A PCDN Optimization Method and System Based on Intelligent Scheduling and Dynamic Governance

Abstract

This invention relates to a PCDN optimization method and system based on intelligent scheduling and dynamic governance, belonging to the field of distributed computing and edge network scheduling technology. The method includes: collecting hardware resource status data of nodes; generating node resource scheduling signals by combining user geographical location and content demand priority with a content demand optimization scheduling algorithm; constructing a multi-level cache priority system based on hot content in different time periods and regions, combined with content timeliness and user access frequency, and allocating optimal nodes and data transmission paths; analyzing the dynamic relationship between node load and resource utilization through path cross-modeling scheduling method, and allocating independent scheduling windows; constructing a point-to-point content distribution network authorization feature library to classify and identify node traffic, and implementing a tiered suppression strategy for unauthorized traffic abuse based on real-time network load, limiting traffic transmission during low load and blocking it during high load, based on the network node load status.

Description

Technical Field

[0001] This invention belongs to the field of distributed computing and edge network scheduling technology, specifically relating to a PCDN optimization method and system based on intelligent scheduling and dynamic governance. Background Technology

[0002] Peer-to-Peer Content Delivery Networks (PCDNs), as a significant evolution in content delivery technology, utilize a large number of distributed nodes to participate in content transmission and caching, effectively reducing the load on central servers and improving content transmission efficiency. With the rapid growth of internet content, especially the widespread adoption of high-definition video, live streaming, and long-form video-on-demand services, traditional centralized CDNs are struggling to meet the demands of high concurrency, high bandwidth, and high real-time performance. PCDNs are gradually becoming an important supplement to content delivery systems. However, existing PCDN networks still have many shortcomings in resource scheduling, node governance, caching strategies, and traffic management.

[0003] First, PCDN has a large number of nodes, a wide geographical distribution, and a complex operating environment. The computing power, storage resources, and bandwidth of each node vary significantly. Existing scheduling algorithms mostly rely on single-dimensional indicators for node selection, making it difficult to accurately assess the resource status of nodes. Second, user access behavior is highly dynamic and varies regionally. Traditional scheduling models cannot respond quickly to changes in real-time content demand, user location, and content popularity, leading to low node resource utilization, scheduling imbalances, and even network congestion.

[0004] Secondly, the content in the PCDN network is mostly high-frequency and time-sensitive media resources. However, existing caching systems typically use static caching levels or replacement mechanisms based on fixed thresholds, which cannot adaptively adjust according to node load, geographical access distribution, and content lifecycle. This leads to decreased cache hit rates, redundant content occupying storage space, and unreasonable content distribution density across regional nodes. Furthermore, the access popularity of hot content varies significantly across different regions. If tiered caching deployment is not implemented based on regional characteristics, overall content distribution efficiency will be further reduced.

[0005] Given their open and decentralized nature, distribution networks are susceptible to abnormal access, unauthorized node requests, and abuse. Existing traffic identification mechanisms largely rely on static features or single behavioral rules, lacking the comprehensive ability to judge multi-dimensional characteristics such as content timeliness and the degree of access deviation. This makes it difficult to identify unauthorized abuse traffic in a timely manner, causing abnormal traffic to consume significant node resources under high network load, and even degrading distribution network performance. Furthermore, existing suppression strategies generally adopt a "one-size-fits-all" approach to rate limiting, lacking the ability to flexibly adjust according to load conditions, which can easily suppress normal traffic.

[0006] Finally, due to the complexity of PCDN data transmission paths and the overlapping relationships between multiple node paths, traditional path scheduling methods cannot effectively model path conflicts, resource competition, and dynamic changes in node load. They also lack a refined timing management mechanism, resulting in a lack of independence and coordination in data transmission scheduling between nodes, which affects overall transmission efficiency.

[0007] Therefore, there is an urgent need for a PCDN optimization method that can achieve real-time assessment of node resource status, intelligent scheduling of content demand, adaptive management of multi-level caching, and dynamic governance of unauthorized traffic abuse, so as to improve the system's resource utilization, content transmission efficiency, and network security. Summary of the Invention

[0008] To address the aforementioned problems in the existing technology, this invention provides a PCDN optimization method based on intelligent scheduling and dynamic governance.

