Video stream scheduling method, system and device of power video cascade network and medium

Abstract

The invention discloses a video stream scheduling method, system and device for a power video cascade network, and a medium. The method comprises the following steps: acquiring state data of each node in the power video cascade network; converting the video retrieval response duration of each node by using a depth compensation factor to obtain a compensation response value of each node, and screening out a plurality of video advantage nodes based on the compensation response value and each piece of first video storage information; the method comprises the following steps: when a to-be-scheduled node in the power video cascade network generates a video retrieval request and a corresponding first forwarding load rate is higher than a preset load threshold value, determining video content corresponding to the video retrieval request; and if the second forwarding load rate corresponding to each acquired video advantage node is lower than a preset load threshold value, sending the video retrieval request to a target scheduling node for processing so as to realize load balancing. According to the invention, the video stream scheduling accuracy of the power video cascade network can be improved.

Description

Technical Field

[0001] This invention relates to the field of power video surveillance, and in particular to a method, system, device, and medium for scheduling video streams in a power video cascaded network. Background Technology

[0002] In power video surveillance systems, inspection video data from critical facilities such as substations needs to be aggregated and accessed through multi-level cascaded networks (such as station-end access, regional aggregation, and dispatch center). This network has a tree-like or hierarchical topology, with significant differences in data processing capabilities, storage configurations, and distances from the core access point among nodes at different levels. Without an effective video stream scheduling mechanism, all access requests will default to transmission along fixed paths or be directly responded to by the original storage nodes, easily leading to two major problems: First, uneven network traffic distribution, with some critical forwarding nodes becoming overloaded due to concentrated requests, becoming performance bottlenecks and affecting system stability; second, inconsistent access response experiences, as video data from deep-level nodes at the network edge requires multiple hops, resulting in excessively long user wait times, making it difficult to meet the emergency needs for real-time viewing of critical videos in power operation and maintenance.

[0003] In existing technologies, a typical approach is real-time load-based scheduling, which monitors the CPU, memory, or instantaneous bandwidth utilization of each node and distributes new requests to the node with the lowest current load. Another common strategy is content location-based proximity scheduling, which prioritizes retrieving data from the node geographically or topologically closest to the requester that stores the required video. However, these methods all suffer from accuracy deficiencies in the specific scenario of power video cascading networks. The fundamental reason is that they ignore the systematic and non-linear impact of node hierarchy depth, a key structural factor, on response performance. Hierarchical depth directly determines the inherent path length (hop count) and cumulative forwarding latency of data transmission. Existing technologies compare nodes at different levels on the same evaluation plane, which leads to a systematic underestimation of the actual service potential of deep nodes due to their inherent path delays (especially when the node has pre-stored high-frequency video due to intelligent caching), while shallow nodes may be overestimated. This performance misjudgment caused by ignoring the depth of the network topology results in insufficient accuracy in scheduling decisions, making it impossible to accurately identify the truly optimal response node after comprehensively considering path cost and local caching advantages. As a result, it is difficult to achieve true cross-level load balancing and overall response time optimization. Summary of the Invention

[0004] This invention provides a video stream scheduling method, system, device, and medium for power video cascade networks, which can improve the accuracy of video stream scheduling in power video cascade networks.

[0005] This invention provides a video stream scheduling method for a power video cascaded network, comprising: Acquire the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth; Based on the node hierarchy depth, a corresponding depth compensation factor is determined for each node, and the video retrieval response time of each node is calculated using the depth compensation factor to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, several video advantage nodes are selected. When a node to be scheduled in a power video cascade network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined. If the video content exists in each of the video advantage nodes and the second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

[0006] This invention, through acquiring multi-dimensional state data including first video storage information, video retrieval response time, and node hierarchy depth, can comprehensively incorporate a node's content storage capacity, historical service performance, and physical location information within the network into the decision-making process, ensuring the accuracy of scheduling decisions from the data source. By determining a depth compensation factor based on node hierarchy depth and calculating the compensation response value by adjusting the response time, the interference of purely physical topology on performance evaluation is eliminated, allowing nodes at different levels to compare their true processing efficiency on a fair benchmark. Based on a comprehensive screening using the compensation response value and first video storage information, nodes that possess both high-efficiency processing capabilities (reflected after compensation) and store frequently accessed content can be accurately identified as video advantage nodes. This process ensures that the selected node set represents high-quality targets within the network that are truly capable and resourceful enough to quickly serve access requests, providing a reliable target pool for subsequent precise scheduling. Further verification of high load on request nodes confirms that the requested content resides within video-advantageous nodes, guaranteeing the effectiveness of scheduling. Further verification confirms that the current load on the target advantage nodes is low, ensuring their redundancy in accepting new requests. Finally, specific target scheduling nodes are determined based on path information. This ensures that each scheduling action "guides the request to the correct and capable target node at the right time, for the right content," thereby maximizing the success rate of each scheduling action and overall service efficiency while achieving load balancing, fundamentally improving scheduling accuracy.

