SDN-based marketing content intelligent scheduling method

Abstract

This invention relates to the field of digital marketing technology, specifically disclosing an intelligent scheduling method for marketing content based on SDN. The method includes device deployment, multi-dimensional status information collection, marketing content request parsing, intelligent scheduling decision-making, flow table distribution and path establishment, and marketing content transmission and dynamic adjustment. This solution decouples the control plane and data plane based on an SDN architecture. A centralized SDN controller provides a real-time global network status awareness, offering a foundational platform for dynamic scheduling of marketing content. It proposes an intelligent scheduling algorithm combining link load, server performance, and user preferences to achieve optimal matching and dynamic adjustment of marketing content traffic. This ensures that each marketing content request is allocated to the optimal service node and transmission path, effectively avoiding network congestion and server overload, and significantly improving user experience.

Description

Technical Field

[0001] This invention relates to the field of digital marketing technology, specifically to an intelligent scheduling method for marketing content based on SDN. Background Technology

[0002] With the rapid development of internet technology and the popularization of digital marketing, the transmission volume of various marketing content (including video ads, graphic and text pushes, interactive pages, etc.) on the network is growing exponentially. The effective distribution and scheduling of marketing content is directly related to user experience, marketing effectiveness, and the efficiency of network resource utilization. However, in traditional network architectures, the control plane and data plane are tightly coupled, resulting in rigid network management and difficulty in flexibly adjusting according to real-time network status and content distribution needs. On the other hand, existing marketing content scheduling methods are mostly focused on application-layer load balancing strategies, such as round-robin and least connections. Although these methods can distribute request pressure to a certain extent, their decision-making basis is relatively singular and fails to comprehensively consider multi-dimensional factors such as real-time bandwidth utilization of network links, transmission latency, and user preferences for different types of content. Summary of the Invention

[0003] To address the above issues and overcome the shortcomings of existing technologies, this invention provides an SDN-based intelligent scheduling method for marketing content. Addressing the problem in traditional network architectures where the control plane and data plane are tightly coupled, leading to rigid network management and difficulty in flexibly adjusting based on real-time network conditions and content distribution needs, this solution decouples the control plane and data plane based on an SDN architecture. A centralized SDN controller monitors the global network status in real time, providing a foundational platform for dynamic scheduling of marketing content. Furthermore, addressing the issue that existing marketing content scheduling methods often focus on application-layer load balancing strategies, failing to comprehensively consider multi-dimensional factors such as real-time bandwidth utilization, transmission latency, and user preferences for different content types, this solution proposes an intelligent scheduling algorithm that combines link load, server performance, and user preferences. This achieves optimal matching and dynamic adjustment of marketing content traffic, ensuring that each marketing content request is allocated to the optimal service node and transmission path, effectively avoiding network congestion and server overload, and significantly improving user experience.

[0004] The technical solution adopted in this invention is as follows: The intelligent scheduling method for marketing content based on SDN provided by this invention includes the following steps:

[0005] Step S1: Equipment deployment, deploying SDN controllers, OpenFlow switch clusters, marketing content server clusters, and user terminals;

[0006] Step S2: Multi-dimensional status information collection. The SDN controller periodically collects network link status information and the load information of the marketing content server cluster, and calculates the real-time bandwidth utilization, transmission latency, and comprehensive load index of the server.

[0007] Step S3: Marketing content request parsing. Parse the marketing content request initiated by the user, extract the marketing content identifier, and construct the user preference vector;

[0008] Step S4: Intelligent scheduling decision-making. Construct an intelligent scheduling model. Based on the collected multi-dimensional status information and request parsing results, calculate the comprehensive scheduling score of each candidate server and select the optimal target server and optimal transmission path.

[0009] Step S5: Flow table distribution and path establishment. The SDN controller generates multi-level flow table entries based on the optimal target server and optimal transmission path, distributes them to the relevant OpenFlow switches on the optimal transmission path, and establishes a data forwarding channel.