[0009] The objective of this invention can be achieved through the following technical solutions:

[0010] S1: Based on the distributed network probe, collect hardware resource status data of each node in the peer-to-peer content delivery network. The hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm.

[0011] S2: The point-to-point content distribution network issues instructions in conjunction with the node resource scheduling signal, and constructs a multi-level cache priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path;

[0012] S3: The path cross modeling scheduling method is used to vectorize the data transmission path of the nodes, analyze the dynamic relationship between the load and resource utilization of each node, divide the resource access time sequence into discrete time slices, and allocate independent scheduling windows for different network nodes.

[0013] S4: Construct a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized abuse traffic, and implement a tiered suppression strategy for unauthorized abuse traffic based on the real-time network load. Depending on the network node load, traffic transmission is limited when the load is low and blocked when the load is high.

[0014] Specifically, the hardware resource status data includes bandwidth load and network latency, and the computing resources and network transmission capabilities of each node are evaluated in real time based on periodically acquired local node indicators.

[0015] Specifically, the content demand priority is predicted by analyzing historical access data through a content demand optimization scheduling algorithm, and the node allocation is dynamically adjusted in combination with geographical location information.

[0016] Specifically, the method for generating the node resource scheduling signal is as follows:

[0017] Based on the content request characteristics of the user side, content type requirement parameters are extracted, and a content requirement optimization scheduling algorithm is used to mobilize weighted scoring nodes according to the content request feature vector, and node resource scores are calculated according to priority queues.

[0018] Based on the node resource scoring-driven node time slice rolling update mechanism, the node resource fluctuation trend is corrected, a node resource scheduling signal is generated, and the validity period of the node resource scheduling signal is adjusted according to the network load and node resource changes.

[0019] Specifically, the method for constructing the multi-level cache priority system is as follows:

[0020] The cache layer is initialized and configured based on node resource capabilities and network topology. The layered structure of hot spot fast cache layer, regional mid-term cache layer and long tail basic cache layer is defined. The content attribute vector is extracted based on the access popularity parameter and regional access distribution of the acquired content.

[0021] The content attribute vectors are classified according to access popularity thresholds and resource consumption costs. Hot content is quickly cached at higher levels using a rule tree weighted scoring algorithm. The regional delivery level of the content is determined based on regional distribution parameters. The residence time of the node resource scheduling signal at different levels is controlled. A cache distribution mapping table is generated to define the distribution density and cache ratio of content at each level in different geographical regions, and a multi-level cache priority system is constructed.

[0022] Specifically, the content timeliness is used to perform deviation analysis on node access traffic to identify unauthorized abuse traffic, including video update frequency and file popularity decay period. By comparing the actual access time of the node request content with the expected access time defined by the content timeliness characteristics, it is determined whether the request meets the authorized access requirements, and the unauthorized abuse traffic is marked and graded according to the degree of deviation.

[0023] Specifically, the multi-level caching priority system prioritizes caching high-priority content to edge nodes based on the bandwidth consumption of network nodes transmitting content back to the peer-to-peer distribution network, caches low-priority content on demand, and reduces invalid cache usage through adaptive eviction.

[0024] Specifically, the path cross-modeling and scheduling method constructs a path matrix of data transmission paths for each node in the peer-to-peer content delivery network. By analyzing the similarity and cross-relationship of path feature vectors in the path matrix, path conflicts are identified, and data transmission paths are dynamically adjusted based on node load and resource utilization.

[0025] Specifically, the independent scheduling window dynamically adjusts its length and start time based on the node's historical load, resource availability, and access request characteristics, and combines this with a discrete time slice partitioning strategy to enable each node to independently execute node data transmission and cache scheduling tasks in different time periods.

[0026] Specifically, the unauthorized abuse traffic refers to traffic generated by unauthorized nodes or abnormal access behavior. It is characterized by abnormal node access frequency, requested content deviating from timeliness indicators, and network resources occupied exceeding the node's carrying capacity. When the peer-to-peer content distribution network authorization feature library identifies features that deviate from the authorized access mode, it is determined to be unauthorized abuse traffic and a tiered suppression strategy is triggered.