[0007] Further, determining the corresponding depth compensation factor for each node based on the node hierarchy depth includes: Based on the node hierarchy depth of each node, several node hierarchy groups are obtained; The average video retrieval response time of each node-level group is calculated to determine the baseline response time corresponding to each node level. Using the baseline response time of a preset reference level as a standard, the depth compensation factor for each other level is calculated, wherein the depth compensation factor is equal to the ratio of the baseline response time of the preset reference level to the baseline response time of the current level.

[0008] By determining the depth compensation factor based on the node hierarchy depth and calculating the compensation response value by adjusting the response time, the interference of pure physical topology on performance evaluation is eliminated, allowing nodes at different levels to compare their true processing efficiency on a fair benchmark.

[0009] Furthermore, the step of selecting several video advantage nodes based on the compensation response value and the first video storage information includes: The high-frequency video segments to be stored are determined based on the first video storage information of each node, and the high-frequency storage ratio of each node is calculated in combination with the high-frequency video segments. The high-frequency video segments are video segments that have been accessed more than a preset number of times within a preset statistical period. Based on the compensation response value, calculate the response speed value of each node relative to its respective node hierarchy group; Based on the high-frequency storage ratio and the response speed value, calculate the comprehensive advantage score of each node; All nodes are ranked according to the comprehensive advantage score, and several video advantage nodes are determined based on the ranking results.

[0010] This comprehensive screening, based on fair compensation response values ​​and primary video storage information reflecting content popularity, can accurately identify nodes that possess both efficient processing capabilities and store frequently accessed content as video advantage nodes. This ensures that the selected node set represents high-quality targets in the network that are truly capable and resourceful in quickly serving access requests, providing a reliable target pool for subsequent precise scheduling.

[0011] Furthermore, when a node to be scheduled in the power video cascade network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined, including: Obtain the request processing rate and traffic transmission rate of the scheduled node that made the video access request within the statistical period, and calculate the first forwarding load rate based on the request processing rate and traffic transmission rate; If the first forwarding load rate is higher than the preset load threshold, the video access request is parsed to extract the video resource identifier; The video resource identifier is matched with the local video index list of each video advantage node; wherein, the local video index list is generated based on the second video storage information corresponding to each video advantage node and is used to record the locally stored video resource identifier. The video content is determined based on the matching query results.

[0012] This method of calculating load rate based on real-time rate ensures that scheduling is triggered only on truly overloaded nodes, avoiding unnecessary scheduling operations and making scheduling timing more precise. Secondly, parsing requests and matching them with video indexes accurately confirms whether the "video advantage node" actually stores the requested specific content, preventing invalid scheduling to nodes without content and ensuring the effectiveness of each scheduling action. These two factors work together to ensure that subsequent load balancing decisions are based on accurate demand and resource matching.

[0013] Further, the calculation of the first forwarding load rate based on the request processing rate and the traffic transmission rate includes: Calculate the request load rate of the node to be scheduled based on the request processing rate and the rated request processing capacity; Calculate the traffic load rate of the node to be scheduled based on the traffic transmission rate and the rated traffic transmission capacity; The first forwarding load rate is obtained by weighting the request load rate and the traffic load rate.

[0014] This method of calculating load rate based on real-time rate ensures that scheduling is triggered only for truly overloaded nodes, avoiding unnecessary scheduling operations and making scheduling timing more precise.

[0015] Further, if the video content exists in each of the video advantage nodes, and the second forwarding load rate corresponding to each of the video advantage nodes is lower than a preset load threshold, then the video retrieval request is sent to the target scheduling node for processing to achieve load balancing, including: Obtain the second forwarding load rate of the video advantage node; The video content is matched with the second video storage information corresponding to the video advantage node. If the match is successful and the second forwarding load rate is lower than the preset load threshold, the remaining processing capacity of each video advantage node, as well as the number of transmission hops and historical transmission delay of the potential forwarding path from each video advantage node to the node to be scheduled are obtained. Based on the transmission hop count, the historical transmission delay, and the remaining processing capacity, calculate the comprehensive path score for each potential forwarding path; The target forwarding path and target scheduling node are determined based on the comprehensive path score, and the video access request is scheduled to the target scheduling node for processing according to the target forwarding path to achieve load balancing.

