WO2024060860A1

WO2024060860A1 - Computing power network-based video transcoding method, apparatus and device and storage medium

Info

Publication number: WO2024060860A1
Application number: PCT/CN2023/111561
Authority: WO
Inventors: 王艳辉
Original assignee: 中移(苏州)软件技术有限公司; 中国移动通信集团有限公司
Priority date: 2022-09-22
Filing date: 2023-08-07
Publication date: 2024-03-28
Also published as: CN116781917A

Abstract

Provided in the embodiments of the present disclosure are a computing power network-based video transcoding method, apparatus and device and a storage medium. The method comprises: determining a computing power resource corresponding to a video to be transcoded; in at least one node meeting the computing power resource, determining a target node on the basis of a priority value corresponding to each node of the at least one node; and by means of the target node, transcoding said video, so as to obtain a transcoded video. Compared with a cloud-computing-based transcoding mode, a computing power network-based transcoding mode in the technical solutions of the embodiments of the present invention determines a target node from at least one node, which meets a computing power resource, in cloud layers, edge layers or terminal layers of a computing power network; therefore, transcoding, by means of the target node, a video to be transcoded can reduce time delays; and transcoding, by means of the target node, a video to be transcoded can also utilize resources to the maximum extent, thus increasing the utilization rate of resources.

Description

Video transcoding methods, devices, equipment and storage media based on computing power network

Cross-references to related applications

This disclosure is based on a Chinese patent application with application number 202211157805.8 and a filing date of September 22, 2022, and claims the priority of the Chinese patent application. The entire content of the Chinese patent application is hereby incorporated into this application by introduction.

Technical field

The present disclosure relates to the field of mobile communication technology, and specifically to a video transcoding method, device, equipment and storage medium based on a computing power network.

Background technique

In the current related technologies, the mainstream video transcoding system is a centralized transcoding system based on cloud computing. The technical solutions in the related technologies have the following problems: On the one hand, the efficiency of online transcoding is low: Since online transcoding is deployed in the cloud center, all videos that need to be transcoded need to be transmitted to the cloud center for online transcoding calculation, and then pushed to the client side. The computing center of the cloud layer is generally far away from the user, and there will be a relatively large delay. The requirements for latency of live video broadcast are very high, so it will cause high latency and poor user experience. On the other hand, the resource utilization rate of offline transcoding is low: offline transcoding has high requirements for computing resources, but the methods in the related technologies are limited to the utilization of computing resources in offline transcoding clusters, and resources cannot be fully utilized. There is currently no effective solution to this problem.

Contents of the invention

In view of this, the main purpose of the present disclosure is to provide a video transcoding method, device, equipment and storage medium based on a computing power network.

In order to achieve the above objectives, the technical solution of the present disclosure is implemented as follows:

Embodiments of the present disclosure provide a video transcoding method based on computing power network. The method includes:

Determine the computing resources corresponding to the video to be transcoded;

In at least one node that meets the computing power resources, determine the target node based on the priority value corresponding to each node in the at least one node;

The target node is used to transcode the video to be transcoded to obtain a transcoded video.

In the above solution, the transcoding includes online transcoding, and the method further includes:

Determine at least one first node in the computing power network; wherein the first node represents any node that meets the computing power resources in the edge layer of the computing power network;

Determine a first target node based on a first priority value corresponding to each node in the at least one first node;

The first target node is used to perform online transcoding on the video to be transcoded to obtain the first video after online transcoding.

In the above solution, determining the first target node based on the first priority value corresponding to each node in the at least one first node includes:

Obtain a first parameter corresponding to each node in the at least one first node; the first parameter represents an attribute parameter of the first node related to the fluency of the video to be transcoded;

Determine a second parameter corresponding to the first parameter based on a preset model; the second parameter represents the influence parameter of the first parameter on the fluency of the video to be transcoded;

Based on the first parameter and the second parameter, determine a first priority value corresponding to each node in the at least one first node;

The first node corresponding to the smallest first priority value is used as the first target node.

In the above solution, the method further includes:

constructing a training sample set based on first sample parameters and second sample parameters associated with each of the at least one first node;

Using the first sample parameters included in the training sample set as input and the second sample parameters included in the training sample set as output, the first model is trained to obtain a preset model.

In the above solution, after determining the first target node based on the first priority value corresponding to each node in the at least one first node, the method further includes:

Determine the first current priority value of the first target node;

Determine the second current priority value corresponding to each node in the at least one first node;

Determine the smallest second current priority value among the second current priority values;

When the first current priority value and the minimum second current priority value meet preset conditions, the first node corresponding to the minimum second current priority value is updated as the first target node.

In the above solution, after the first target node is used to perform online transcoding on the video to be transcoded, and the first video after online transcoding is obtained, the method further includes:

When the first target node is closest to the user, the first video is played through the first target node.

In the above solution, the transcoding includes offline transcoding, and the method further includes:

Determine at least one second node in the computing power network; wherein the second node represents any node that meets the computing power resources in the terminal layer, edge layer or cloud layer of the computing power network;

Determine the second target node based on the second priority value corresponding to each node in the at least one second node;

The second target node is used to perform offline transcoding on the video to be transcoded to obtain the second video after offline transcoding.

In the above solution, determining the second target node based on the second priority value corresponding to each node in the at least one second node includes:

Obtain a third parameter corresponding to each node in the at least one second node; the third parameter represents an attribute parameter of the second node related to the fluency of the video to be transcoded;

Determine a fourth parameter corresponding to the third parameter; the fourth parameter represents the influence parameter of the third parameter on the fluency of the video to be transcoded;

Based on the third parameter and the fourth parameter, determine a second priority value corresponding to each node in the at least one second node;

The second node corresponding to the largest second priority value is used as the second target node.

In the above solution, the second target node is used to perform offline transcoding on the video to be transcoded, and the second video after offline transcoding is obtained, including:

When the second target node is in the cloud layer or edge layer of the computing power network, determine a first scheduling method for offline transcoding of the video to be transcoded;

Based on the first scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.

When the second target node is in the terminal layer of the computing power network, determine a second scheduling method for offline transcoding of the video to be transcoded;

Based on the second scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.

In the above solution, the method further includes:

Obtain the first computing power scheduled by the terminal layer of the computing power network;

Obtain the second computing power to transcode the video to be transcoded;

Based on the first computing power and the second computing power, a revenue parameter of the terminal layer is determined.

An embodiment of the present disclosure provides a video transcoding device based on a computing power network. The device includes:

The first determination module is configured to determine the computing resources corresponding to the video to be transcoded;

The second determination module is configured to determine the target node based on the priority value corresponding to each node in the at least one node among at least one node that meets the computing power resource;

A transcoding module configured to use the target node to transcode the video to be transcoded to obtain a transcoded video.