[0010] Step S6: Marketing content transmission and dynamic adjustment. Transmit marketing content according to the optimal transmission path, and dynamically monitor network status changes during transmission, and set up a rescheduling mechanism.

[0011] Furthermore, in step S2, the acquisition of multidimensional state information specifically includes the following steps:

[0012] Step S21: Network link status information collection. The SDN controller discovers the network topology through the LLDP protocol (Link Layer Discovery Protocol), periodically acquires port statistics information of the OpenFlow switches, including the number of bytes sent, the number of bytes received, the number of data packets, and the number of lost packets. Based on the difference in the number of bytes between two consecutive sampling periods, the real-time traffic rate of the link is calculated, and the real-time bandwidth utilization of the link is calculated. The round-trip time of the link is measured by sending probe packets, and the transmission delay is calculated. The formula used is as follows: ; ;

[0013] In the formula, Represents the network link index. This indicates the real-time bandwidth utilization of the link. Indicates the current traffic rate of the link. Indicates the maximum bandwidth capacity of the link. Indicates the transmission delay of the link. Indicates the link round-trip time;

[0014] Step S22: Server load information collection. Real-time monitoring of CPU utilization, memory utilization, current active connections, and request response time of each server in the marketing content server cluster. This data is then reported to the SDN controller via the southbound interface to construct a comprehensive load index to assess the server's load status. The formula used is as follows: ; ;

[0015] In the formula, Indicates the server index. This indicates the overall load index of the server. This indicates the server's CPU utilization. This indicates the server's memory usage. This indicates the current number of active connections to the server. This indicates the maximum number of connections the server can handle. , , These represent the corresponding weighting coefficients.

[0016] Furthermore, in step S3, the marketing content request parsing specifically includes the following steps:

[0017] Step S31: When a user initiates a marketing content request, the request data packet arrives at the edge OpenFlow switch, which encapsulates the request data packet and forwards it to the SDN controller;

[0018] Step S32: The SDN controller performs deep parsing on the request data packet to extract the user identifier, the requested marketing content identifier, and the user level information;

[0019] Step S33: Obtain user's historical behavior data, determine the user's preference weight for different types of marketing content, calculate the percentage of historical consumption time for each type of marketing content, and construct a user preference vector using the following formula: ; ;

[0020] In the formula, Indicates user, Represents a user preference vector. This indicates the total number of marketing content types. Indicates an index of marketing content. Indicates the user's opinion on the first The preference weight of marketing-related content.

[0021] Furthermore, in step S4, the intelligent scheduling decision specifically includes the following steps:

[0022] Step S41: Candidate server filtering. Based on the marketing content identifier obtained from request parsing, find candidate servers that can provide the marketing content.

[0023] Step S42: Network path quality assessment. For each candidate server, the K-shortest path algorithm is used. Combining the real-time bandwidth utilization and transmission latency of each link on the path, the comprehensive quality score of all feasible paths from the user's access to the OpenFlow switch to the candidate server is calculated. The formula used is as follows: ;

[0024] In the formula, Indicates candidate servers, This represents the quality score of the optimal path to the candidate server. Represents the set of all feasible paths. This represents any path in the set of feasible paths. Indicates the links in the path. This indicates the real-time bandwidth utilization of the link. Indicates the transmission delay of the link. and These represent the weighting coefficients for real-time bandwidth utilization and transmission delay, respectively.

[0025] Step S43: Server processing capacity assessment. Combining the candidate server's overall load index and the network distance between the candidate server and the user, evaluate the candidate server's processing capacity score for the current request. The formula used is as follows: ;

[0026] In the formula, This indicates the candidate server's rating of its ability to process the current request. This represents the overall load index of the candidate servers. Indicates the network distance between the user and the candidate server. This represents the maximum distance among all candidate servers. and These represent the weighting coefficients for the overall load index and network distance, respectively.