[0016] This approach ensures scheduling effectiveness by matching content and storage information; it filters out nodes with service capacity by comparing load rates; and then, by comprehensively considering multiple factors such as the remaining processing capacity of nodes, path hop count, and historical latency, it calculates a comprehensive path score. This achieves a deeper decision-making process from schedulable to optimal scheduling, ensuring that while achieving load balancing, it selects the target with the optimal network path and the strongest node processing capacity, thereby accurately improving the overall scheduling accuracy and efficiency.

[0017] Further, the calculation of the path comprehensive score for each potential forwarding path based on the transmission hop count, the historical transmission delay, and the remaining processing capacity includes: The transmission hop count, the historical transmission delay, and the remaining processing capacity are normalized respectively to obtain the corresponding hop count normalization value, delay normalization value, and capacity normalization value. Based on preset hop count weight coefficients, delay weight coefficients, and capability weight coefficients, the normalized values ​​of the hop count, delay, and capability are weighted and summed to calculate the comprehensive path score for each potential forwarding path.

[0018] Another embodiment of the present invention provides a video stream scheduling system for a power video cascaded network, comprising: The acquisition module is used to acquire the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth; The filtering module determines a corresponding depth compensation factor for each node based on the node hierarchy depth, and uses the depth compensation factor to calculate the video retrieval response time of each node to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, several video advantage nodes are filtered out. The scheduling module is used to determine the video content corresponding to the video access request when a node to be scheduled in the power video cascaded network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold. If the video content exists in each of the video advantage nodes and the obtained second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

[0019] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the video stream scheduling method for the power video cascaded network as described in the present invention.

[0020] Another embodiment of the present invention provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the video stream scheduling method of the power video cascaded network as described in the present invention. Attached Figure Description

[0021] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating an embodiment of the video stream scheduling method for a power video cascaded network provided in this application; Figure 2 This is a flowchart illustrating one embodiment of steps S201 to S204 provided in this application; Figure 3 This is a flowchart illustrating one embodiment of steps S301 to S304 provided in this application; Figure 4 This is a flowchart illustrating one embodiment of steps S401 to S404 provided in this application; Figure 5 This is a system schematic diagram of one embodiment of the video stream scheduling system for the power video cascading network provided in this application; Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0025] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0028] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0029] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0030] In power video surveillance systems, inspection video data from critical facilities such as substations needs to be aggregated and accessed through multi-level cascaded networks (such as station-end access, regional aggregation, and dispatch center). This network exhibits a tree-like or hierarchical topology, with significant differences in data processing capabilities, storage configurations, and distances from the core access point among nodes at different levels. Without an effective video stream scheduling mechanism, user waiting times will be excessively long, making it difficult to meet the emergency needs for real-time viewing of critical videos in power operation and maintenance. Existing technologies often employ video stream scheduling strategies based on simple load indicators or static rules, but these methods cannot accurately identify the truly optimal response node after comprehensively considering path cost and local caching advantages, thus hindering the achievement of true cross-level load balancing and overall response time optimization.

[0031] See Figure 1 To improve the accuracy of video stream scheduling in power video cascade networks, an embodiment of the present invention provides a video stream scheduling method for power video cascade networks, comprising steps S101 to S103: Step S101: Obtain the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth; In some embodiments, a centralized dispatch manager or distributed status collector is deployed in the power video cascade network to actively collect status data from each network node through periodic polling or event-triggered reporting. After obtaining status data such as the first video storage information, video retrieval response time, and node hierarchy depth, these data are updated to the status database of the dispatch manager in real time to facilitate subsequent selection of advantageous video nodes and scheduling decisions.

[0032] It should be noted that the acquisition of the first video storage information is achieved by querying the local video storage database or file system metadata of each node to extract the identifier list, storage time, segment size, and storage location of the video segments stored by the node, and then summarizing them to form the video storage list for that node. The acquisition of video retrieval response time is achieved by deploying monitoring probes on each node to record the timestamp of each video retrieval request processed by that node, i.e., the time interval from receiving the retrieval request to starting to return video stream data, and calculating the average response time within a preset period (e.g., the most recent hour) as the current video retrieval response time for that node. The acquisition of node hierarchy depth is based on a pre-configured network topology diagram, defining the core node as hierarchy depth 0, with each downstream node increasing in depth by 1, and storing this hierarchy depth as a static attribute bound to the node identifier in the network topology database.