An embodiment of the present disclosure provides a video transcoding device based on a computing power network, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the program, the above-mentioned any of the methods described.

Embodiments of the present disclosure provide a storage medium that stores executable instructions. When the executable instructions are executed by a processor, any one of the methods described above is implemented.

The present disclosure provides a method, apparatus, device and storage medium for video transcoding based on a computing network, wherein the method comprises: determining computing resources corresponding to the video to be transcoded; In at least one node of the computing power resources, a target node is determined based on the priority value corresponding to each node in the at least one node; the target node is used to transcode the video to be transcoded to obtain the transcoded video. Compared with the transcoding method based on cloud computing, the technical solution of the embodiment of the present invention is adopted, and the transcoding method based on the computing power network is adopted. The target node is determined in at least one node that meets the computing power resources in the cloud layer, edge layer or terminal layer of the computing power network. The target node is used to transcode the video to be transcoded, which can overcome the problem that the computing center of the cloud layer is generally far away from the user, which not only reduces the delay of video playback, but also maximizes the use of resources and improves resource utilization.

Description of drawings

FIG1 is a schematic diagram of a transcoding system architecture in a related art of a video transcoding method based on a computing network according to an embodiment of the present disclosure;

Figure 2 is a schematic flow chart of a video transcoding method based on a computing power network according to an embodiment of the present disclosure;

Figure 3 is a schematic diagram of the transcoding system architecture of the video transcoding method based on the computing power network according to an embodiment of the present disclosure;

Figure 4 is a schematic flowchart of the online transcoding implementation process of the video transcoding method based on the computing power network according to the embodiment of the present disclosure;

Figure 5 is a schematic flowchart of the offline transcoding implementation process of the video transcoding method based on the computing power network according to the embodiment of the present disclosure;

Figure 6 is a schematic structural diagram of a video transcoding device based on a computing power network according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram of the hardware structure of a video transcoding device based on a computing power network according to an embodiment of the present disclosure.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure more clear, the specific technical solutions disclosed will be further described in detail below in conjunction with the drawings in the embodiments of the present disclosure. The following examples serve to illustrate the disclosure but are not intended to limit the scope of the disclosure.

Figure 1 is a schematic diagram of the transcoding system architecture in the related art of a video transcoding method based on computing power network according to an embodiment of the present disclosure. As shown in Figure 1, the technical architecture of a typical transcoding system in the related art includes: Source terminal, management node, cloud storage, offline transcoding module and online transcoding module. Among them, the source end includes various types of terminals, including personal computers (Personal Computer, PC), mobile phones, cameras, etc. The source end is responsible for collecting video data and uploading it to the cloud platform. The management node is responsible for managing system data, processing business logic, and providing application program (Application Programming Interface, API) interfaces, etc. Cloud storage generally uses object storage. Object storage generally uses central storage. The user first manually selects the video file and transfers it to his own object storage. Then select the files to transcode. When transcoding, the transcoding cluster first downloads the video file from the cloud storage center, and then stores it back to the object storage after the transcoding is completed. Offline transcoding refers to converting one video file into another or multiple video files with different bitrates to adapt to different network bandwidths, different terminal processing capabilities, and different user needs. Online transcoding refers to converting live streams into multiple live streams with different resolutions or bit rates, so that they can be distributed to different live broadcast platforms.

In the transcoding method in the related technology, after the video transcoding is completed, the video continues to be placed in the cloud storage center. Just like uploading, when users download and play, they also face the problem of being far away from the video service address and low playback and download efficiency.

In view of the deficiencies of the above-mentioned related technologies, the embodiments of the present disclosure provide a video transcoding method, device, equipment and storage medium based on a computing power network, wherein the computing power network system includes three levels: cloud layer, edge layer and terminal layer. The terminal layer is responsible for collecting videos and pushing the collected videos to the edge side in real time. By performing online transcoding at the edge layer, the real-time video is pushed to the client (playing end) through the nearest edge node through edge-to-edge collaboration to ensure the real-time transmission. For offline transcoding with low real-time requirements, the three-level computing power of the cloud layer, edge layer and terminal layer can be coordinated uniformly to select the most suitable node for transcoding. The transcoding result is transmitted back to the cloud storage center for storage. When playing, the video of the cloud storage center is used as the source station, and each edge node is used as an acceleration node to improve the smoothness of video playback. In the entire system, the terminal layer consumes and contributes computing power, and the system management module provides a computing power-based billing method.

The embodiment of the present disclosure proposes a video transcoding method based on computing power network. The functions implemented by this method can be realized by calling the program code by the processor in the video transcoding device based on computing power network. Of course, the program code can be saved in In computer storage media, it can be seen that the computing device at least includes a processor and a storage medium.

Figure 2 is a schematic flow chart of a video transcoding method based on computing power network according to an embodiment of the present disclosure, as shown in Figure As shown in 2, the method includes:

Step 201: Determine the computing resources corresponding to the video to be transcoded;

Step 202: Among at least one node that meets the computing power resources, determine the target node based on the priority value corresponding to each node in the at least one node;

Step 203: Use the target node to transcode the video to be transcoded to obtain a transcoded video.

In step 201, the application scenario of the video transcoding method based on the computing power network can be determined according to the actual situation, and is not limited here. As an example, the video transcoding method based on computing power network can be applied to a computing power network service system oriented to video transcoding. The computing power network service system is a video transcoding method oriented to computing power network. system.

The video to be transcoded can be determined according to actual conditions and is not limited here. As an example, the video to be transcoded may be a live stream that needs to be transcoded online, or a video file that needs to be transcoded offline.

The determination process of the computing power resources corresponding to the video to be transcoded can be determined according to the actual situation, and is not limited here. As an example, the computing resources corresponding to the video to be transcoded can be determined according to the transcoding template and the parameters of the video to be transcoded. The parameters of the video to be transcoded can be determined according to actual conditions and are not limited here. As an example, the parameters of the video to be transcoded may be any one or more of video format, resolution, bit rate and other parameters. The transcoding template can be determined according to actual conditions and is not limited here. As an example, the transcoding template may be a template associated with the video to be transcoded.

In step 202, the at least one node may be determined according to actual conditions, and is not limited here. As an example, the at least one node may be a node in the cloud layer, edge layer or terminal layer of the computing power network that meets the computing power resources. The number of the at least one node can be determined according to actual conditions and is not limited here.

The priority value can be determined according to actual conditions and is not limited here. As an example, the priority value may be data characterizing the fluency relationship between each node in the at least one node and the video to be transcoded. In at least one node that meets the computing power resources, based on the at least one Determining the target node based on the priority value corresponding to each node in the nodes may be to determine the target node to be transcoded based on the priority value corresponding to each node in the at least one node in at least one node that meets the computing power resources. The optimal fluency of the video; use the node corresponding to the optimal fluency as the target node.