[0027] Step S44: User preference matching evaluation. Based on the degree of matching between the marketing content type identified by the marketing content identifier and the user preference vector, calculate the preference matching score of the candidate server for the current user request. The formula used is as follows: ;

[0028] In the formula, This represents the score indicating how well the candidate servers match the current user's request in terms of preference. As an indicator function, the marketing content types stored on the candidate server include the first... The value is 1 if the condition is met, and 0 otherwise.

[0029] Step S45: Calculate the comprehensive scheduling score by integrating network path quality, server processing capacity, and user preference matching degree. The formula used is as follows: ; ;

[0030] In the formula, This represents the overall scheduling score of the candidate servers. , , These represent the corresponding weight coefficients;

[0031] Step S46: Optimal target server selection. Select the candidate server with the highest comprehensive scheduling score as the optimal target server and record the optimal transmission path corresponding to the server.

[0032] Further, in step S5, the flow table distribution and path establishment specifically involve: the SDN controller generating multi-level flow table entries based on the optimal target server and optimal transmission path, including a matching field and an instruction set: the matching field is set to the user IP address, the optimal target server IP address, and the input port, and the instruction set is set to forward to the output port corresponding to the next hop; the SDN controller distributes the flow table entries to all OpenFlow switches on the optimal transmission path via the OpenFlow protocol, the user request data packet is forwarded to the optimal target server along the optimal transmission path, and the marketing content data responded by the optimal target server is returned to the user along the reverse path.

[0033] The beneficial effects achieved by the present invention using the above solution are as follows:

[0034] (1) In the traditional network architecture, the control plane and the data plane are tightly coupled, which leads to rigid network management and makes it difficult to make flexible adjustments according to the real-time network status and content distribution needs. This solution is based on the SDN architecture to decouple the control plane and the data plane. Through a centralized SDN controller, the global network status is perceived in real time, providing a basic platform for the dynamic scheduling of marketing content.

[0035] (2) In view of the fact that existing marketing content scheduling methods are mostly focused on application layer load balancing strategies, and fail to comprehensively consider multi-dimensional factors such as real-time bandwidth utilization of network links, transmission delay and user preferences for different types of content, this solution proposes an intelligent scheduling algorithm that combines link load, server performance and user preferences to achieve optimal matching and dynamic adjustment of marketing content traffic, ensuring that each marketing content request can be allocated to the optimal service node and transmission path, effectively avoiding network congestion and server overload, and significantly improving user experience. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the SDN-based intelligent scheduling method for marketing content proposed in this invention.

[0037] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0038] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0039] Example 1, see Figure 1 The present invention provides an intelligent marketing content scheduling method based on SDN, which includes the following steps:

[0040] Step S1: Equipment deployment, deploying SDN controllers, OpenFlow switch clusters, marketing content server clusters, and user terminals;

[0041] Step S2: Multi-dimensional status information collection. The SDN controller periodically collects network link status information and the load information of the marketing content server cluster, and calculates the real-time bandwidth utilization, transmission latency, and comprehensive load index of the server.

[0042] Step S3: Marketing content request parsing. Parse the marketing content request initiated by the user, extract the marketing content identifier, and construct the user preference vector;

[0043] Step S4: Intelligent scheduling decision-making. Construct an intelligent scheduling model. Based on the collected multi-dimensional status information and request parsing results, calculate the comprehensive scheduling score of each candidate server and select the optimal target server and optimal transmission path.

[0044] Step S5: Flow table distribution and path establishment. The SDN controller generates multi-level flow table entries based on the optimal target server and optimal transmission path, distributes them to the relevant OpenFlow switches on the optimal transmission path, and establishes a data forwarding channel.

[0045] Step S6: Marketing content transmission and dynamic adjustment. Transmit marketing content according to the optimal transmission path, and dynamically monitor network status changes during transmission, and set up a rescheduling mechanism.