[0033] Step S102: Determine the corresponding depth compensation factor for each node based on the node hierarchy depth, and use the depth compensation factor to calculate the video retrieval response time of each node to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, select several video advantage nodes. In some embodiments, determining the corresponding depth compensation factor for each node based on the node level depth includes: grouping each node according to its node level depth to obtain several node level groups; calculating the average video retrieval response time of each node level group to determine the baseline response time corresponding to each node level; and calculating the depth compensation factor for other levels using the baseline response time of a preset reference level as a standard, wherein the depth compensation factor is equal to the ratio of the baseline response time of the preset reference level to the baseline response time of the current level. Specifically, firstly, after obtaining the node level depth (e.g., value 1 represents the core layer, 2 represents the aggregation layer, and 3 represents the access layer), all nodes are grouped according to this value, and nodes with the same level depth value are grouped into the same logical set, thereby forming several node level groups (such as core layer group, aggregation layer group, and access layer group) that correspond one-to-one with the network level. Next, for each node-level group formed above, the video retrieval response time reported by all nodes within the group in the most recent complete statistical period is extracted, and the arithmetic mean of all these video retrieval response times within the group is calculated. This arithmetic mean is then determined as the baseline response time corresponding to that node level. Finally, the network core layer or the layer with the most stable performance and lowest latency is selected as the preset reference layer, and the depth compensation factor for the current layer is calculated using the formula: Depth Compensation Factor = Baseline Response Time of Preset Reference Layer / Baseline Response Time of Current Layer.

[0034] It should be noted that nodes within each node-level group are considered to be in similar network locations and face similar underlying network transmission delays.

[0035] For example, if the average response time of all core layer nodes is calculated to be 50 milliseconds, then the baseline response time of the core layer is 50 milliseconds; similarly, the baseline response time of the aggregation layer is calculated to be 120 milliseconds, and that of the access layer is 200 milliseconds.

[0036] By determining the depth compensation factor based on the node hierarchy depth and calculating the compensation response value by adjusting the response time, the interference of pure physical topology on performance evaluation is eliminated, allowing nodes at different levels to compare their true processing efficiency on a fair benchmark.

[0037] In some embodiments, the video retrieval response time of each node is converted using the depth compensation factor to obtain the compensation response value of each node. Specifically, after obtaining the depth compensation factor for each level, a simple multiplication operation is used to calculate the compensation response value, i.e., compensation response value = video retrieval response time × depth compensation factor. For example, a node located in the access layer (assuming its depth compensation factor is 0.25) with an original video retrieval response time of 240 milliseconds will have its compensation response value converted to 240 × 0.25 = 60 milliseconds; while a node located in the core layer (compensation factor is 1.0) with an original response time of 55 milliseconds will have its compensation response value remain at 55 milliseconds.

[0038] Please refer to Figure 2 In some embodiments, the step of selecting several video advantage nodes based on the compensation response value and the first video storage information includes steps S201 to S204: Step S201: Determine the stored high-frequency video segments based on the first video storage information of each node, and calculate the high-frequency storage ratio of each node in combination with the high-frequency video segments, wherein the high-frequency video segments are video segments that have been accessed more than a preset number of times within a preset statistical period. In some embodiments, for each node, the first video storage information of each node is identified to identify all video segments that have been accessed more than a preset threshold number (e.g., 10 times) within a preset statistical period (e.g., the past 24 hours), and these segments are marked as high-frequency video segments for that node. Subsequently, the high-frequency storage ratio of that node is calculated. The formula is: .in, The number of high-frequency video clips stored in this node. The total number of video clips stored for this node.

[0039] Step S202: Based on the compensation response value, calculate the response speed value of each node relative to its respective node hierarchy group; In some embodiments, for each node, the baseline response time of its node hierarchy group is obtained. Then calculate the response speed value of the node. The formula is: ,in, This is the compensation response value of that node.

[0040] It should be noted that a larger response speed value (V>1) indicates that the node's processing speed is ahead of the average level of its layer; a smaller response speed value (V<1) indicates that its speed is behind the average level.

[0041] Step S203: Calculate the comprehensive advantage score of each node based on the high-frequency storage ratio and the response speed value. In some embodiments, when each of the aforementioned high-frequency storage ratios is obtained With the aforementioned response speed value Next, the two values ​​are normalized to obtain normalized high-frequency storage ratio and normalized response speed values ​​(e.g., scaled to the 0-1 range); then, a linear weighted summation model is applied, with the following formula: ,in, and The preset weighting coefficients ( This is used to adjust the emphasis on content value and processing capability in the evaluation, thereby calculating the comprehensive advantage score for each node. .