In step 203, the target node can be determined according to the actual situation, and is not limited here. As an example, the target node may be a node determined based on the priority value corresponding to each node in the at least one node in the cloud layer, edge layer or terminal layer of the computing power network.

The transcoding can be determined according to actual conditions and is not limited here. As an example, the transcoding may include online transcoding and offline transcoding. The transcoded video can be determined according to actual conditions and is not limited here. As an example, the transcoded video may be a live stream and a video file. Specifically, the method of using the target node to transcode the video to be transcoded to obtain the transcoded video may be: using the target node to convert the live stream to obtain a plurality of different resolutions. rate or code rate of the live stream; it may also be that the target node is used to convert the one video file to obtain another or multiple video files with different code rates.

Embodiments of the present disclosure provide a video transcoding method based on a computing power network. The method includes: determining the computing power resources corresponding to the video to be transcoded; in at least one node that meets the computing power resources, based on the at least The priority value corresponding to each node in a node determines the target node; the target node is used to transcode the video to be transcoded to obtain the transcoded video. Compared with the transcoding method based on cloud computing, the technical solution of the embodiment of the present disclosure is adopted, and the transcoding method based on the computing power network satisfies the requirements of the computing power resources in the cloud layer, edge layer or terminal layer of the computing power network. Determining a target node in at least one node, and using the target node to transcode the video to be transcoded can overcome the problem that the computing center of the cloud layer is generally far away from the user, not only reducing the delay of video playback, but also achieving Maximize the use of resources and improve resource utilization.

In an optional embodiment of the present disclosure, the transcoding includes online transcoding, and the method further includes:

Determine the first target node based on the first priority value corresponding to each node in the at least one first node;

In this embodiment, the determination of at least one first node in the computing power network can be determined according to actual conditions, and is not limited here. As an example, the at least one first node may be at least one first node that satisfies the computing power resource determined in the edge layer of the computing power network. The at least one first node may be determined in the edge layer of the computing power network. The edge node that satisfies the computing power resources is used as at least one first node in the edge layer of the computing power network. Different from the cloud computing-centered online transcoding process in related technologies, the online transcoding based on the computing power network in this embodiment mainly relies on the edge nodes of the terminal layer and edge layer.

The first priority value can be determined according to actual conditions and is not limited here. As an example, the first priority value may be data representing the fluency relationship between each node in at least one first node of the edge layer of the computing power network and the video to be transcoded. Determining the first target node based on the first priority value corresponding to each node in the at least one first node may be: determining the first target node based on the first priority value corresponding to each node in the at least one first node. Describe the first optimal fluency of the video to be transcoded; use the first node corresponding to the first optimal fluency as the first target node.

The first video can be determined according to actual conditions and is not limited here. As an example, in the case of online transcoding of the video to be transcoded, the first video may be a live stream that converts the video to be transcoded into multiple different resolutions or code rates. .

The method of using the first target node to transcode the video to be transcoded online to obtain the first video after online transcoding may be to use the first target node in the edge layer of the computing power network to transcode the video to be transcoded online. The video to be transcoded is transcoded online to obtain the first video after online transcoding.

In an optional embodiment of the present disclosure, determining the first target node based on the first priority value corresponding to each node in the at least one first node includes:

Determine a first priority value corresponding to each node in the at least one first node based on the first parameter and the second parameter;

In this embodiment, the first parameter can be determined according to actual conditions and is not limited here. As an example, the first parameter may be the distance between the terminal layer of the computing power network and each edge node in the edge layer of the computing power network, each edge node in the edge layer of the computing power network The performance and any one or more parameters of the load of each edge node in the edge layer of the computing power network.

The preset model can be determined according to actual conditions and is not limited here. As an example, the preset model may represent the corresponding relationship between the fluency of the video to be transcoded and the second parameter. Determining the second parameter corresponding to the first parameter based on the preset model may include: determining the second parameter corresponding to the second optimal fluency based on the preset model; The corresponding second parameter is used as the second parameter corresponding to the first parameter.

The second parameter can be determined according to actual conditions and is not limited here. As an example, the second parameter may be an influencing parameter of the distance between the terminal layer of the computing power network and each edge node in the edge layer of the computing power network, Any one or more of the influencing parameters on the performance of each edge node and the influencing parameters on the load of each edge node in the edge layer of the computing power network.

Determining the first priority value corresponding to each node in the at least one first node based on the first parameter and the second parameter may be: based on the first parameter and the second parameter The products are added together to obtain the first priority value corresponding to each node in the at least one first node.

In some embodiments, determining the first target node based on the first priority value corresponding to each node in the at least one first node can be: determining the fluency of the video to be transcoded based on the first priority value corresponding to each node in the at least one first node; taking the fluency of the video to be transcoded corresponding to the smallest first priority value as the first optimal fluency; and taking the first node corresponding to the first optimal fluency as the first target node.

In an optional embodiment of the present disclosure, the method further includes:

Based on a first sample parameter associated with each of the at least one first node and a second sample This parameter is used to construct a training sample set;

In this embodiment, the first sample parameter can be determined according to actual conditions and is not limited here. As an example, the first sample parameter may be the distance between the terminal layer of the computing power network and each edge node in the edge layer of the computing power network, the distance between each edge node in the edge layer of the computing power network, Any one or more parameters of the performance of edge nodes, the load of each edge node in the edge layer of the computing power network, and the user's experience value. The user's experience value may be a parameter that is positively correlated with the fluency of the video to be transcoded.

The second sample parameter can be determined according to actual conditions and is not limited here. As an example, the second sample parameter may be an influencing parameter of the distance between the terminal layer of the computing power network and each edge node in the edge layer of the computing power network, the edge layer of the computing power network Any one or more of the influencing parameters on the performance of each edge node in the computing power network and the influencing parameters on the load of each edge node in the edge layer of the computing power network.

In some embodiments, determining the second parameter corresponding to the first parameter based on the preset model may be to determine the second optimal smoothness based on the user's experience value in the first sample parameter of the preset model. degree; based on the distance between the terminal layer of the computing power network and each edge node in the edge layer of the computing power network in the first sample parameter of the preset model, the distance between the edge nodes in the edge layer of the computing power network Determine the second parameter corresponding to the second optimal fluency based on any one or more parameters of the performance of each edge node and the load of each edge node in the edge layer of the computing power network; The second parameter corresponding to the excellent fluency is used as the second parameter corresponding to the first parameter.

In an optional embodiment of the present disclosure, after determining the first target node based on the first priority value corresponding to each node in the at least one first node, the method further includes:

Determine the first current priority value of the first target node;

Determine a second current priority value corresponding to each node in the at least one first node;

When the first current priority value and the minimum second current priority value meet preset conditions Next, the first node corresponding to the smallest second current priority value is updated as the first target node.