[0027] Specifically, the tiered suppression strategy progressively increases the restriction intensity based on multi-level suppression thresholds, including sequentially implementing request frequency reduction, bandwidth allocation restrictions, task queuing delay injection, and forced interruption rescheduling for target nodes or target traffic, thereby gradually suppressing and scheduling abnormal behavior.

[0028] Specifically, a PCDN optimization system based on intelligent scheduling and dynamic governance includes:

[0029] Intelligent node scheduling module: Based on distributed network probes to collect hardware resource status data of each node in the peer-to-peer content delivery network, the hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm.

[0030] Dynamic content caching optimization module: The peer-to-peer content distribution network issues instructions in conjunction with the node resource scheduling signals, and constructs a multi-level caching priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path;

[0031] Independent scheduling and allocation module: The node data transmission path is vectorized using the path cross modeling scheduling method, the dynamic relationship between the load and resource utilization of each node is analyzed, and the resource access time sequence is divided into discrete time slices to allocate independent scheduling windows to different network nodes;

[0032] Unauthorized traffic abuse management module: Constructs a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized traffic abuse, and implements a tiered suppression strategy for unauthorized traffic abuse based on real-time network load, limiting traffic transmission during low load and blocking it during high load according to the network node load status.

[0033] The beneficial effects of this invention are as follows:

[0034] The PCDN optimization method proposed in this invention, based on intelligent scheduling and dynamic governance, can significantly improve multi-dimensional resource awareness, content scheduling strategies, and abnormal traffic management. First, by collecting node hardware resource status in real time through distributed network probes, and combining this with a content demand optimization scheduling algorithm to jointly analyze user geographical location, content priority, and node resources, precise node resource scheduling signals can be dynamically generated. This enables targeted and differentiated scheduling during content distribution, significantly improving node resource utilization and load balancing capabilities.

[0035] The multi-level caching priority system can implement adaptive caching strategies across different levels based on content popularity, content timeliness, geographical access distribution, and node bandwidth feedback. Through rule trees and a weighted scoring mechanism, popular content can be quickly stored in high-priority cache layers, while long-tail content is allocated on demand, significantly improving cache hit rates and reducing content transmission latency and cross-domain access pressure. Furthermore, the access deviation analysis mechanism based on content timeliness further enhances the accuracy of identifying abnormal node requests, making the distinction between authorized access and unauthorized traffic abuse more reliable.

[0036] A path intersection modeling and scheduling method is employed to perform vectorized analysis of node transmission paths. By identifying path conflicts and resource contention, dynamic optimization of data transmission paths is achieved. Combined with an independent scheduling window time-series management mechanism, each node can execute tasks within separate time slices, effectively avoiding transmission efficiency degradation caused by resource contention.

[0037] By introducing a tiered suppression strategy, measures such as frequency limiting, bandwidth limiting, latency injection, or forced interruption are implemented in stages based on the degree of network load and traffic deviation. This allows for gradual control of unauthorized traffic abuse, from mild to strict. While ensuring the stability of normal services, the impact of abnormal traffic on network performance is reduced.

[0038] In summary, this invention can significantly improve the resource scheduling accuracy, cache allocation efficiency, path management capabilities, and abnormal traffic management capabilities of PCDN networks, thereby achieving a more efficient, controllable, and stable distributed content distribution system. Attached Figure Description

[0039] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0040] Figure 1 This is a schematic diagram of the PCDN optimization method and system based on intelligent scheduling and dynamic governance according to the present invention.

[0041] Figure 2 This is a diagram illustrating the overall technical framework of a PCDN optimization method and system based on intelligent scheduling and dynamic governance according to the present invention. Detailed Implementation

[0042] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0043] Please see Figure 1 A PCDN optimization method based on intelligent scheduling and dynamic governance:

[0044] S1: Based on the distributed network probe, collect hardware resource status data of each node in the peer-to-peer content delivery network. The hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm.