[0042] It should be noted that the overall advantage score This comprehensively reflects the advantages of the node in terms of both content reserves and response performance.

[0043] Step S204: Sort all nodes according to the comprehensive advantage score, and determine several video advantage nodes based on the sorting results.

[0044] In some embodiments, the overall advantage score of all nodes is calculated. The nodes are sorted in descending order to form a global list of advantageous nodes. Then, video advantageous nodes are selected from the global list of advantageous nodes according to preset rules, and the corresponding information (such as node ID, stored content index, current rating, etc.) is updated to an independent advantageous node resource pool for real-time querying and selection during subsequent video access request scheduling.

[0045] It should be noted that the preset rules can be: (1) selecting the top K nodes (K is a fixed number); or (2) selecting all nodes whose score S exceeds a certain advantage threshold.

[0046] This comprehensive screening, based on fair compensation response values ​​and primary video storage information reflecting content popularity, can accurately identify nodes that possess both efficient processing capabilities and store frequently accessed content as video advantage nodes. This ensures that the selected node set represents high-quality targets in the network that are truly capable and resourceful in quickly serving access requests, providing a reliable target pool for subsequent precise scheduling.

[0047] Step S103: When a node to be scheduled in the power video cascade network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined. If the video content exists in each of the video advantage nodes and the second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

[0048] Please refer to Figure 3 In some embodiments, when a node to be scheduled in the power video cascade network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined, including steps S301 to S304: Step S301: Obtain the request processing rate and traffic transmission rate of the scheduled node that made the video access request within the statistical period, and calculate the first forwarding load rate based on the request processing rate and the traffic transmission rate. In some embodiments, when any node in the power video cascade network, i.e., the node to be scheduled, receives a new video access request, the distributed state collector deployed on that node immediately triggers a load assessment, counting the number of requests successfully processed per unit time within the most recent sliding time window (e.g., the past minute) as the request processing rate. The total amount of video data forwarded per unit time is used as the traffic transmission rate. .

[0049] In some embodiments, calculating the first forwarding load rate based on the request processing rate and the traffic transmission rate includes: calculating the request load rate of the node to be scheduled based on the request processing rate and the rated request processing capacity; calculating the traffic load rate of the node to be scheduled based on the traffic transmission rate and the rated traffic transmission capacity; and performing a weighted calculation on the request load rate and the traffic load rate to obtain the first forwarding load rate. Specifically, firstly, when the request processing rate is obtained... Then, based on the request processing rate and rated request processing capacity The request load rate of the node to be scheduled is calculated. The calculation formula is: ,in, It is the actual request processing rate (e.g., the number of requests processed per second) within the statistical period. This is the node's pre-defined or configured rated request processing capacity (i.e., the theoretical upper limit of requests it can process per unit time). Then, based on the traffic transmission rate... and the node's rated flow transmission capacity The flow load rate was calculated. Finally, regarding the request load rate... and traffic load rate The weighted calculation is performed to obtain the final first forwarding load rate. Its calculation formula is Among them, the weighting coefficient and A positive value preset based on the importance of the control plane and the data plane (e.g., both set to 0.5), and satisfying... .

[0050] It should be noted that the first forwarding load rate is a comprehensive indicator between 0 and 1 (or 0% to 100%), used to accurately reflect the current composite load level of the node.

[0051] This method of calculating load rate based on real-time rate ensures that scheduling is triggered only for truly overloaded nodes, avoiding unnecessary scheduling operations and making scheduling timing more precise.

[0052] Step S302: If the first forwarding load rate is higher than the preset load threshold, the video access request is parsed to extract the video resource identifier. In some embodiments, after calculating the first forwarding load rate L, the first forwarding load rate L is compared with a preset load threshold (e.g., 0.8). If L > threshold, it is determined that the node is in an overloaded state and the scheduling process needs to be started. At this time, the currently triggered video access request message (e.g., RTSP DESCRIBE / PLAY request or HTTP GET request) is immediately parsed, and a unique video resource identifier (VID) is extracted according to the predetermined protocol specifications (e.g., from the URL or SDP information).

[0053] In some embodiments, if the load does not exceed the threshold, the request is processed at this node according to the normal process, that is, no video stream scheduling is performed.

[0054] Step S303: Match the video resource identifier with the local video index list of each video advantage node; wherein, the local video index list is generated based on the second video storage information corresponding to each video advantage node and is used to record the locally stored video resource identifier. In some embodiments, the node to be scheduled initiates a parallel query to a pre-built resource pool of video advantage nodes in the system using the extracted Video Resource Identifier (VID). At this time, the query request is sent to all video advantage nodes recorded in the resource pool. After receiving the query request, each advantage node performs a fast search (such as a hash lookup) in its local index list to determine whether the requested VID exists.

