Algorithm scheduling method, device and equipment, articulated naturality web platform and readable storage medium

Abstract

The invention discloses an algorithm scheduling method, apparatus and device, an articulated naturality web platform and a readable storage medium, the method is applied to a first algorithm management device deployed in a first-level platform of the articulated naturality web platform, and the method comprises the steps of receiving an AI algorithm order request sent by a whole network service management device, the AI algorithm order request comprising an algorithm identifier of a first algorithm; sending an AI analysis task request to the computing network calling equipment according to the AI algorithm order request, wherein the AI analysis task request comprises an algorithm identifier of a first algorithm; receiving an AI analysis task response sent by the computing network calling equipment, wherein the AI analysis task response carries an algorithm service address of the first algorithm; and sending an AI analysis task response to the whole network service management equipment. According to the scheme, the first algorithm management equipment is introduced to perform unified management of the algorithm, so that the whole network service management equipment can obtain the algorithm service address from the first algorithm management equipment, and cross-resource reuse and sharing of the algorithm can be realized.

Description

Technical Field

[0001] The present application belongs to the field of communication technology, and in particular relates to an algorithm scheduling method, device, equipment, visual networking platform and readable storage medium. Background Art

[0002] The computing power network is centered on computing and based on the network. It connects distributed computing nodes through ubiquitous networks to achieve automated deployment of services, optimal routing and load balancing, thereby building a new network infrastructure that can perceive computing power, ensuring that the network can dispatch computing resources in different locations on demand and in real time, improve the utilization rate of network and computing resources, and further enhance user experience, thus realizing the vision of omnipresent networks, ubiquitous computing power, and all-encompassing intelligence.

[0003] At present, the visual Internet of Things has the following problems and needs in improving the level of intelligence:

[0004] 1. Currently, the visual network still uses a fixed resource pool and resource-based calling method. Different algorithms are deployed in different resource pools, and the storage and computing power of different resource pools are unevenly distributed. Cross-resource reuse is impossible, resulting in resource waste and may cause video intelligent processing tasks to be queued in some cases, which is time-consuming and inefficient.

[0005] 2. Currently, each provincial secondary platform develops or introduces its own algorithm, resulting in high algorithm homogeneity and the inability to share algorithms. Summary of the invention

[0006] The embodiments of the present application provide an algorithm scheduling method, apparatus, device, visual networking platform and readable storage medium to solve the problem that current visual networking algorithms cannot be shared and cannot be reused across resources.

[0007] In order to solve the above technical problems, an embodiment of the present application provides an algorithm scheduling method, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including:

[0008] Receive an artificial intelligence (AI) algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0009] Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0010] Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0011] Send the AI ​​analysis task response to the whole network business management device.

[0012] Optionally, before sending the AI ​​analysis task request to the computing network calling device according to the AI ​​algorithm order request, the method further includes:

[0013] Sending an algorithm synchronization request to a computing device, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm;

[0014] In a case where the computing power device needs to download the first algorithm, an algorithm file of the first algorithm is sent to the computing power device.

[0015] Optionally, the method further includes:

[0016] Receiving an algorithm application sent by the network-wide service management device, wherein the algorithm application carries an algorithm identifier of a second algorithm;

[0017] Determining whether the network-wide service management device has download authority for the second algorithm;

[0018] In a case where the whole-network service management device has the download authority for the second algorithm, the algorithm encapsulation file download address of the second algorithm is sent to the whole-network service management device.

[0019] Optionally, the method further includes:

[0020] Receiving algorithm subscription information sent by the network-wide service management device, wherein the algorithm subscription information includes an algorithm information receiving address;

[0021] According to the algorithm subscription information, an algorithm information notification is sent to the network-wide service management device.

[0022] The embodiment of the present application also provides an algorithm scheduling method, which is applied to a computing network calling device deployed in a primary platform of a visual network platform, including:

[0023] Receiving an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0024] Sending the AI ​​analysis task request to the computing device;

[0025] receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0026] The AI ​​analysis task response is sent to the first algorithm management device.

[0027] Optionally, the method further includes:

[0028] Receiving the analysis result sent by the computing power device;

[0029] The analysis result is sent to the service management device of the whole network.

[0030] The embodiment of the present application also provides an algorithm scheduling method, which is applied to a network-wide service management device, including:

[0031] Sending an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0032] receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0033] Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0034] Receive the analysis result fed back by the computing power device.

[0035] Optionally, the receiving the analysis result fed back by the computing power device includes:

[0036] Receive the analysis results fed back by the computing power device through the computing network calling device.

[0037] Optionally, the method further includes:

[0038] Sending an algorithm application to the first algorithm management device, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0039] Receive the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device;

[0040] The algorithm encapsulation file download address and algorithm attribute information of the second algorithm are sent to the second algorithm management device.

[0041] The present application also provides a visual networking platform, including:

[0042] First algorithm management equipment, whole network service management equipment, and network calculation calling equipment;

[0043] Wherein, the first algorithm management device, the whole network service management device and the computing network calling device are deployed on a first-level platform;

[0044] The first algorithm management device receives an artificial intelligence (AI) algorithm order request sent by the whole network service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; sends an AI analysis task request to the computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0045] The computing network calling device sends the AI ​​analysis task request to the computing power device; receives the AI ​​analysis task response fed back by the computing power device, wherein the AI ​​analysis task response carries the algorithm service address of the first algorithm; and sends the AI ​​analysis task response to the first algorithm management device;

[0046] The first algorithm management device sends the AI ​​analysis task response to the whole network service management device;

[0047] The whole network business management device sends an AI analysis request to the computing power device, wherein the AI ​​analysis request includes the algorithm service address of the first algorithm; and receives the analysis result fed back by the computing power device.