[0046] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the SDN controller is connected to the OpenFlow switch cluster and the control plane respectively, and is used to collect network status information and issue flow tables; the OpenFlow switch cluster is used to forward user requests and marketing content data according to the flow tables; the marketing content server cluster is deployed at the network edge or center and is used to store and distribute marketing content.

[0047] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, the multi-dimensional state information acquisition specifically includes the following steps:

[0048] Step S21: Network link status information collection. The SDN controller discovers the network topology through the LLDP protocol, periodically acquires port statistics information of the OpenFlow switches, including the number of bytes sent, the number of bytes received, the number of data packets, and the number of lost packets. Based on the difference in the number of bytes between two consecutive sampling periods, the real-time traffic rate of the link is calculated, and the real-time bandwidth utilization of the link is calculated. The round-trip time of the link is measured by sending probe packets, and the transmission delay is calculated. The formula used is as follows: ; ;

[0049] In the formula, Represents the network link index. This indicates the real-time bandwidth utilization of the link. Indicates the current traffic rate of the link. Indicates the maximum bandwidth capacity of the link. Indicates the transmission delay of the link. Indicates the link round-trip time;

[0050] Step S22: Server load information collection. Real-time monitoring of CPU utilization, memory utilization, current active connections, and request response time of each server in the marketing content server cluster. This data is then reported to the SDN controller via the southbound interface to construct a comprehensive load index to assess the server's load status. The formula used is as follows: ; ;

[0051] In the formula, Indicates the server index. This indicates the overall load index of the server. This indicates the server's CPU utilization. This indicates the server's memory usage. This indicates the current number of active connections to the server. This indicates the maximum number of connections the server can handle. , , These represent the corresponding weighting coefficients.

[0052] Example 4, see Figure 1 This embodiment is based on the above embodiment. In step S3, the marketing content request parsing specifically includes the following steps:

[0053] Step S31: When a user initiates a marketing content request, the request data packet arrives at the edge OpenFlow switch, which encapsulates the request data packet using a Packet-in message and forwards it to the SDN controller;

[0054] Step S32: The SDN controller performs deep parsing on the request data packet to extract the user identifier, the requested marketing content identifier, and the user level information;

[0055] Step S33: Obtain user historical behavior data, including click records, viewing time, and interaction behavior; determine the user's preference weight for different types of marketing content; calculate the percentage of historical consumption time for each type of marketing content; and construct a user preference vector using the following formula: ; ;

[0056] In the formula, Indicates user, Represents a user preference vector. This indicates the total number of marketing content types. Indicates an index of marketing content. Indicates the user's opinion on the first The preference weight of marketing-related content.

[0057] By performing the aforementioned operations, this solution addresses the problem in traditional network architectures where the control plane and data plane are tightly coupled, leading to rigid network management and difficulty in flexibly adjusting according to real-time network status and content distribution needs. Based on an SDN architecture, this solution decouples the control plane and data plane, and provides a basic platform for dynamic scheduling of marketing content by using a centralized SDN controller to perceive the global network status in real time.

[0058] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the intelligent scheduling decision specifically includes the following steps:

[0059] Step S41: Candidate server filtering. Based on the marketing content identifier obtained from request parsing, find candidate servers that can provide the marketing content.

[0060] Step S42: Network path quality assessment. For each candidate server, the K-shortest path algorithm is used. Combining the real-time bandwidth utilization and transmission latency of each link on the path, the comprehensive quality score of all feasible paths from the user's access to the OpenFlow switch to the candidate server is calculated. The formula used is as follows: ;

[0061] In the formula, Indicates candidate servers, This represents the quality score of the optimal path to the candidate server. Represents the set of all feasible paths. This represents any path in the set of feasible paths. Indicates the links in the path. This indicates the real-time bandwidth utilization of the link. Indicates the transmission delay of the link. and These represent the weighting coefficients for real-time bandwidth utilization and transmission delay, respectively.