[0055] It should be noted that each video advantage node maintains a local video index list, which is a core component of its second video storage information (i.e., a snapshot of its actual local storage content). Essentially, it is a hash table or database index that stores all local VIDs.

[0056] Step S304: Determine the video content based on the matching query results.

[0057] In some embodiments, it is determined whether at least one video advantage node has confirmed that the VID is stored. If so, it is determined that "the video content exists in each of the video advantage nodes", and the specific video content corresponding to the VID (such as the recording of a certain period of a specific camera in a substation) is determined as the target content of this scheduling decision.

[0058] In some embodiments, if all dominant nodes return no match, it is determined that the content does not exist in the dominant nodes, and this request cannot trigger load scheduling based on the dominant nodes. Other alternative strategies may be adopted (such as pulling from the origin server or being handled directly by the overloaded nodes).

[0059] This method of calculating load rate based on real-time rate ensures that scheduling is triggered only on truly overloaded nodes, avoiding unnecessary scheduling operations and making scheduling timing more precise. Secondly, parsing requests and matching them with video indexes accurately confirms whether the "video advantage node" actually stores the requested specific content, preventing invalid scheduling to nodes without content and ensuring the effectiveness of each scheduling action. These two factors work together to ensure that subsequent load balancing decisions are based on accurate demand and resource matching.

[0060] Please refer to Figure 4 If the video content exists in each of the video advantage nodes, and the second forwarding load rate corresponding to each video advantage node is lower than a preset load threshold, then the video retrieval request will be sent to the target scheduling node for processing to achieve load balancing, including steps S401 to S404: Step S401: Obtain the second forwarding load rate of the video advantage node; In some embodiments, the node to be scheduled initiates a real-time status query to the video advantage nodes that have been initially matched. Each queried video advantage node calculates its own second forwarding load rate by its distributed status collector. The calculation method is exactly the same as the first forwarding load rate (i.e., based on its own request processing rate and traffic transmission rate, and weighted with reference to its rated capacity), which will not be elaborated here. Afterwards, the second forwarding load rate is immediately returned to the querying party to determine whether these candidate nodes currently have the capacity space to receive additional scheduling requests (i.e., whether the load is lower than the same preset load threshold).

[0061] Step S402: Match the video content with the second video storage information corresponding to the video advantage node. If the match is successful and the second forwarding load rate is lower than the preset load threshold, obtain the remaining processing capacity of each video advantage node, as well as the number of transmission hops and historical transmission delay of the potential forwarding path from each video advantage node to the node to be scheduled. In some embodiments, after receiving load rate feedback from each video advantage node, the querying party performs a filtering process: only nodes with a second forwarding load rate lower than a preset load threshold (e.g., 0.7) are retained as valid candidate nodes. For each valid candidate node, the scheduler further obtains the remaining processing capacity, directly queries the network topology management system to obtain the transmission hop count of the potential forwarding path of the node to be scheduled, and obtains its recent (e.g., the average latency of the past 5 minutes) average latency value from the network performance monitoring historical database, thus determining the historical transmission latency of the path.

[0062] It should be noted that the remaining processing capacity can be simplified to (1 - second forwarding load rate) or calculated based on the difference between the rated capacity and the current rate.

[0063] Step S403: Calculate the comprehensive path score for each potential forwarding path based on the transmission hop count, the historical transmission delay, and the remaining processing capacity. In some embodiments, step S403 includes: normalizing the transmission hop count, the historical transmission delay, and the remaining processing capacity to obtain corresponding hop count normalized values, delay normalized values, and capacity normalized values; and calculating a weighted sum of the hop count normalized values, delay normalized values, and capacity normalized values ​​based on preset hop count weighting coefficients, delay weighting coefficients, and capacity weighting coefficients to obtain a comprehensive path score for each potential forwarding path. Specifically, to quantitatively compare each potential path, the transmission hop count, the historical transmission delay, and the remaining processing capacity are normalized, for example: hop count normalized value = (maximum hop count - current hop count) / (maximum hop count - minimum hop count); the calculation method for delay normalized values ​​and capacity normalized values ​​is similar. Then, a weighted summation formula is used: ,in, , and These are the normalized jump count, latency, and ability value, respectively. , and These are the corresponding preset weighting coefficients, and the sum of the three is 1.