[0045] S2: The point-to-point content distribution network issues instructions in conjunction with the node resource scheduling signal, and constructs a multi-level cache priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path;

[0046] S3: The path cross modeling scheduling method is used to vectorize the data transmission path of the nodes, analyze the dynamic relationship between the load and resource utilization of each node, divide the resource access time sequence into discrete time slices, and allocate independent scheduling windows for different network nodes.

[0047] S4: Construct a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized abuse traffic, and implement a tiered suppression strategy for unauthorized abuse traffic based on the real-time network load. Depending on the network node load, traffic transmission is limited when the load is low and blocked when the load is high.

[0048] In this embodiment, the hardware resource status data includes bandwidth load and network latency, and the computing resources and network transmission capabilities of each node are evaluated in real time based on the periodically acquired local node indicators.

[0049] In this embodiment, the content demand priority is predicted by analyzing historical access data through a content demand optimization scheduling algorithm, and the node allocation is dynamically adjusted in combination with geographical location information.

[0050] In this embodiment, the method for generating the node resource scheduling signal is as follows:

[0051] Based on the content request characteristics of the user side, content type requirement parameters are extracted, and a content requirement optimization scheduling algorithm is used to mobilize weighted scoring nodes according to the content request feature vector, and node resource scores are calculated according to priority queues.

[0052] Based on the node resource scoring-driven node time slice rolling update mechanism, the node resource fluctuation trend is corrected, a node resource scheduling signal is generated, and the validity period of the node resource scheduling signal is adjusted according to the network load and node resource changes.

[0053] In this embodiment, the method for constructing the multi-level cache priority system is as follows:

[0054] The cache layer is initialized and configured based on node resource capabilities and network topology. The layered structure of hot spot fast cache layer, regional mid-term cache layer and long tail basic cache layer is defined. The content attribute vector is extracted based on the access popularity parameter and regional access distribution of the acquired content.

[0055] The content attribute vectors are classified according to access popularity thresholds and resource consumption costs. Hot content is quickly cached at higher levels using a rule tree weighted scoring algorithm. The regional delivery level of the content is determined based on regional distribution parameters. The residence time of the node resource scheduling signal at different levels is controlled. A cache distribution mapping table is generated to define the distribution density and cache ratio of content at each level in different geographical regions, and a multi-level cache priority system is constructed.

[0056] In this embodiment, a distributed architecture is adopted, consisting of a resource acquisition module, a scheduling signal generation module, a multi-level cache priority management module, a path modeling and scheduling module, and a traffic identification and governance module. The modules interact with each other through message queues and a high-speed cache database to achieve dynamic scheduling and governance of the PCDN network.

[0057] The resource acquisition module is deployed on each PCDN node via distributed network probes to collect real-time hardware resource status data. Collected metrics include CPU load, bandwidth utilization, storage space usage, and network latency. The collected data is written to a message queue and consumed in real-time by the scheduling signal generation module. A content demand optimization scheduling algorithm is used to integrate node hardware resource status with user geographic location and content demand priority. This algorithm constructs a content demand vector for user requests and calculates a node score based on the matching degree between node resources and location, ultimately dynamically generating a node resource scheduling signal. This scheduling signal guides subsequent cache allocation and path selection.

[0058] The multi-level cache priority management module issues cache management instructions in the network based on node resource scheduling signals. The system prioritizes content for caching based on the characteristics of hot content in different time periods and regions, combined with the timeliness of the content and the frequency of user access.

[0059] The system initialized three cache levels:

[0060] (1) Hotspot caching layer: used to cache highly popular and time-sensitive content;

[0061] (2) Regional mid-term cache layer: used to cache trending content that is frequently accessed in the region;

[0062] (3) Long-tail basic cache layer: used to cache low-frequency long-tail content.

[0063] The management service dynamically generates a cache distribution mapping table based on scheduling signals and content attributes, and synchronizes it to each edge node to achieve differentiated cache allocation for different regions and priorities.

[0064] Based on the PCDN network topology, a path feature vector is generated for each data transmission path. The service constructs a path matrix and calculates the similarity and intersection degree between paths. Potential conflicts are identified through path intersection modeling and scheduling. The transmission path is dynamically adjusted according to node load, resource utilization, and scheduling signals. At the same time, the service allocates an independent scheduling window to each node, isolating the node's data transmission tasks in a discrete time slice manner to avoid resource contention and improve overall transmission efficiency.