[0062] Step S43: Server processing capacity assessment. Combining the candidate server's overall load index and the network distance between the candidate server and the user, evaluate the candidate server's processing capacity score for the current request. The formula used is as follows: ;

[0063] In the formula, This indicates the candidate server's rating of its ability to process the current request. This represents the overall load index of the candidate servers. Indicates the network distance between the user and the candidate server. This represents the maximum distance among all candidate servers. and These represent the weighting coefficients for the overall load index and network distance, respectively.

[0064] Step S44: User preference matching evaluation. Based on the degree of matching between the marketing content type identified by the marketing content identifier and the user preference vector, calculate the preference matching score of the candidate server for the current user request. The formula used is as follows: ;

[0065] In the formula, This represents the score indicating how well the candidate servers match the current user's request in terms of preference. As an indicator function, the marketing content types stored on the candidate server include the first... The value is 1 if the condition is met, and 0 otherwise.

[0066] Step S45: Calculate the comprehensive scheduling score by integrating network path quality, server processing capacity, and user preference matching degree. The formula used is as follows: ; ;

[0067] In the formula, This represents the overall scheduling score of the candidate servers. , , These represent the corresponding weight coefficients;

[0068] Step S46: Optimal target server selection. Select the candidate server with the highest comprehensive scheduling score as the optimal target server and record the optimal transmission path corresponding to the server.

[0069] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S5, the flow table distribution and path establishment are specifically as follows: The SDN controller generates multi-level flow table entries based on the optimal target server and the optimal transmission path, including a matching field and an instruction set: the matching field is set to the user IP address, the optimal target server IP address and the input port, and the instruction set is set to forward to the output port corresponding to the next hop; the SDN controller distributes the flow table entries to all OpenFlow switches on the optimal transmission path through the OpenFlow protocol, the user request data packet is forwarded to the optimal target server along the optimal transmission path, and the marketing content data responded by the optimal target server is returned to the user along the reverse path.

[0070] Example 7, see Figure 1 This embodiment is based on the above embodiment. In step S6, the transmission and dynamic adjustment of marketing content specifically includes the following steps:

[0071] Step S61: The SDN controller continuously monitors the network link status and server load changes during marketing content transmission, with a monitoring period of 5 seconds;

[0072] Step S62: Set the bandwidth threshold to 80% and the load threshold to 0.8. When the real-time bandwidth utilization of a link on the optimal transmission path exceeds the bandwidth threshold, or the comprehensive load index of the optimal target server exceeds the load threshold, the rescheduling mechanism is triggered.

[0073] Step S63: The rescheduling mechanism re-executes step S4, and calculates the new optimal target server and optimal transmission path for the current user request based on the current established network connection status;

[0074] Step S64: The SDN controller adopts an elegant rerouting method. First, it creates a flow table entry for the new path, sets its priority to be higher than that of the original path, monitors the real-time traffic rate on the new path to confirm whether the request packets of the new path are forwarded normally, and deletes the flow table entry of the original path after normal forwarding begins, thus achieving seamless switching.

[0075] By performing the aforementioned operations, this solution addresses the problem that existing marketing content scheduling methods, which mostly focus on application-layer load balancing strategies, fail to comprehensively consider multi-dimensional factors such as real-time bandwidth utilization of network links, transmission latency, and user preferences for different types of content. It proposes an intelligent scheduling algorithm that combines link load, server performance, and user preferences to achieve optimal matching and dynamic adjustment of marketing content traffic. This ensures that each marketing content request is allocated to the optimal service node and transmission path, effectively avoiding network congestion and server overload, and significantly improving user experience.