[0064] Step S404: Determine the target forwarding path and target scheduling node based on the comprehensive path score, and schedule the video viewing request to the target scheduling node for processing according to the target forwarding path to achieve load balancing.

[0065] In some embodiments, the comprehensive scores of the paths corresponding to all valid candidate nodes are compared, and the path with the highest score is selected as the final target forwarding path. The video advantage node corresponding to this path is then determined as the target scheduling node. Subsequently, a scheduling instruction is sent to the node to be scheduled. This instruction contains the address information of the target scheduling node and necessary relay routing information. Based on this, the node to be scheduled schedules the video access request (either through redirection signaling or by establishing a streaming media forwarding channel via the target node) to the target scheduling node, which then directly provides video streaming services to the requester. This migrates the load from the overloaded original node to the advantage node with lower load, ready content, and a better path, thereby achieving global load balancing.

[0066] This approach ensures scheduling effectiveness by matching content and storage information; it filters out nodes with service capacity by comparing load rates; and then, by comprehensively considering multiple factors such as the remaining processing capacity of nodes, path hop count, and historical latency, it calculates a comprehensive path score. This achieves a deeper decision-making process from schedulable to optimal scheduling, ensuring that while achieving load balancing, it selects the target with the optimal network path and the strongest node processing capacity, thereby accurately improving the overall scheduling accuracy and efficiency.

[0067] This invention, through acquiring multi-dimensional state data including first video storage information, video retrieval response time, and node hierarchy depth, can comprehensively incorporate a node's content storage capacity, historical service performance, and physical location information within the network into the decision-making process, ensuring the accuracy of scheduling decisions from the data source. By determining a depth compensation factor based on node hierarchy depth and calculating the compensation response value by adjusting the response time, the interference of purely physical topology on performance evaluation is eliminated, allowing nodes at different levels to compare their true processing efficiency on a fair benchmark. Based on a comprehensive screening using the compensation response value and first video storage information, nodes that possess both high-efficiency processing capabilities (reflected after compensation) and store frequently accessed content can be accurately identified as video advantage nodes. This process ensures that the selected node set represents high-quality targets within the network that are truly capable and resourceful enough to quickly serve access requests, providing a reliable target pool for subsequent precise scheduling. Further verification of high load on request nodes confirms that the requested content resides within video-advantageous nodes, guaranteeing the effectiveness of scheduling. Further verification confirms that the current load on the target advantage nodes is low, ensuring their redundancy in accepting new requests. Finally, specific target scheduling nodes are determined based on path information. This ensures that each scheduling action "guides the request to the correct and capable target node at the right time, for the right content," thereby maximizing the success rate of each scheduling action and overall service efficiency while achieving load balancing, fundamentally improving scheduling accuracy.

[0068] like Figure 5 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a video stream scheduling system for a power video cascade network, comprising: The acquisition module 100 is used to acquire the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth. The filtering module 200 determines a corresponding depth compensation factor for each node based on the node hierarchy depth, and uses the depth compensation factor to calculate the video access response time of each node to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, a number of video advantage nodes are filtered out. The scheduling module 300 is used to determine the video content corresponding to the video access request when a node to be scheduled in the power video cascaded network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold. If the video content exists in each of the video advantage nodes and the obtained second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

[0069] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the video stream scheduling method for power video cascaded networks provided by any of the above-described method embodiments of the present invention.

[0070] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0071] Based on the above embodiments of the video stream scheduling method for power video cascaded networks, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the video stream scheduling method for power video cascaded networks according to any embodiment of the present invention.

[0072] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0073] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0074] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0075] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the video stream scheduling method for the power video cascaded network described in any of the above-described method embodiments of the present invention.

[0076] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0077] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A video stream scheduling method for a power video cascaded network, characterized in that, include: Acquire the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth; Based on the node hierarchy depth, a corresponding depth compensation factor is determined for each node, and the video retrieval response time of each node is calculated using the depth compensation factor to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, several video advantage nodes are selected. When a node to be scheduled in a power video cascade network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined. If the video content exists in each of the video advantage nodes and the second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

2. The video stream scheduling method for a power video cascaded network according to claim 1, characterized in that, The process of determining the corresponding depth compensation factor for each node based on the node hierarchy depth includes: Based on the node hierarchy depth of each node, several node hierarchy groups are obtained; The average video retrieval response time of each node-level group is calculated to determine the baseline response time corresponding to each node level. Using the baseline response time of a preset reference level as a standard, the depth compensation factor for each other level is calculated, wherein the depth compensation factor is equal to the ratio of the baseline response time of the preset reference level to the baseline response time of the current level.

