Video patch stream scheduling method and device based on multi-dimensional priority and medium

Abstract

The invention provides a multi-dimensional priority-based video patch stream scheduling method, which comprises the following steps of: firstly, defining parameters which comprise current time, patch windows and program individual parameters which comprise conventional streams, conventional stream establishment time, dynamic service value priorities and waiting queue lengths, and then dividing patch sets, dividing all program waiting queues into a patching set and an unpatching set based on a matching relationship between program individual parameters and patching windows, performing comprehensive priority formula calculation based on a classification result and the determined service value priority and waiting queue length, preferentially distributing a patch stream for the highest queue in the patching set, and performing priority calculation on the highest queue in the patching set; when the set is empty, a new conventional flow is established for the highest queue in the unpatching set; according to the method, the service value, the technical feasibility and the resource efficiency are comprehensively considered, and the optimal configuration of the server bandwidth and the computing resources is realized, so that the overall service quality and the operation benefit are improved.

Description

Technical Field

[0001] This invention relates to the technical field of streaming media transmission and server resource scheduling, specifically a video patch stream scheduling method, device, and medium based on multi-dimensional priority. Background Technology

[0002] With the rapid development of internet infrastructure and the widespread adoption of smart devices, online video services have become a core way for people to access information and entertainment. From on-demand movies and TV series to live sports events and news broadcasts, users have increasingly demanding requirements for the immediacy, smoothness, and picture quality of video content. In video service systems, when thousands of users simultaneously request to watch the same popular content, the traditional method of establishing a separate video stream for each user places enormous, almost catastrophic, pressure on servers and network bandwidth. This not only leads to high operating costs but also easily causes poor user experiences such as video buffering and loading failures due to resource bottlenecks.

[0003] To overcome this challenge, patch streaming technology emerged and has been recognized by academia and industry as an effective resource-saving solution. Its core idea is to utilize the concept of multicast or shared streaming.

[0004] Specifically, when the first user requests a video program, the server creates a complete regular stream for them, playing from the beginning. In the subsequent period, other users requesting the same program no longer need to start a separate complete stream from scratch. Instead, the server provides these subsequent users with a very short patch stream. This patch stream contains only the data from the moment the user requests the program to the current playback position of the first regular stream. Once a user receives the patch stream, they catch up with the playback progress of the regular stream and then share the same regular stream with the first user to receive subsequent complete video data.

[0005] While patching techniques can theoretically save significant bandwidth, their effectiveness is highly dependent on the quality of the scheduling algorithm—specifically, "which program's waiting queue should be allocated a patch stream next?". Existing classic scheduling strategies often focus on a single technical or efficiency dimension, exhibiting significant limitations. 1. Earliest Request First: This method strictly serves requests in the order they arrive, but completely ignores business value and resource efficiency. For example, an earlier request to watch a less popular archived video might block thousands of later requests from users watching crucial moments of a live final. This clearly contradicts business logic and user experience expectations, creating the unreasonable phenomenon of "low-value requests blocking high-value requests."

[0006] 2. Shortest Patch Priority: This approach may cause requests that are slightly later, require longer patch flows, but have extremely high business value, such as popular live streams, to be shelved for extended periods. However, it fails to consider queue length, potentially leading to a short patch request with only a few users being prioritized over a long patch request with a large number of users. From a macro perspective, this results in poor resource savings and lower average user wait times.

[0007] 3. Maximum Queue Length Priority: This approach fails to differentiate the importance of content requests within a queue. A queue of regular on-demand content with a large number of users will have a higher scheduling priority than a queue of ongoing live sports events with fewer users. Furthermore, it treats all queues equally, without distinguishing between patchable and non-patternable requests. Serving a queue that has exceeded its patchable window means the system needs to establish a completely new regular stream, consuming tens or even hundreds of times more resources than a patched stream. Prioritizing a queue simply because it is large can lead to system resources being quickly exhausted by a few such requests, while failing to efficiently serve a large number of patchable requests. From a global perspective, this is inefficient in terms of resource utilization.

[0008] In summary, the main problem with existing technologies lies in the singularity of their decision-making dimensions. They either focus only on time fairness, local efficiency, or service scale, lacking a comprehensive scheduling framework that can simultaneously weigh business value, what content is more important, the state of the technology, whether it can be patched at low cost, resource efficiency, how many people are served, and how many resources are consumed.