[0076] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0077] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0078] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. An intelligent marketing content scheduling method based on SDN, characterized by: The method includes the following steps: Step S1: Equipment deployment, deploying SDN controllers, OpenFlow switch clusters, marketing content server clusters, and user terminals; Step S2: Multi-dimensional status information collection. The SDN controller periodically collects network link status information and the load information of the marketing content server cluster, and calculates the real-time bandwidth utilization, transmission latency, and comprehensive load index of the server. Step S3: Marketing content request parsing. Parse the marketing content request initiated by the user, extract the marketing content identifier, and construct the user preference vector; Step S4: Intelligent scheduling decision-making. Construct an intelligent scheduling model. Based on the collected multi-dimensional status information and request parsing results, calculate the comprehensive scheduling score of each candidate server and select the optimal target server and optimal transmission path. Step S5: Flow table distribution and path establishment. The SDN controller generates multi-level flow table entries based on the optimal target server and optimal transmission path, distributes them to the relevant OpenFlow switches on the optimal transmission path, and establishes a data forwarding channel. Step S6: Marketing content transmission and dynamic adjustment. Transmit marketing content according to the optimal transmission path, and dynamically monitor network status changes during transmission, and set up a rescheduling mechanism.

2. The SDN-based intelligent scheduling method for marketing content according to claim 1, characterized in that: In step S2, the acquisition of multidimensional state information specifically includes the following steps: Step S21: Network link status information collection. The SDN controller discovers the network topology through the LLDP protocol, periodically obtains the port statistics information of the OpenFlow switch, calculates the real-time traffic rate of the link, calculates the real-time bandwidth utilization of the link, measures the round-trip time of the link by sending probe packets, and calculates the transmission delay. Step S22: Collect server load information, monitor the CPU utilization, memory utilization, current active connection count and request response time of each server in the marketing content server cluster in real time, and report to the SDN controller through the southbound interface to build a comprehensive load index to evaluate the server load status.

3. The SDN-based intelligent scheduling method for marketing content according to claim 1, characterized in that: In step S3, the marketing content request parsing specifically includes the following steps: Step S31: When a user initiates a marketing content request, the request data packet arrives at the edge OpenFlow switch, which encapsulates the request data packet and forwards it to the SDN controller; Step S32: The SDN controller performs deep parsing on the request data packet to extract the user identifier, the requested marketing content identifier, and the user level information; Step S33: Obtain user's historical behavior data, determine the user's preference weight for different types of marketing content, calculate the percentage of historical consumption time for each type of marketing content, and construct a user preference vector.

4. The SDN-based intelligent scheduling method for marketing content according to claim 1, characterized in that: In step S4, the intelligent scheduling decision specifically includes the following steps: Step S41: Candidate server filtering. Based on the marketing content identifier obtained from request parsing, find candidate servers that can provide the marketing content. Step S42: Network path quality assessment. For each candidate server, the K-shortest path algorithm is used. Combining the real-time bandwidth utilization and transmission delay of each link on the path, the comprehensive quality score of all feasible paths from the user's access to the OpenFlow switch to the candidate server is calculated. Step S43: Server processing capacity assessment. Based on the comprehensive load index of the candidate server and the network distance between the candidate server and the user, evaluate the candidate server's processing capacity score for the current request. Step S44: User preference matching evaluation. Based on the degree of matching between the marketing content type identified by the marketing content identifier and the user preference vector, calculate the preference matching score of the candidate server for the current user request. Step S45: Calculate the overall scheduling score by integrating the overall quality score, processing capacity score, and preference matching score to calculate the overall scheduling score for each candidate server; Step S46: Optimal target server selection. Select the candidate server with the highest comprehensive scheduling score as the optimal target server and record the optimal transmission path corresponding to the server.

5. The SDN-based intelligent scheduling method for marketing content according to claim 1, characterized in that: In step S5, the flow table distribution and path establishment specifically involve: the SDN controller generating multi-level flow table entries based on the optimal target server and optimal transmission path, including a matching field and an instruction set: the matching field is set to the user's IP address, the optimal target server's IP address, and the input port; the instruction set is set to forward to the output port corresponding to the next hop; the SDN controller distributes the flow table entries to all OpenFlow switches on the optimal transmission path via the OpenFlow protocol; the user request data packet is forwarded to the optimal target server along the optimal transmission path; and the marketing content data responded by the optimal target server is returned to the user along the reverse path.

Images