[0065] The traffic identification and governance module constructs a peer-to-peer content delivery network authorization feature library for classifying and identifying node traffic. The module comprehensively judges access behavior based on content type, access time, access frequency, and the degree of deviation from content timeliness, distinguishing between authorized traffic and unauthorized abuse traffic. It can also dynamically adjust the suppression intensity based on the overall network load, achieving refined governance.

[0066] When unauthorized traffic abuse is detected, the system executes a tiered suppression strategy based on the current load status of the network nodes:

[0067] When the node is under low load, abnormal traffic is rate-limited.

[0068] When a node is under high load, abnormal traffic is blocked directly to prevent resources from being abused.

[0069] In this embodiment, the content timeliness is used to perform deviation analysis on node access traffic to identify unauthorized abuse traffic, including video update frequency and file popularity decay period. By comparing the actual access time of the node request content with the expected access time defined by the content timeliness feature, it is determined whether the request meets the authorized access, and the unauthorized abuse traffic is marked and graded according to the degree of deviation.

[0070] In this embodiment, the multi-level caching priority system prioritizes caching high-priority content to edge nodes and caches low-priority content on demand based on the bandwidth consumption of the network nodes that transmit content back to the point-to-point distribution network. It also reduces the use of invalid cache through adaptive eviction.

[0071] In this embodiment, the path cross-modeling and scheduling method constructs the data transmission paths of each node in the point-to-point content delivery network into a path matrix. By analyzing the similarity and cross-relationship of the path feature vectors in the path matrix, path conflicts are identified, and the data transmission paths are dynamically adjusted based on node load and resource utilization.

[0072] In this embodiment, the system deploys multiple distributed network probes in the PCDN network to periodically collect hardware resource status data of each node. Resource data: Ri = {C} i M i B i S i L i}, where C i For CPU utilization, M i For memory utilization, B i S represents the remaining network bandwidth. i The available storage space is dist(u, i), which represents the geographic location of the node.

[0073] Based on content requirement priority P u Based on the node's geographical location dist(u, i), a resource scheduling scoring function is constructed:

[0074] ,

[0075] Among them, α1~α5 are weighting factors that can be dynamically adjusted according to the network strategy. The scheduling control center calculates the score of all candidate nodes, generates node resource scheduling signals through softmax normalization, and sends the signals to the PCDN network to guide the dynamic matching and allocation of node resources.

[0076] Based on node resource scheduling signals, combined with the distribution of regional hot content, content timeliness, and user access frequency, a multi-level content caching priority system is constructed. In order to analyze the dynamic relationship of network node load changes over time, the system constructs a path cross-modeling scheduling method, which vectorizes the transmission paths between nodes and defines the node transmission path feature vector. The larger the scheduling window, the more content transmission tasks the node receives, thereby achieving load balancing and dynamic scheduling.

[0077] The PCDN authorized content distribution feature library includes: node authorization certificate hash, content distribution identifier, node behavior pattern feature vector, and a matching score calculated for the traffic packet Pkt entering the node.

[0078] ,

[0079] Wherein, AuthScore(Pkt) is the authenticity score; δ1, δ2, and δ3 are weight coefficients; Match(CertHash) is the certificate hash matching item; Match(CID) is the cloud service provider matching item; and Sim(Behav) is the behavioral similarity.

[0080] When AuthScore(Pkt) < θ, it is determined to be unauthorized abuse of traffic.

[0081] The stepped suppression strategy assumes that the real-time load of the node is Load. i Then the flow suppression coefficient is:

[0082] ,

[0083] Among them, L low Low load threshold; L high High load threshold; Φ i This is the flow suppression coefficient.

[0084] Rate limiting is applied when nodes are under low load, and blocking is applied when nodes are under high load, to prevent unauthorized abuse from consuming resources in the PCDN network.