[0009] Therefore, there is an urgent need in this field for an innovative and intelligent scheduling method that can break through the limitations of traditional algorithms and make comprehensive judgments on multiple key dimensions, thereby maximizing the utilization of global resources while ensuring the experience of high-value business. Summary of the Invention

[0010] This invention provides a video patch stream scheduling method, device, and medium based on multi-dimensional priority. By comprehensively considering business value, technical feasibility, and resource efficiency, it optimizes the allocation of server bandwidth and computing resources, thereby improving overall service quality and operational efficiency, and solving the problems in the background technology.

[0011] To achieve the above objectives, the technical solution of the present invention is as follows: A video patch stream scheduling method based on multi-dimensional priority involves the following steps performed via computer equipment: S1 defines parameters, including the current time and patchable window, as well as program-specific parameters. Program-specific parameters include the regular stream, regular stream establishment time, dynamic service value priority, and waiting queue length. S2, divide the patch set. Based on the matching relationship between the regular stream, the current time and the patchable window, divide all program waiting queues into a patchable set and an unpatched set. S3, priority calculation and allocation: Based on the classification results of S2, the comprehensive priority formula is calculated using the business value priority determined by S1 and the waiting queue length. Priority is given to allocating patch flows to the highest queue in the patchable set. If the set is empty, a new regular flow is established for the highest queue in the unpattable set.

[0012] Preferably, the specific implementation steps of S1 are as follows: S11 defines global system parameters within the video platform, sets the current time parameter t, and synchronizes the clock in real time. Set the patchable window size parameter W to define the time range during which subsequent users are allowed to join the shared stream via patching after the regular stream is established; S12 defines individual program parameters and defines the regular stream as... This represents the first complete video stream established for program i, defining the regular stream establishment time parameter. Representing records The moment of transmission begins is a critical time node for calculating the length of the patch stream and determining whether it is within the patchable window. S13, Dynamically calculate business value priority parameters Quantify the importance of the program; S14, Set the waiting queue length parameter This counts the number of user requests currently waiting to receive program i-patch streams or regular streams.

[0013] Preferably, the specific algorithm flow for S13 is as follows: Three core network parameters are selected: packet loss rate z, latency rate y, and jitter rate j. The constraints are satisfied:

[0014] Define the potential user count parameter , and These represent the potential user base of program i under scenarios with packet loss rate z, latency rate y, and jitter rate j, respectively. Calculate the service value priority P of program i to achieve quantification of the linkage between network status and user scale:

[0015] The program was then divided into four segments, a, b, c, and d, based on the size of P. belong , belong , belong , belong The range thresholds are calibrated based on platform operation data and industry characteristics.

[0016] Preferably, the specific implementation steps of S2 are as follows: S21: Create two empty sets, a patchable queue set and an unpatched queue set, to store program waiting queues that can be patched and those that cannot be patched, respectively, and clearly define the classification and attribution targets; S22: Traverse the waiting queues of all media programs i within the video platform, and extract the individual parameters of each program, including the regular stream parameters. Existence state, regular flow establishment time Combine the current system time t with the patchable window W to perform dual technical status verification and classification judgment; S23: Repeat the traversal, verification, and classification judgment of S22 until all programs in the waiting queue of the video platform have completed the affiliation judgment and have been assigned to the corresponding set without omission; output the final classification result.

[0017] Preferably, the dual technical status verification and classification determination includes: Dual technical status verification includes: Technical Status Verification 1: This involves checking whether the real-time transport stream identifier, transport status record, and data buffer corresponding to the program are present in the server to determine the regular stream of program i. Does it exist? Technical State 2 Verification: Determine whether the current time t is within the patchable window, i.e., whether the constraint conditions are met.

[0018] in, This represents the duration of the regular stream. If this duration does not exceed the patchable window W, then the basic requirement for patch stream sharing is met. The classification and determination process is as follows: If both of the above checks pass, add program i's waiting queue to the patchable queue set; If any check fails, add program i's waiting queue to the unpatched queue set.