3. The video stream scheduling method for a power video cascaded network according to claim 1, characterized in that, Based on the compensation response value and the first video storage information, several video advantage nodes are selected, including: The high-frequency video segments to be stored are determined based on the first video storage information of each node, and the high-frequency storage ratio of each node is calculated in combination with the high-frequency video segments. The high-frequency video segments are video segments that have been accessed more than a preset number of times within a preset statistical period. Based on the compensation response value, calculate the response speed value of each node relative to its respective node hierarchy group; Based on the high-frequency storage ratio and the response speed value, calculate the comprehensive advantage score of each node; All nodes are ranked according to the comprehensive advantage score, and several video advantage nodes are determined based on the ranking results.

4. The video stream scheduling method for a power video cascaded network according to claim 1, characterized in that, When a node in the power video cascade network requests video access and the corresponding first forwarding load rate is higher than a preset load threshold, the video content corresponding to the video access request is determined, including: Obtain the request processing rate and traffic transmission rate of the scheduled node that made the video access request within the statistical period, and calculate the first forwarding load rate based on the request processing rate and traffic transmission rate; If the first forwarding load rate is higher than the preset load threshold, the video access request is parsed to extract the video resource identifier; The video resource identifier is matched with the local video index list of each video advantage node; wherein, the local video index list is generated based on the second video storage information corresponding to each video advantage node and is used to record the locally stored video resource identifier. The video content is determined based on the matching query results.

5. The video stream scheduling method for a power video cascaded network according to claim 4, characterized in that, The calculation of the first forwarding load rate based on the request processing rate and the traffic transmission rate includes: Calculate the request load rate of the node to be scheduled based on the request processing rate and the rated request processing capacity; Calculate the traffic load rate of the node to be scheduled based on the traffic transmission rate and the rated traffic transmission capacity; The first forwarding load rate is obtained by weighting the request load rate and the traffic load rate.

6. The video stream scheduling method for a power video cascaded network according to claim 1, characterized in that, If the video content exists in each of the video advantage nodes, and the second forwarding load rate corresponding to each video advantage node is lower than a preset load threshold, then the video access request is sent to the target scheduling node for processing to achieve load balancing, including: Obtain the second forwarding load rate of the video advantage node; The video content is matched with the second video storage information corresponding to the video advantage node. If the match is successful and the second forwarding load rate is lower than the preset load threshold, the remaining processing capacity of each video advantage node, as well as the number of transmission hops and historical transmission delay of the potential forwarding path from each video advantage node to the node to be scheduled are obtained. Based on the transmission hop count, the historical transmission delay, and the remaining processing capacity, calculate the comprehensive path score for each potential forwarding path; The target forwarding path and target scheduling node are determined based on the comprehensive path score, and the video access request is scheduled to the target scheduling node for processing according to the target forwarding path to achieve load balancing.

7. The video stream scheduling method for a power video cascaded network according to claim 6, characterized in that, The calculation of a comprehensive path score for each potential forwarding path based on the transmission hop count, the historical transmission delay, and the remaining processing capacity includes: The transmission hop count, the historical transmission delay, and the remaining processing capacity are normalized respectively to obtain the corresponding hop count normalization value, delay normalization value, and capacity normalization value. Based on preset hop count weight coefficients, delay weight coefficients, and capability weight coefficients, the normalized values ​​of the hop count, delay, and capability are weighted and summed to calculate the comprehensive path score for each potential forwarding path.

8. A video stream scheduling system for a power video cascaded network, characterized in that, include: The acquisition module is used to acquire the status data of each node in the power video cascade network, wherein the status data includes first video storage information, video retrieval response time and node hierarchy depth; The filtering module determines a corresponding depth compensation factor for each node based on the node hierarchy depth, and uses the depth compensation factor to calculate the video retrieval response time of each node to obtain the compensation response value of each node. Based on the compensation response value and the first video storage information, several video advantage nodes are filtered out. The scheduling module is used to determine the video content corresponding to the video access request when a node to be scheduled in the power video cascaded network makes a video access request and the corresponding first forwarding load rate is higher than a preset load threshold. If the video content exists in each of the video advantage nodes and the obtained second forwarding load rate corresponding to each of the video advantage nodes is lower than the preset load threshold, the video access request is sent to the target scheduling node for processing to achieve load balancing. The target scheduling node is determined based on the forwarding path between each of the video advantage nodes and the node to be scheduled.

9. A terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the video stream scheduling method for a power video cascaded network as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the steps of the video stream scheduling method for a power video cascaded network as described in any one of claims 1-7.

Images