[0085] Through node resource status awareness, intelligent content scheduling, multi-level cache priority, path vector model, and authorized traffic governance mechanism, the following can be achieved:

[0086] Node resource utilization rate increased by 20%–40%;

[0087] Content delivery latency reduced by 15%–30%;

[0088] Unauthorized traffic abuse reduced by 60%–90%;

[0089] The significant improvement in overall network throughput ensures that the PCDN network maintains high performance and stability in large-scale content distribution scenarios over the long term.

[0090] In this embodiment, the independent scheduling window dynamically adjusts its length and start time based on the node's historical load, resource availability, and access request characteristics, and combines this with a discrete time slice partitioning strategy to enable each node to independently execute node data transmission and cache scheduling tasks in different time periods.

[0091] In this embodiment, the unauthorized abuse traffic refers to traffic generated by unauthorized nodes or abnormal access behavior. It is characterized by abnormal node access frequency, requested content deviating from timeliness indicators, and network resources occupied exceeding the node's carrying capacity. When the peer-to-peer content delivery network authorization feature library identifies features that deviate from the authorized access mode, it is determined to be unauthorized abuse traffic and a tiered suppression strategy is triggered.

[0092] In this embodiment, the tiered suppression strategy gradually increases the restriction intensity according to the multi-level suppression threshold, including sequentially implementing request frequency reduction, bandwidth allocation restriction, task queuing delay injection and forced interruption rescheduling on the target node or target traffic, and performing progressive suppression scheduling for abnormal behavior.

[0093] This invention also provides a PCDN optimization system based on intelligent scheduling and dynamic governance, specifically including:

[0094] Intelligent node scheduling module: Based on distributed network probes to collect hardware resource status data of each node in the peer-to-peer content delivery network, the hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm.

[0095] Dynamic content caching optimization module: The peer-to-peer content distribution network issues instructions in conjunction with the node resource scheduling signals, and constructs a multi-level caching priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path;

[0096] Independent scheduling and allocation module: The node data transmission path is vectorized using the path cross modeling scheduling method, the dynamic relationship between the load and resource utilization of each node is analyzed, and the resource access time sequence is divided into discrete time slices to allocate independent scheduling windows to different network nodes;

[0097] Unauthorized traffic abuse management module: Constructs a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized traffic abuse, and implements a tiered suppression strategy for unauthorized traffic abuse based on real-time network load, limiting traffic transmission during low load and blocking it during high load according to the network node load status.

[0098] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A PCDN optimization method based on intelligent scheduling and dynamic governance, characterized in that, include: S1: Based on the distributed network probe, collect hardware resource status data of each node in the peer-to-peer content delivery network. The hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm. S2: The point-to-point content distribution network issues instructions in conjunction with the node resource scheduling signal, and constructs a multi-level cache priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path; S3: The path cross modeling scheduling method is used to vectorize the data transmission path of the nodes, analyze the dynamic relationship between the load and resource utilization of each node, divide the resource access time sequence into discrete time slices, and allocate independent scheduling windows for different network nodes. S4: Construct a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized abuse traffic, and implement a tiered suppression strategy for unauthorized abuse traffic based on the real-time network load. Depending on the network node load, traffic transmission is limited when the load is low and blocked when the load is high.

2. The method according to claim 1, characterized in that, The hardware resource status data includes bandwidth load and network latency. The computing resources and network transmission capabilities of each node are evaluated in real time based on the periodically acquired local node indicators.

3. The method according to claim 1, characterized in that, The content demand priority is predicted by analyzing historical access data through a content demand optimization scheduling algorithm, and the node allocation is dynamically adjusted in combination with geographical location information.

4. The method according to claim 1, characterized in that, The method for generating the node resource scheduling signal is as follows: Based on the content request characteristics of the user side, content type requirement parameters are extracted, and a content requirement optimization scheduling algorithm is used to mobilize weighted scoring nodes according to the content request feature vector, and node resource scores are calculated according to priority queues. Based on the node resource scoring-driven node time slice rolling update mechanism, the node resource fluctuation trend is corrected, a node resource scheduling signal is generated, and the validity period of the node resource scheduling signal is adjusted according to the network load and node resource changes.