[0019] Preferably, the specific implementation process of S3 is as follows: S31: Call the patchable queue set and the non-patternable queue set output by S2 to confirm the queue list of the two sets and the program individual parameters of the corresponding program i; Define scheduling priority rules: the first priority is the set of patchable queues, the second priority is the set of unpatched queues, and the next priority is triggered only when there are no available queues in the first priority. S32: When the set of patchable queues is not empty, calculate the priority of each queue in the set and assign a patch stream of the corresponding length to it; S33: When the set of patchable queues is empty and the set of unpatched queues is not empty, the same queue priority calculation and filtering rules as in S32 are used to select the optimal queue and establish a new regular flow for it. At the same time, the establishment time of the new regular flow is updated to the current system time.

[0020] S34: If both the patchable queue set and the unpatched queue set are empty, it indicates that there are currently no user requests to be processed. The scheduling result, including the scheduling queue identifier, queue priority, and execution operation type, will be synchronized to the system log.

[0021] Preferably, the specific implementation steps of S32 are as follows: S321, Decision condition: If the set of patchable queues is not empty; S322, Priority Calculation: For each queue in the set, based on the parameters defined in S1, calculate the queue priority score of queue i one by one according to the comprehensive priority formula:

[0022] in, Indicates the priority of business value. The total number of user requests for patched or regular streams of i, obtained through real-time statistics from the video platform, represents the number of users that the service queue can satisfy. This is used to reflect the efficiency and benefits of this scheduling operation. S323, Optimal queue selection: For all The values ​​are sorted in descending order, and the queue at the top of the sorted list is selected as the scheduling target; if there are multiple For queues with equal values, priority is given to filtering. Larger queues If they are still equal, prioritize screening. Larger queues; S324, Execute scheduling: Allocate a patch stream to the selected optimal queue, with a patch stream length of [length missing]. After receiving the patch stream, the user terminal buffers it and automatically and seamlessly switches to the corresponding regular stream after playback is complete.

[0023] Preferably, the specific implementation steps of S33 are as follows: S331, Decision condition: If the set of patchable queues is empty and the set of unpatchable queues is not empty; S332, Priority Calculation: The priority formula in S32 is used to calculate the priority of each queue within the set of unpatched queues. value:

[0024] S333, Optimal Queue Selection: Select according to the sorting and selection rules in S32. The queue with the highest value is selected as the scheduling target; S334, Execute Scheduling: Establish a new regular flow for the selected optimal queue. and the establishment time of this regular stream Updated to the current system time t, synchronized to S1, from From now on, new waiting users can join the regular stream via the patch stream within the patchable window W.

[0025] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0026] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0027] As can be seen from the above technical solution compared with the prior art, the present invention has the following beneficial effects: 1. This invention introduces dynamically configured business value priorities, enabling the scheduling system to precisely align with business strategies, prioritizing the service quality of core businesses, and directly improving user satisfaction and platform brand value. At the same time, by leveraging the cost-aware division of patchable / non-patchable sets, the system is forced to prioritize the discovery of low-cost patch service opportunities, significantly reducing server bandwidth output and concurrent streams, saving hardware costs and extending the system expansion cycle, which is more in line with actual operational decision-making logic. 2. The algorithm of this invention has clear logic, a simple hierarchical structure, and low computational complexity, involving only basic comparison and multiplication operations. It has low requirements for system real-time performance and can be seamlessly integrated into the scheduling modules of existing streaming media servers such as Nginx-rtmp, SRS, and Wowza without modifying the underlying hardware or transmission protocol. At the same time, the priority definition can be flexibly expanded, easily incorporating new weighting factors such as "content exclusive to paid users" and "advertising load value," effectively adapting to more complex business models in the future, and possessing strong feasibility and adaptability. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the method steps of the present invention; Figure 2This is a schematic diagram of the determination process in step two of the embodiments of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0030] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention, but should not be used to limit the scope of the present invention.

[0031] This invention provides a video patch stream scheduling method based on multi-dimensional priority, such as... Figure 1 As shown, perform the following steps using a computer device: S1 defines parameters, including the current time and patchable window, as well as program-specific parameters. Program-specific parameters include the regular stream, regular stream establishment time, dynamic service value priority, and waiting queue length. S2, divide the patch set. Based on the matching relationship between the regular stream, the current time and the patchable window, divide all program waiting queues into a patchable set and an unpatched set. S3, priority calculation and allocation: Based on the classification results of S2, the comprehensive priority formula is calculated using the business value priority determined by S1 and the waiting queue length. Priority is given to allocating patch flows to the highest queue in the patchable set. If the set is empty, a new regular flow is established for the highest queue in the unpattable set.