In this embodiment, considering that the terminal layer of the computing power network may be constantly moving, the calculated first priority value is also constantly changing. Then it is necessary to determine the first current priority value of the first target node, and also need to determine the second current priority value corresponding to each node in the at least one first node in real time, and re-determine the smallest second current priority value. The corresponding first node.

In some embodiments, the method further includes: determining a switching parameter corresponding to the first current priority value and the minimum second current priority value; wherein the switching parameter represents a proportional parameter of the performance improvement of the switching node. .

The preset conditions can be determined according to actual conditions and are not limited here. As an example, the preset condition may be to determine whether the switching parameter is greater than a preset threshold. The preset threshold can be determined according to actual conditions and is not limited here.

In the case where the first current priority value and the minimum second current priority value meet a preset condition, the first node corresponding to the minimum second current priority value is updated as the first target. The node may be, when the switching parameter corresponding to the first current priority value and the minimum second current priority value is greater than a preset threshold, change the first value corresponding to the minimum second current priority value to The node is updated to the first target node.

In an optional embodiment of the present disclosure, after the first target node is used to perform online transcoding on the video to be transcoded, and the first video after online transcoding is obtained, the method further includes:

In this embodiment, playing the first video through the first target node closest to the user can reduce the distance between the terminal layer and the playback end, thereby reducing delay.

In some embodiments, if the first target node is far from the user, and if the first target node is not the node closest to the user, determine the first node in the edge layer of the computing power network that is closest to the user. Three target nodes; push the first video to the third target node; play the first video through the third target node.

In an optional embodiment of the present disclosure, the transcoding includes offline transcoding, and the method further includes:

In this embodiment, the determination of at least one second node in the computing power network can be determined according to actual conditions, and is not limited here. As an example, nodes that meet the computing power resources may be determined in the terminal layer, edge layer or cloud layer of the computing power network.

The second priority value can be determined according to actual conditions and is not limited here. As an example, the second priority value may be data representing the fluency relationship between each node in at least one node in the terminal layer, edge layer, or cloud layer of the computing power network and the video to be transcoded. Determining the second target node based on the second priority value corresponding to each node in the at least one second node may be: determining the second target node based on the second priority value corresponding to each node in the at least one second node. Describe the third optimal fluency of the video to be transcoded; use the second node corresponding to the third optimal fluency as the second target node.

The second video can be determined according to actual conditions and is not limited here. As an example, in the case of offline transcoding of the video to be transcoded, the second video may be a video file that converts the video to be transcoded into another or more different code rates.

The offline transcoding of the video to be transcoded by using the second target node to obtain the second video after offline transcoding can be performed by using the second target node in the terminal layer, edge layer or cloud layer of the computing power network to offline transcode the video to be transcoded to obtain the second video after offline transcoding.

In an optional embodiment of the present disclosure, determining the second target node based on the second priority value corresponding to each node in the at least one second node includes:

Determine a fourth parameter corresponding to the third parameter; the fourth parameter represents the third parameter pair Parameters that influence the smoothness of the video to be transcoded;

In this embodiment, the third parameter can be determined according to actual conditions and is not limited here. As an example, the third parameter can be any one or more parameters of the performance of each node in the terminal layer, edge layer or cloud layer of the computing network, the load of each node in the terminal layer, edge layer or cloud layer of the computing network, and the transportation cost of each node in the terminal layer, edge layer or cloud layer of the computing network.

The fourth parameter can be determined according to actual conditions and is not limited here. As an example, the fourth parameter may be a parameter that affects the performance of each node in the terminal layer, edge layer or cloud layer of the computing power network, and each node in the terminal layer, edge layer or cloud layer of the computing power network. Any one or more of the influencing parameters of the load and the influencing parameters of the transportation cost of each node in the terminal layer, edge layer or cloud layer of the computing power network.

Determining the second priority value corresponding to each node in the at least one second node based on the third parameter and the fourth parameter may be, based on the third parameter and the fourth parameter The products are added together to obtain the second priority value corresponding to each node in the at least one second node.

In some embodiments, determining the second target node based on the second priority value corresponding to each node in the at least one second node may be based on the second priority value corresponding to each node in the at least one second node. The second priority value determines the fluency of the video to be transcoded; the fluency of the video to be transcoded corresponding to the largest second priority value is regarded as the third optimal fluency; the third optimal The second node corresponding to the fluency serves as the second target node.

In an optional embodiment of the present disclosure, using the second target node to perform offline transcoding on the video to be transcoded to obtain the second video after offline transcoding includes:

Based on the first scheduling method, the video to be transcoded is detached through the second target node. Online transcoding to obtain the second video after offline transcoding.

In this embodiment, the first scheduling method can be determined according to actual conditions, and is not limited here. As an example, the first scheduling method may be to use k8s (Kubernetes) for cluster computing scheduling.

Based on the first scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding, which may be k8s based on the cloud layer or edge layer. Create a sub-scheduling unit (pod); perform offline transcoding on the video to be transcoded through the second target node to obtain the second video after offline transcoding.

In this embodiment, the second scheduling method can be determined according to actual conditions and is not limited here. As an example, the second scheduling method may be to use lightweight k3s for computing scheduling.

Based on the second scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding, which can be: a sub-scheduling unit is created based on the k3s of the terminal layer; the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.

Obtaining a first computing power for scheduling by a terminal layer of the computing power network;

Obtain the second computing power to transcode the video to be transcoded;

In this embodiment, the first computing power can be determined according to actual conditions and is not limited here. As an example, the first computing power may be the computing power contributed by scheduling of the computing power network by the terminal layer of the computing power network.

The second computing power can be determined according to actual conditions and is not limited here. As an example, the second computing power may be the computing power consumed by the computing power network for transcoding the video to be transcoded collected by the terminal layer of the computing power network.

Determining the revenue parameter of the terminal layer based on the first computing power and the second computing power may be: performing a difference based on the first computing power and the second computing power to obtain the terminal The income parameters of the layer.

In order to facilitate understanding of the embodiments of the present disclosure, the following description takes the video transcoding method based on the computing power network applied to a computing power network service system for video transcoding as an example.

Figure 3 is a schematic diagram of the transcoding system architecture of the video transcoding method based on the computing power network according to an embodiment of the present disclosure. As shown in Figure 3, the entire computing power network service system is divided into three layers, namely the terminal layer, the edge layer and the cloud layer. Among them, the terminal layer can be the source layer; the edge layer can be the edge computing layer; and the cloud layer can be the cloud computing center layer. These three levels of computing, storage, and network resources form a unified computing network with unified management, unified scheduling, and clear division of labor to complete the process of online and offline video transcoding in the most efficient manner.