5. The method according to claim 2, characterized in that, The method for constructing the multi-level cache priority system: The cache layer is initialized and configured based on node resource capabilities and network topology. The layered structure of hot spot fast cache layer, regional mid-term cache layer and long tail basic cache layer is defined. The content attribute vector is extracted based on the access popularity parameter and regional access distribution of the acquired content. The content attribute vectors are classified according to access popularity thresholds and resource consumption costs. Hot content is quickly cached at higher levels using a rule tree weighted scoring algorithm. The regional delivery level of the content is determined based on regional distribution parameters. The residence time of the node resource scheduling signal at different levels is controlled. A cache distribution mapping table is generated to define the distribution density and cache ratio of content at each level in different geographical regions, and a multi-level cache priority system is constructed.

6. The method according to claim 5, characterized in that, The content timeliness is used to perform deviation analysis on node access traffic to identify unauthorized abuse traffic, including video update frequency and file popularity decay period. By comparing the actual access time of the node request content with the expected access time defined by the content timeliness feature, it is determined whether the request meets the authorized access, and the unauthorized abuse traffic is marked and graded according to the degree of deviation.

7. The method according to claim 4, characterized in that, The multi-level caching priority system prioritizes high-priority content to edge nodes and caches low-priority content on demand based on the bandwidth consumption of network nodes transmitting content in a point-to-point distribution. It also reduces invalid cache usage through adaptive eviction.

8. The method according to claim 2, characterized in that, The path cross-modeling and scheduling method constructs a path matrix of data transmission paths for each node in a peer-to-peer content delivery network. By analyzing the similarity and cross-relationship of path feature vectors in the path matrix, path conflicts are identified, and data transmission paths are dynamically adjusted based on node load and resource utilization.

9. The method according to claim 4, characterized in that, The independent scheduling window dynamically adjusts its length and start time based on the node's historical load, resource availability, and access request characteristics, and combines this with a discrete time slice partitioning strategy to enable each node to independently execute node data transmission and cache scheduling tasks in different time periods.

10. The method according to claim 4, characterized in that, The unauthorized abuse traffic refers to traffic generated by unauthorized nodes or abnormal access behavior. It is characterized by abnormal node access frequency, requested content deviating from timeliness indicators, and network resources occupied exceeding the node's carrying capacity. When the peer-to-peer content distribution network authorization feature library identifies features that deviate from the authorized access mode, it is determined to be unauthorized abuse traffic and a tiered suppression strategy is triggered.

11. The method according to claim 7, characterized in that, The tiered suppression strategy progressively increases the restriction intensity based on multi-level suppression thresholds, including sequentially implementing request frequency reduction, bandwidth allocation restrictions, task queuing delay injection, and forced interruption rescheduling for target nodes or target traffic, and gradually suppressing and scheduling abnormal behavior.

12. A PCDN optimization system based on intelligent scheduling and dynamic governance, used to execute the method as described in any one of claims 1-11, characterized in that, include: Intelligent node scheduling module: Based on distributed network probes to collect hardware resource status data of each node in the peer-to-peer content delivery network, the hardware resource status data is dynamically used to generate node resource scheduling signals by combining the user's geographical location and content demand priority through a content demand optimization scheduling algorithm. Dynamic content caching optimization module: The peer-to-peer content distribution network issues instructions in conjunction with the node resource scheduling signals, and constructs a multi-level caching priority system based on the hot content of different time periods and regions, combined with the timeliness of the content and the frequency of user access, and allocates the optimal node and data transmission path; Independent scheduling and allocation module: The node data transmission path is vectorized using the path cross modeling scheduling method, the dynamic relationship between the load and resource utilization of each node is analyzed, and the resource access time sequence is divided into discrete time slices to allocate independent scheduling windows to different network nodes; Unauthorized traffic abuse management module: Constructs a peer-to-peer content delivery network authorization feature library to classify and identify node traffic, distinguish between authorized peer-to-peer content delivery network traffic and unauthorized traffic abuse, and implements a tiered suppression strategy for unauthorized traffic abuse based on real-time network load, limiting traffic transmission during low load and blocking it during high load according to the network node load status.

Images