[0032] Example: In this embodiment, a video platform faces high concurrency requests during prime time in the evening: there are live broadcasts of the World Cup qualifiers and key events, which attract a large number of users, as well as newly launched popular suspense dramas with concentrated user on-demand viewing, and on-demand requests for regular variety shows and classic old movies.

[0033] The platform server needs to schedule the patch stream using the MPQL method to ensure the smoothness of core business, live streaming, and popular dramas, while avoiding resource waste. Simultaneously, it is also necessary to address the special needs of some live streaming users who are VIP members. Therefore, this invention introduces a video patch stream scheduling method based on multi-dimensional priority, such as... Figure 2 As shown: Step 1: System initialization and multi-dimensional parameter definition, corresponding to technical status and business value dimensions; Define system global state parameters: Let the current time be t; The patchable window size is W, a preset system parameter that indicates the time range within which subsequent users are allowed to join the shared stream via patching after the regular stream is established. Its value is typically set based on a combination of factors including video bitrate, network latency, and server performance.

[0034] Define the individual state parameters for each program i: Normal flow: This refers to the first complete video stream established for the program.

[0035] Normal stream setup time: refer to The moment transmission begins.

[0036] Business value priority: It is a dynamically or statically assigned weighting coefficient used to quantify the business importance of the program content. Its assignment rules follow explicit business logic, for example: It dynamically calculates priorities based on the following three network parameters. : z represents packet loss rate, y represents latency rate, and j represents jitter rate;

[0037]

[0038] in, , and These represent the potential user base of program i under scenarios with packet loss rate z, latency rate y, and jitter rate j, respectively. The program was then divided into four segments based on the size of P. belong , belong , belong , belong .

[0039] Parameter 'a' indicates that the program is either live or has just ended, such as popular events within the patchable window W. This type of content is highly time-sensitive, has the highest density of concurrent user requests, and therefore has the greatest negative impact from service failures.

[0040] Parameter 'b' indicates the likelihood that the program belongs to high-concurrency on-demand content, such as newly released blockbusters or the latest episodes of popular dramas. Such content generates concentrated user requests and is key to attracting and retaining users.

[0041] Parameter 'c' indicates that the program belongs to the category of regular on-demand content, such as movies or regular TV series that have been available for some time. This type of content is part of the platform's basic library, and user requests for it are stable.

[0042] The parameter 'd' indicates the likelihood that the program belongs to archived content, such as classic old films, educational videos, or outdated content. This type of content has low access volume and relatively minimal business value.

[0043] Let the length of the waiting queue be... This refers to the number of user requests currently waiting to receive the patch stream or regular stream of program i. This is a key metric for measuring resource efficiency.

[0044] Step 2: Queue Classification Based on Technical Status and Resource Efficiency This step is a preprocessing stage for efficient scheduling, and its goal is to divide the waiting queue into two sets with distinctly different resource consumption characteristics based on technical feasibility.

[0045] 1. Initialize two empty sets: a set of patchable queues and a set of unpatchable queues.

[0046] 2. For each media program i in the system's waiting queue, perform the following judgment: Technical status check, check the regular stream corresponding to this program. Does it exist? That is, are there any users watching? And is the current time t within a patchable window? That is, does it meet the conditions? .

[0047] Classification and resource efficiency prediction: If the above conditions are met, then the waiting queue of program i is added to the set of patchable queues. This decision is based on profound considerations of resource efficiency. Because such queues provide services only by consuming a short-lived queue of length (t - ...). This is a patch stream resource. It is a "resource-efficient" operation that can serve a large number of users at a very low cost.

[0048] Otherwise, that is, the normal flow does not exist or Then the waiting queue of program i will be added to the set of unpatched queues. Serving these queues means that the system needs to build a completely new and complete regular stream for them, which consumes resources proportional to the total length of the video and is costly, belonging to the category of "low resource efficiency" operation.

[0049] Repeat steps 1 and 2 of step two until all program waiting queues have been processed and added to their corresponding sets.