Process 1: Online transcoding process.

Unlike cloud-centric online transcoding processes. Online transcoding based on computing power network mainly relies on edge nodes at the terminal layer and edge layer. Figure 4 is a schematic diagram of the online transcoding implementation process of the video transcoding method based on the computing power network according to the embodiment of the present disclosure. Figure 4 shows the process of video access and playback.

Online transcoding includes two stages: streaming and playback. The terminal layer selects the most appropriate edge node of the edge layer for streaming or uploading. After the video to be transcoded is online transcoded at the edge node, it is transcoded into a video format suitable for playback on various devices.

There are two core indicators that affect live broadcast user experience. First: delay, which is the time delay from the playback source to the playback end. Second: Stuttering refers to the frame lag during video playback, which makes people obviously feel "stuck". The statistics of the number of playback stutters per unit time is called the stuttering rate.

There are three main parameters that affect the above two core indicators. First: the distance between the terminal layer and each edge node of the edge layer. Generally speaking, the longer the distance, the greater the delay and the higher the possibility of lagging. Second: the performance of edge nodes in the edge layer. Generally speaking, the better the performance of the edge node, the smaller the delay, and the lower the possibility of lag. Third: The load of edge nodes in the edge layer. Generally speaking, edge nodes The higher the load on the point, the smaller the delay and the lower the possibility of lagging.

The first stage: the push stage or the upload stage. Before the push or upload stage, these three parameters are obtained from each edge node of the edge layer in turn.

Generally speaking, the farther the distance, the higher the transmission cost, and the worse the delay and smoothness. The higher the node performance, the stronger the transcoding capability, and the stronger the latency and fluency. The higher the load on the node, the weaker the transcoding capability, the weaker the latency and fluency.

Step 1: Calculate the user’s experience value E1. The calculation method of the user’s experience value E1 is as shown in Equation (1).
E1＝α*delay+β*lag (1)

In formula (1), E1 represents the user experience value; delay represents delay; lag represents the stuck frame rate; α and β represent the parameters of delay and stuck frame rate respectively, which can be determined by the user according to the actual situation. There is a corresponding relationship between the user's experience value E1 and the second optimal fluency. The second optimal fluency can be the user's playback experience level. The greater the user's experience value E1, the worse the user's playback experience level.

Step 2: Calculate the first priority value E2 of the edge node. The calculation method of the first priority value E2 of the edge node is as shown in Equation (2).
E2＝δ*d+ε*l+φ/p (2)

In formula (2), E2 represents the first priority value of the edge node; d represents the distance between the source end and the edge node in the edge layer; p represents the performance of the edge node; l represents the load of the edge node; δ, ε, and φ represent respectively The influence parameters of d, l and p on the first priority value. Among them, the first priority value E2 of the edge node has a corresponding relationship with the first optimal fluency. The first optimal fluency can be the priority level of the edge node. The larger the first priority value E2 of the edge node, the higher the priority of the edge node. The lower the priority.

Step 3: Tune the influencing parameters in E2.

When the first priority value E2 of the edge node is larger, the user's experience value E1 is also larger. Because the priority level of the edge node is poor, if the edge node is selected, the user's playback experience level will also be poor.

It is assumed that the first priority value E2 of the edge node and the user's experience value E1 satisfy a certain linear relationship as shown in formula (3):

E1＝λE2 (3)

Substitute to get Through a large amount of data, existing testing methods are used for learning and tuning, and the values of these parameters δ, ε, and φ are obtained. Among them, a large amount of data can be training data including E1, d, l, and p; the existing testing method can be a model training method.

Then before selecting the edge node, the source end can first obtain the values of d, p, and l and then obtain the first priority value E2 value of the edge node. The greater the first priority value E2 of the edge node, the worse the user's playback experience level will be. .

Step 4: Determine the source location and switch the edge node to be connected.

Considering that the source may be constantly moving, the calculated first priority value E2 of the edge node is also constantly changing. If the source end calculates that the current optimal edge node is no longer the previously selected edge node, it needs to switch. Because switching requires a relatively large cost. Therefore, it needs to be switched when certain conditions are met. Follow the following rules when making the final selection of nodes.

1) Calculate the first priority value of the current access edge node. The calculation method of the first priority value of the current access edge node is as shown in Equation (4),
E2 _o ＝δ*d _o +ε _o *l _o +φ/p _o (4)

2) Calculate the smallest first priority value among all edge nodes. The calculation method for the smallest first priority value among all edge nodes is as shown in Equation (5),
E2 _m =Min(δ*d+ε*l+φ/p) (5)

3) Determine the final selected edge node. When the terminal layer accesses for the first time, select the edge node corresponding to E2 _m . Otherwise, calculate the handover parameter θ. The calculation method of the handover parameter θ is as shown in Equation (6),
θ＝(E2 _o -E2 _m )/E2 _o (6)

In equation (6), θ represents the performance improvement ratio of switching nodes. Set the threshold. When θ is greater than the threshold, select the edge node corresponding to E2 _m , otherwise continue to use the edge node corresponding to E2 _o . The threshold can be determined by the user based on the actual situation.

The second stage: the playback stage. If the playback end is closest to the above-mentioned edge node used for transcoding, then the video of the edge node used for transcoding will be played directly. Among them, the playback end includes users and transcoding machines. If the playback end is far away from the edge node used for transcoding, through edge-to-edge collaboration, the edge node used for transcoding is used as the source station, the stream is pushed to the edge node closest to the user, and then the edge closest to the playback end is played. Node video. In this way, the distance between the source end and the playback end can be shortened to the greatest extent and the delay can be reduced to the greatest extent. Online videos that are completely transcoded at edge nodes will be cached and played on each edge node through edge-to-edge collaboration. When users play through the content delivery network (Content Delivery Network, CDN), they will choose the nearest edge node to access, maximizing the Reduce latency.

Process 2: Offline transcoding process.

Offline transcoding: Historical videos recorded for historical videos are transcoded into different formats through offline transcoding to facilitate playback by different clients (players). Offline transcoding does not have high requirements for real-time performance, but has high requirements for computing. Therefore, it is necessary to maximize the use of computing resources at the cloud layer, edge layer, and even terminal layer. The cloud layer and edge layer are relatively large in scale and have strong computing power. K8s can be used for cluster computing scheduling. The computing capabilities of various terminal layers are limited, so lightweight K3s is used for computing scheduling. k3s is a Kubernetes distribution open sourced by rancher. Compared with k8s, k3s has been greatly tailored and optimized. The binary program is less than 50MB, takes up less resources, and only requires 512MB of memory to run. Because of the above advantages of k3s, k3s can be used in web browsers (Edge), Internet of Things (IoT), corporate identification systems (Corporate Identity System, CIS), processors (Advanced RISC Machines, ARM) and other scenarios , which can coordinate and control the computing power of various terminal layers in a unified manner.

