information indicates a source node, and the sink node information indicates a sink node;

a storage module, configured to use the storage network topology information, where the network topology information includes a network topology and a first attribute and a second attribute of each edge in the network topology, where the first attribute is the network topology An attribute of the link corresponding to the middle side, where the second attribute is another attribute of the link corresponding to the edge in the network topology that is different from the first attribute;

a processing module, configured to determine a multiplier corresponding to a shortest path determined by a Lagrangian relaxed LALAC algorithm from a source node to a sink node in a network topology; determining, according to the multiplier, a weight of each edge in the network topology And determining K paths according to the weight of each edge, where the K paths are K paths whose weights in the path from the source node to the sink node in the network topology are small to large, K is a positive integer greater than 1; a weight of the path is a sum of weights of all edges in the path; a sum of first attributes of the paths in the K paths is less than or equal to a preset threshold, and a second attribute of the path And the smallest path is determined as a first path, the first path being a shortest path from the source node to the sink node in the network topology, wherein a sum of the first attributes of the path is a first attribute of all edges in the path And the sum of the second attribute of the path and the sum of the second attributes of all edges in the path.
The controller according to claim 8, wherein the processing module is further configured to:

Make sure the following preset conditions are met:

The node density of the network topology is greater than or equal to a preset node density, and the average difference coefficient of the shortest path determined based on the LALAC algorithm is less than or equal to a preset difference coefficient.
The controller of claim 9 wherein said node density of said network topology satisfies the following formula:

Av(D)=|E|×2/|V|

Where av(D) is the node density of the network topology, E is the number of edges included in the network topology, and V is the number of nodes included in the network topology.
The controller according to claim 9, wherein the average difference coefficient of said second path satisfies the following formula:

Where av(CV) is the average difference coefficient of the second path,
CV i is a difference coefficient of the weight of the heavy edge between the adjacent two nodes corresponding to the i-th edge of the second path, and S i is the weight between the adjacent two nodes corresponding to the i-th edge in the second path The standard deviation of the edge weights,
An average value of the weights of the heavy edges between the adjacent two nodes corresponding to the i-th edge of the second path, where n is a positive integer greater than or equal to 1, and n is the number of edges included in the second path.
The controller according to any one of claims 8 to 11, wherein the processing module is specifically configured to:

For any one of the network topologies, the sum of the product of the first attribute of the edge and the multiplier and the second attribute of the edge is determined as the weight of the edge.
The controller according to any one of claims 8 to 12, wherein the processing module is further configured to:

Generating a routing table from the source node to the sink node according to the first path.
The controller according to claim 13, wherein the storage module is further configured to store a routing table;

The processing module is further configured to instruct the storage module to update the routing table, where the routing table is used to route a message between the source node and the sink node.
The controller according to any one of claims 8 to 14, wherein the first attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate;

The second attribute is any one of a cost, a delay, a delay jitter, and a packet loss rate.
A computer readable storage medium comprising computer readable instructions which, when read and executed by a computer, cause a computer to perform the method of any of claims 1-7.
A computer program product comprising computer readable instructions which, when read and executed by a computer, cause a computer to perform the method of any of claims 1-7.