[0050] Step 3: Prioritizing scheduling that integrates business value and resource efficiency This step is the core decision-making process of this invention, which cleverly integrates business value and resource efficiency within a hierarchical structure.

[0051] First priority: High-value queues within the "patchable" service set. Check: If the patchable queue set is not empty.

[0052] Decision: Instead of directly calling traditional MQL methods, a weighted queue length-first strategy is adopted. The overall priority score for each queue i in the patchable queue set is calculated. .

[0053] Calculation formula:

[0054] in, Representing the business value dimension, it acts as a multiplier to ensure that the queue of high-value content is amplified in the calculation; a high... It can make up for its The temporary shortage. This indicates the number of users that the queue can serve, directly reflecting the efficiency gains of this scheduling operation.

[0055] Execution: Select from the set of patchable queues The queue with the largest value.

[0056] Return the queue and assign a patch stream to it.

[0057] This invention ensures that the system always prioritizes serving the queues with the highest business value and the most users within the "resource-efficient zone," which is the key to achieving global optimization.

[0058] Second priority: High-value queues within the set of services that cannot be patched; Judgment: If the patchable queue set is empty, it indicates that there are currently no low-cost service opportunities, but the unpatchable queue set is not empty.

[0059] Decision: The same weighted queue length priority strategy is adopted. Calculate the overall priority score for each queue i in the unpatched queue set:

[0060] Execution: Select from the unpatched queue set Find the queue with the largest value, return that queue, and create a new regular stream for it.

[0061] This step ensures that when the system has to perform high-cost operations, resources are still directed to the queues with the highest business value and the most urgent needs, thus avoiding waste of resources.

[0062] Idle state: If both sets are empty, it indicates that there are currently no waiting requests, and an empty value is returned.

[0063] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0064] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0065] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the video patch stream scheduling methods based on multi-dimensional priority in the above embodiments.

[0066] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0067] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0068] For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media.

[0069] The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0070] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0071] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0072] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0073] The embodiments of the present invention are given for the purposes of illustration and description. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A video patch stream scheduling method based on multi-dimensional priority, characterized in that, Perform the following steps using a computer device: S1 defines parameters, including the current time and patchable window, as well as program-specific parameters. Program-specific parameters include the regular stream, regular stream establishment time, dynamic service value priority, and waiting queue length. S2, divide the patch set. Based on the matching relationship between the regular stream, the current time and the patchable window, divide all program waiting queues into a patchable set and an unpatched set. S3, priority calculation and allocation: Based on the classification results of S2, the comprehensive priority formula is calculated using the business value priority determined by S1 and the waiting queue length. Priority is given to allocating patch flows to the highest queue in the patchable set. If the set is empty, a new regular flow is established for the highest queue in the unpattable set.

2. The video patch stream scheduling method based on multi-dimensional priority as described in claim 1, characterized in that: The specific implementation steps of S1 are as follows: S11 defines global system parameters within the video platform, sets the current time parameter t, and synchronizes the clock in real time. Set the patchable window size parameter W to define the time range during which subsequent users are allowed to join the shared stream via patching after the regular stream is established; S12 defines individual program parameters and defines the regular stream as... This represents the first complete video stream established for program i, defining the regular stream establishment time parameter. Representing records The moment of transmission begins is a critical time node for calculating the length of the patch stream and determining whether it is within the patchable window. S13, Dynamically calculate business value priority parameters Quantify the importance of the program; S14, Set the waiting queue length parameter This counts the number of user requests currently waiting to receive program i-patch streams or regular streams.

3. The video patch stream scheduling method based on multi-dimensional priority as described in claim 2, characterized in that: The specific algorithm flow for S13 is as follows: Three core network parameters are selected: packet loss rate z, latency rate y, and jitter rate j. The constraints are satisfied: Define the potential user count parameter , and These represent the potential user base of program i under scenarios with packet loss rate z, latency rate y, and jitter rate j, respectively. Calculate the service value priority P of program i to achieve quantification of the linkage between network status and user scale: The program was then divided into four segments, a, b, c, and d, based on the size of P. belong , belong , belong , belong The range thresholds are calibrated based on platform operation data and industry characteristics.