Therefore, within the computing power network, a two-level Kubernetes + one-level k3s collaborative scheduling pattern has been formed, that is, k8s at the cloud layer and edge layer does cluster computing scheduling, and the terminal layer uses lightweight k3s for computing scheduling. By mixing Kubernetes and k3s, resources can be more flexibly scheduled at the cloud and edge layers, and the computing power of the terminal layer can be incorporated into the entire network to maximize the use of resources within the entire network.

FIG5 is a schematic diagram of the offline transcoding implementation process of the video transcoding method based on the computing power network according to an embodiment of the present disclosure. FIG5 is a whole flowchart of offline transcoding.

After the terminal layer collects the video, the recorded historical video is uploaded to the object storage of the cloud layer through the edge node of the edge layer. Each video is associated with a corresponding transcoding template, where the transcoding template is specified by the terminal layer. The computing power network service system calculates the required computing power based on the transcoding template and video parameters. human resources, and then create an offline transcoding task.

Consider the following factors when making task selection:

First: Whether each node in the three-level network of cloud layer, edge layer, and terminal layer can meet the computing power requirements.

Second: Among the nodes that meet the computing power requirements, consider the performance of each node in the three-level network of cloud layer, edge layer, and terminal layer. Set the value of the performance of each node in the three-level network as p′, and its influencing factor as γ (which can be determined by the user according to the actual situation).

Third: Among the nodes that meet the computing power requirements, consider the load of each node in the three-level network of cloud layer, edge layer, and terminal layer. Set the value of the load of each node in the three-level network to l', and its influencing factor is η (which can be determined by the user according to actual conditions).

Fourth: Among the nodes that meet the computing power requirements, consider the transmission cost of each node in the three-level network from the terminal layer to the cloud layer, edge layer, and terminal layer. Set the value of the transmission cost of each node in the three-level network as d′, and set its influencing factor as λ (can be determined by the user according to the actual situation).

Finally, the second priority value V of each node in the three-level network is V=γp′+ηl′+λ/d′.

Among them, there is a corresponding relationship between the second priority value V value of each node in the three-level network and the third optimal fluency. The third optimal fluency can be the priority level of each node in the three-level network. Each node in the three-level network The larger the second priority value V of a node is, the higher the priority level of each node in the three-level network is. Select the node with the largest V among the nodes that meet the conditions as the optimal node.

After calculating the optimal node, the corresponding node in the computing power network is triggered for scheduling. If the optimal node is in the cloud layer, the pod is created by k8s in the cloud layer. If the optimal node is at the edge layer, the pod is created by k8s at the corresponding edge layer. If the optimal node is at the terminal layer, the pod is created by K3S at the terminal layer. After the task execution is completed, the corresponding pod will be destroyed.

When users on the playback end need to play historical videos, the video stored in the cloud layer is used as the source station, and the cache and transmission optimization of each edge node in the edge layer are used to accelerate the playback experience of users on each playback end.

Process 3: Management and transaction process of transcoding computing power.

It is different from the usual transcoding system based on the calculation method of video size and transcoding format. For computing power network The network service system supports billing based on computing power. Each terminal layer can both provide computing power and consume computing power. The terminal layer is registered in the computing power network, which means it becomes a service grid of the computing power network. The terminal layer itself provides computing power and consumes computing power. Videos collected at the terminal layer need to be transcoded online and transcoded offline and stored in object storage, which represents the consumption of computing power by the terminal layer. At the same time, the terminal layer, as a grid of K3s, can be scheduled by the entire computing power network, which means that the terminal layer can contribute computing power. The computing power network service system has a computing power management module. Every time the terminal layer is scheduled, it can be regarded as a contribution to the computing power and the contribution value will be included. At the same time, the online transcoding computing power and offline transcoding computing power consumed by the video collected at the terminal layer will also be recorded. The final revenue (expenditure) of each terminal is calculated as shown in Equation (7),
∑α ₁ ct+β ₁ mt-∑α ₂ ct+β ₂ mt-∑α ₃ ct+β ₃ mt (7)

In formula (7): c represents the number of cores consuming the CPU; m represents the memory consumed (unit g); t represents the time of use; α ₁ and β ₁ represent the parameters of the computing power contributed by the terminal; α ₂ and β ₂ represent Parameters that consume the computing power consumed in the online transcoding process of the collected videos; α ₃ and β ₃ represent parameters that consume the computing power consumed in the offline transcoding process of the collected videos.

The embodiment of the present disclosure provides an overall design solution for realizing online video transcoding and offline video transcoding based on computing power network. For online transcoding with high real-time requirements, in the upload stage, the calculation formula of the optimal edge node is introduced, and the existing test method is used for learning and optimization, the optimal edge node is selected, and the online process is completed at the edge node of the edge layer. Transcoding. In the playback stage, through edge-side collaboration, the content is pushed to the edge node closest to the customer to achieve real-time playback. For offline transcoding that has low real-time requirements but high computing power requirements, coordinate three-level resources at the cloud layer, edge layer, and terminal layer, schedule through k8s+k3s, specify pods through labels, and select the optimal node for execution. Scheduling. Based on the management and payment methods of computing power, the terminal layer is both a contributor and a consumer of computing power.

Compared with the online transcoding method based on cloud computing, the transcoding method based on computing power network can reduce the delay and improve the smoothness of playback. Compared with the offline transcoding method based on cloud computing, the offline transcoding method based on the computing power network can maximize the use of resources and improve resource utilization. Management and billing methods based on computing power can technically reduce transcoding costs to the greatest extent, improve task execution efficiency, and achieve a win-win situation for video transcoding service providers and buyers.

The embodiment of the present disclosure uses a three-level computing power network composed of cloud layer, edge layer, and terminal layer. For online transcoding, the efficiency of uploading and downloading can be improved through direct transcoding at the edge and edge-to-edge collaboration. For offline transcoding, computing resources within the entire system can be scheduled to the maximum extent. By introducing computing power-based billing, the connected terminal layer is both a consumer and a contributor of computing power, which can achieve a win-win situation for computing power providers and buyers.

An embodiment of the present disclosure provides a video transcoding device based on a computing power network. Figure 6 is a schematic structural diagram of a video transcoding device based on a computing power network according to an embodiment of the disclosure. As shown in Figure 6, the device 600 includes:

The first determination module 601 is configured to determine the computing resources corresponding to the video to be transcoded;

The second determination module 602 is configured to determine the target node based on the priority value corresponding to each node in the at least one node among at least one node that meets the computing power resources;

The transcoding module 603 is configured to use the target node to transcode the video to be transcoded to obtain a transcoded video.