4. The video patch stream scheduling method based on multi-dimensional priority as described in claim 3, characterized in that: The specific implementation steps of S2 are as follows: S21: Create two empty sets, a patchable queue set and an unpatched queue set, to store program waiting queues that can be patched and those that cannot be patched, respectively, and clearly define the classification and attribution targets; S22: Traverse the waiting queues of all media programs i within the video platform, and extract the individual parameters of each program, including the regular stream parameters. Existence state, regular flow establishment time Combine the current system time t with the patchable window W to perform dual technical status verification and classification judgment; S23: Repeat the traversal, verification, and classification judgment of S22 until all programs in the waiting queue of the video platform have completed the affiliation judgment and have been assigned to the corresponding set without omission; output the final classification result.

5. The video patch stream scheduling method based on multi-dimensional priority as described in claim 4, characterized in that: The execution of dual-technology status verification and classification determination includes: The dual-technology status verification includes: Technical Status Verification 1: This involves checking whether the real-time transport stream identifier, transport status record, and data buffer corresponding to the program are present in the server to determine the regular stream of program i. Does it exist? Technical State 2 Verification: Determine whether the current time t is within the patchable window, i.e., whether the constraint conditions are met. in, This represents the duration of the regular stream. If this duration does not exceed the patchable window W, then the basic requirement for patch stream sharing is met. The classification determination process is as follows: If both of the above checks pass, add program i's waiting queue to the patchable queue set; If any check fails, add program i's waiting queue to the unpatched queue set.

6. The video patch stream scheduling method based on multi-dimensional priority as described in claim 5, characterized in that: The specific implementation process of S3 is as follows: S31: Call the patchable queue set and the non-patternable queue set output by S2 to confirm the queue list of the two sets and the program individual parameters of the corresponding program i; Define scheduling priority rules: the first priority is the set of patchable queues, the second priority is the set of unpatched queues, and the next priority is triggered only when there are no available queues in the first priority. S32: When the set of patchable queues is not empty, calculate the priority of each queue in the set and assign a patch stream of the corresponding length to it; S33: When the set of patchable queues is empty and the set of unpatched queues is not empty, the same queue priority calculation and filtering rules as in S32 are used to select the optimal queue and establish a new regular flow for it. At the same time, the establishment time of the new regular flow is updated to the current system time. S34: If both the patchable queue set and the unpatched queue set are empty, it indicates that there are currently no user requests to be processed. The scheduling result, including the scheduling queue identifier, queue priority, and execution operation type, will be synchronized to the system log.

7. The video patch stream scheduling method based on multi-dimensional priority as described in claim 6, characterized in that: The specific implementation steps of S32 are as follows: S321, Decision condition: If the set of patchable queues is not empty; S322, Priority Calculation: For each queue in the set, based on the parameters defined in S1, calculate the queue priority score of queue i one by one according to the comprehensive priority formula: in, Indicates the priority of business value. The total number of user requests for patched or regular streams of i, obtained through real-time statistics from the video platform, represents the number of users that the service queue can satisfy. This is used to reflect the efficiency and benefits of this scheduling operation. S323, Optimal queue selection: For all The values ​​are sorted in descending order, and the queue at the top of the sorted list is selected as the scheduling target; if there are multiple For queues with equal values, priority is given to filtering. Larger queues If they are still equal, prioritize screening. Larger queues; S324, Execute scheduling: Allocate a patch stream to the selected optimal queue, with a patch stream length of [length missing]. After receiving the patch stream, the user terminal buffers it and automatically and seamlessly switches to the corresponding regular stream after playback is complete.

8. The video patch stream scheduling method based on multi-dimensional priority as described in claim 7, characterized in that: The specific implementation steps of S33 are as follows: S331, Decision condition: If the set of patchable queues is empty and the set of unpatchable queues is not empty; S332, Priority Calculation: The priority formula in S32 is used to calculate the priority of each queue within the set of unpatched queues. value: S333, Optimal Queue Selection: Select according to the sorting and selection rules in S32. The queue with the highest value is selected as the scheduling target; S334, Execute Scheduling: Establish a new regular flow for the selected optimal queue. and the establishment time of this regular stream Updated to the current system time t, synchronized to S1, from From now on, new waiting users can join the regular stream via the patch stream within the patchable window W.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it causes the processor to perform the steps of the method as described in any one of claims 1 to 8.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method as described in any one of claims 1 to 8.

Images