In other embodiments, the transcoding includes online transcoding, and the second determination module 602 further includes: a first determination unit and a second determination unit; wherein,

The first determination unit is configured to determine at least one first node in the computing power network; wherein the first node represents any of the computing power resources in the edge layer of the computing power network. node;

The second determining unit is configured to determine the first target node based on the first priority value corresponding to each node in the at least one first node;

The transcoding module 603 is also configured to use the first target node to perform online transcoding on the video to be transcoded to obtain the first video after online transcoding.

In other embodiments, the second determining unit is further configured to obtain a first parameter corresponding to each node in the at least one first node; the first parameter represents the relationship between the first node and the Describe attribute parameters related to the smoothness of the video to be transcoded; determine a second parameter corresponding to the first parameter based on a preset model; the second parameter represents the smoothness of the first parameter to the video to be transcoded degree of influence parameters; based on the first parameter and the second parameter, determine the at least one first node The first priority value corresponding to each node; the first node corresponding to the smallest first priority value is used as the first target node.

In other embodiments, the device 600 further includes: a building module and a training module; wherein,

The building module is configured to construct a training sample set based on the first sample parameter and the second sample parameter related to each node in the at least one first node;

The training module is configured to use the first sample parameter included in the training sample set as input, and use the second sample parameter included in the training sample set as output to train the first model to obtain the preset Model.

In other embodiments, the second determining unit is further configured to determine the first target node based on the first priority value corresponding to each node in the at least one first node. The first current priority value of a target node; determining the second current priority value corresponding to each node in the at least one first node; determining the smallest second current priority value among the second current priority values; in the When the first current priority value and the minimum second current priority value meet the preset conditions, the first node corresponding to the minimum second current priority value is updated as the first target node.

In other embodiments, the transcoding module 603 is further configured to use the first target node to perform online transcoding on the video to be transcoded, and after obtaining the online transcoded first video, after the When the first target node is closest to the user, the first video is played through the first target node.

In other embodiments, the transcoding includes offline transcoding, and the second determination module 602 further includes: a third determination unit and a fourth determination unit; wherein,

The third determination unit is configured to determine at least one second node in the computing power network; wherein the second node represents a condition that satisfies the computing power requirements in the terminal layer, edge layer or cloud layer of the computing power network. Any node of human resources;

The fourth determination unit is configured to determine the second target node based on the second priority value corresponding to each node in the at least one second node;

The transcoding module 603 is also configured to use the second target node to perform offline transcoding on the video to be transcoded to obtain the second video after offline transcoding.

In other embodiments, the fourth determining unit is further configured to obtain a third parameter corresponding to each node in the at least one second node; the third parameter represents the relationship between the second node and the to-be-determined node. Attribute parameters related to the fluency of the transcoded video; determine a fourth parameter corresponding to the third parameter; the fourth parameter represents the influence parameter of the third parameter on the fluency of the video to be transcoded; based on The third parameter and the fourth parameter determine the second priority value corresponding to each node in the at least one second node; the second node corresponding to the largest second priority value is used as the second priority value. Two target nodes.

In other embodiments, the transcoding module 603 is further configured to determine to perform offline processing of the video to be transcoded when the second target node is in the cloud layer or edge layer of the computing power network. A first scheduling method for transcoding; based on the first scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.

In other embodiments, the transcoding module 603 is further configured to determine to perform offline transcoding on the video to be transcoded when the second target node is in the terminal layer of the computing power network. The second scheduling method; based on the second scheduling method, perform offline transcoding on the video to be transcoded through the second target node to obtain the second video after offline transcoding.

In other embodiments, the device 600 further includes: a first acquisition module, a second acquisition module and a third determination module; wherein,

The first acquisition module is configured to acquire the first computing power scheduled by the terminal layer of the computing power network;

The second acquisition module is configured to acquire the second computing power for transcoding the video to be transcoded;

The third determination module is configured to determine the revenue parameter of the terminal layer based on the first computing power and the second computing power.

The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiments of the present disclosure, please refer to the description of the method embodiments of the present disclosure for understanding.

It should be noted that in the embodiments of the present disclosure, if the above-mentioned video transcoding method based on computing power network is implemented in the form of a software function module and is sold or used as an independent product, it can also be stored in a computer-readable in the storage medium. Based on this understanding, the technical implementation of the embodiments of the present disclosure For example, the part that essentially contributes to the existing technology can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions to enable a video transcoding system based on a computing power network. The device (which may be a personal computer, a server, a network device, etc.) executes all or part of the methods described in various embodiments of the present disclosure. The aforementioned storage media include: U disk, mobile hard disk, read only memory (Read Only Memory, ROM), magnetic disk or optical disk and other various media that can store program codes. As such, disclosed embodiments are not limited to any specific combination of hardware and software.

Correspondingly, embodiments of the present disclosure provide a video transcoding device based on a computing power network, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the program, , implement any of the above methods.

Correspondingly, embodiments of the present disclosure provide a storage medium that stores executable instructions. When the executable instructions are executed by a processor, any one of the methods described above is implemented.

It should be pointed out here that the above description of the storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium and device embodiments of the present disclosure, please refer to the description of the method embodiments of the present disclosure for understanding.

It should be noted that Figure 7 is a schematic structural diagram of a hardware entity of a video transcoding device based on a computing power network according to an embodiment of the present disclosure. As shown in Figure 7 , the hardware entity of the video transcoding device 700 includes: a processor 701 and Memory 703. Optionally, the video transcoding device 700 may also include a communication interface 702.

It can be understood that the memory 703 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM, Erasable Programmable Read-Only Memory). Only Memory), Electrically Erasable Programmable Read-Only Memory (EEPROM, Electrically Erasable Programmable Read-Only Memory), Magnetic Random Access Memory (FRAM, ferromagnetic random access memory), Flash Memory, Magnetic Surface Memory , CD, or CD-ROM (Compact Disc Read-Only Memory); Magnetic surface storage can be disk storage or tape storage. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of illustration, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory Memory (DRAM, Dynamic Random Access Memory), synchronous dynamic random access memory (SDRAM, Synchronous Dynamic Random Access Memory), double data rate synchronous dynamic random access memory (DDRSDRAM, Double Data Rate Synchronous Dynamic Random Access Memory), enhanced Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), Direct Rambus Random Access Memory (DRRAM) ). The memory 703 described in embodiments of the present disclosure is intended to include, but is not limited to, these and any other suitable types of memory.

The methods disclosed in the above embodiments of the present disclosure can be applied to the processor 701 or implemented by the processor 701 . The processor 701 may be an integrated circuit chip with signal processing capabilities. During the implementation process, each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor 701 . The above-mentioned processor 701 can be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 701 can implement or execute the disclosed methods, steps and logical block diagrams in the embodiments of the present disclosure. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly implemented by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, and the storage medium is located in the memory 703. The processor 701 reads the information in the memory 703, and completes the steps of the foregoing method in combination with its hardware.

In an exemplary embodiment, the video transcoding device may be one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices ( CPLD, Complex Programmable Logic Device), Field-Programmable Gate Array (FPGA, Field-Programmable Gate Array), general-purpose processor, controller, microcontroller (MCU, Micro Controller Unit), microprocessor (Microprocessor), or other electronic components , used to execute the aforementioned method.

In the several embodiments provided in this disclosure, it should be understood that the disclosed methods and devices can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components may be combined, or can be integrated into another observation, or some features can be ignored, or not implemented. In addition, the communication connection between the various components shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical or other forms.

The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the purpose of this embodiment.

A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), magnetic disks or optical disks, and other media that can store program codes.

Alternatively, if the above-mentioned integrated units in the embodiments of the present disclosure are implemented in the form of software functional units and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical embodiments of the embodiments of the present disclosure can be embodied in the form of software products in nature or in part that contribute to the existing technology. The computer software products are stored in a storage medium and include a number of instructions. So that a video transcoding device (which can be a personal computer, a server, a network device, etc.) executes all or part of the methods described in various embodiments of the present disclosure. The aforementioned storage media include: mobile storage devices, ROMs, magnetic disks or optical disks and other media that can store program codes.

The video transcoding method, device, monitor and storage medium based on the computing power network recorded in the embodiments of the disclosure are only examples of the embodiments of the disclosure, but are not limited thereto. As long as the video transcoding based on the computing power network is involved, The transcoding method, device, monitor and storage medium are all within the protection scope of the present disclosure.

It will be understood that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present disclosure. Thus, the appearances of "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that in various embodiments of the present disclosure, the size of the sequence numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its functions and internal logic, and should not be used in the embodiments of the present disclosure. The implementation process constitutes any limitation. The above serial numbers of the embodiments of the present disclosure are only for description and do not represent the advantages and disadvantages of the embodiments.

It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or apparatus that includes that element.

The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure. should be covered by the protection scope of this disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

A video transcoding method based on computing power network, the method includes:

Determine the computing resources corresponding to the video to be transcoded;

In at least one node that meets the computing power resources, determine the target node based on the priority value corresponding to each node in the at least one node;

The target node is used to transcode the video to be transcoded to obtain a transcoded video.
The method of claim 1, wherein the transcoding includes online transcoding, and the method further includes:

Determine at least one first node in the computing power network; wherein the first node represents any node that meets the computing power resources in the edge layer of the computing power network;

Determine the first target node based on the first priority value corresponding to each node in the at least one first node;

The first target node is used to perform online transcoding on the video to be transcoded to obtain the first video after online transcoding.
The method according to claim 2, wherein determining the first target node based on the first priority value corresponding to each node in the at least one first node includes:

Obtain a first parameter corresponding to each node in the at least one first node; the first parameter represents an attribute parameter of the first node related to the fluency of the video to be transcoded;

Determine a second parameter corresponding to the first parameter based on a preset model; the second parameter represents the influence parameter of the first parameter on the fluency of the video to be transcoded;

Based on the first parameter and the second parameter, determine a first priority value corresponding to each node in the at least one first node;

The first node corresponding to the smallest first priority value is used as the first target node.
The method of claim 1, further comprising:

Construct a training sample set based on the first sample parameter and the second sample parameter associated with each node in the at least one first node;

Using the first sample parameters included in the training sample set as input and the second sample parameters included in the training sample set as output, the first model is trained to obtain a preset model.
The method according to claim 2, wherein, after determining the first target node based on the first priority value corresponding to each node in the at least one first node, the method further comprises:

Determine the first current priority value of the first target node;

Determine the second current priority value corresponding to each node in the at least one first node;

Determine the smallest second current priority value among the second current priority values;

When the first current priority value and the minimum second current priority value meet preset conditions, the first node corresponding to the minimum second current priority value is updated as the first target node.
The method according to claim 2, wherein after using the first target node to perform online transcoding on the video to be transcoded to obtain the first video after online transcoding, the method further includes:

When the first target node is closest to the user, the first video is played through the first target node.
The method of claim 1, wherein the transcoding includes offline transcoding, and the method further includes:

Determine at least one second node in the computing power network; wherein the second node represents any node that meets the computing power resources in the terminal layer, edge layer or cloud layer of the computing power network;

Determine a second target node based on a second priority value corresponding to each node in the at least one second node;

The video to be transcoded is offline transcoded using the second target node to obtain a second video after offline transcoding.
The method of claim 7, wherein determining the second target node based on the second priority value corresponding to each node in the at least one second node includes:

Obtain a third parameter corresponding to each node in the at least one second node; the third parameter represents an attribute parameter of the second node related to the fluency of the video to be transcoded;

Determine a fourth parameter corresponding to the third parameter; the fourth parameter represents the influence parameter of the third parameter on the fluency of the video to be transcoded;

Based on the third parameter and the fourth parameter, determine a second priority value corresponding to each node in the at least one second node;

The second node corresponding to the largest second priority value is used as the second target node.
The method according to claim 7, wherein said using the second target node to perform offline transcoding on the video to be transcoded to obtain the second video after offline transcoding includes:

When the second target node is in the cloud layer or edge layer of the computing power network, determine a first scheduling method for offline transcoding of the video to be transcoded;

Based on the first scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.
The method according to claim 7, wherein said using the second target node to perform offline transcoding on the video to be transcoded to obtain the second video after offline transcoding includes:

When the second target node is in the terminal layer of the computing power network, determine a second scheduling method for offline transcoding of the video to be transcoded;

Based on the second scheduling method, the video to be transcoded is offline transcoded through the second target node to obtain the second video after offline transcoding.
The method of claim 1, further comprising:

Obtain the first computing power scheduled by the terminal layer of the computing power network;

Obtain the second computing power to transcode the video to be transcoded;

Based on the first computing power and the second computing power, a revenue parameter of the terminal layer is determined.
A video transcoding device based on computing power network, wherein the device includes:

The first determination module is configured to determine the computing resources corresponding to the video to be transcoded;

The second determination module is configured to determine the target node based on the priority value corresponding to each node in the at least one node among at least one node that meets the computing power resource;

A transcoding module configured to use the target node to transcode the video to be transcoded to obtain a transcoded video.
A video transcoding device based on a computing power network, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the program, Implement the method of any one of claims 1 to 11.
A storage medium, wherein the storage medium stores executable instructions, and when the executable instructions are executed by a processor, the method described in any one of claims 1 to 11 is implemented.