[0048] Optionally, the method further includes:

[0049] A second algorithm management device deployed on the secondary platform;

[0050] Among them, the whole network business management device sends an algorithm application to the first algorithm management device, and the algorithm application carries the algorithm identifier of the second algorithm; receives the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device; sends the algorithm encapsulation file download address and algorithm attribute information of the second algorithm to the second algorithm management device; the second algorithm management device downloads the algorithm file from the first algorithm management device according to the algorithm encapsulation file download address and the algorithm attribute information.

[0051] The embodiment of the present application further provides an algorithm scheduling device, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including:

[0052] A first receiving module is used to receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0053] A first sending module, configured to send an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0054] A second receiving module is used to receive an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0055] The second sending module is used to send the AI ​​analysis task response to the whole network business management device.

[0056] The embodiment of the present application also provides an algorithm scheduling device, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including a transceiver and a processor;

[0057] The transceiver is used to: receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0058] Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0059] Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0060] Send the AI ​​analysis task response to the whole network business management device.

[0061] An embodiment of the present application also provides an algorithm management device, which is a first algorithm management device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned algorithm scheduling method when executing the program.

[0062] The embodiment of the present application further provides an algorithm scheduling device, which is applied to a computing network calling device deployed in a primary platform of a visual network platform, including:

[0063] A third receiving module is used to receive an AI analysis task request sent by the first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0064] A third sending module is used to send the AI ​​analysis task request to the computing power device;

[0065] A fourth receiving module, configured to receive an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0066] The fourth sending module is used to send the AI ​​analysis task response to the first algorithm management device.

[0067] The embodiment of the present application also provides an algorithm scheduling device, which is applied to a computing network calling device deployed in a primary platform of a visual network platform, including a transceiver and a processor;

[0068] The transceiver is used to: receive an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0069] Sending the AI ​​analysis task request to the computing device;

[0070] receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0071] The AI ​​analysis task response is sent to the first algorithm management device.

[0072] An embodiment of the present application also provides a computing network calling device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned algorithm scheduling method when executing the program.

[0073] The embodiment of the present application also provides an algorithm scheduling device, which is applied to a network-wide service management device, including:

[0074] A fifth sending module, configured to send an AI algorithm order request to the first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0075] a fifth receiving module, configured to receive an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0076] a sixth sending module, configured to send an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0077] The sixth receiving module is used to receive the analysis results fed back by the computing power device.

[0078] The embodiment of the present application also provides an algorithm scheduling device, which is applied to a network-wide service management device, including a transceiver and a processor;

[0079] The transceiver is used to: send an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0080] receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0081] Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0082] Receive the analysis result fed back by the computing power device.

[0083] An embodiment of the present application also provides a whole-network service management device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned algorithm scheduling method when executing the program.

[0084] An embodiment of the present application also provides a readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps in the above method are implemented.

[0085] An embodiment of the present application also provides a computer program product, including computer instructions, which implement the steps of the above method when executed by a processor.

[0086] The beneficial effects of this application are:

[0087] The above scheme receives an AI algorithm order request including an algorithm identifier of a first algorithm sent by a network-wide business management device, sends an AI analysis task request including an algorithm identifier of the first algorithm to a computing network calling device according to the AI ​​algorithm order request, receives an AI analysis task response carrying an algorithm service address of the first algorithm sent by the computing network calling device, and sends the AI ​​analysis task response to the network-wide business management device. By introducing the first algorithm management device to perform unified management of algorithms, the network-wide business management device can obtain the algorithm service address from the first algorithm management device, thereby realizing cross-resource reuse and sharing of algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 This is one of the flowcharts of the algorithm scheduling method according to the embodiment of the present application;

[0089] Figure 2 It is an architecture diagram of the visual networking platform of an embodiment of the present application;

[0090] Figure 3 It is a schematic diagram of the task scheduling process;

[0091] Figure 4 It is a flowchart of algorithm synchronization;

[0092] Figure 5 This is the second flow chart of the algorithm scheduling method of the embodiment of the present application;

[0093] Figure 6 This is the third flow chart of the algorithm scheduling method of the embodiment of the present application;

[0094] Figure 7 It is one of the module schematic diagrams of the algorithm scheduling device in the embodiment of the present application;

[0095] Figure 8 A structural diagram showing an algorithm management device according to an embodiment of the present application;

[0096] Fig. 9 This is the second module schematic diagram of the algorithm scheduling device in the embodiment of the present application;

[0097] Fig.10 This is the third module schematic diagram of the algorithm scheduling device of the embodiment of the present application. DETAILED DESCRIPTION

[0098] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0099] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0100] The algorithm scheduling method, device, equipment, visual networking platform and readable storage medium provided in the embodiments of the present application are described in detail below in combination with the accompanying drawings through specific embodiments and their application scenarios.

[0101] like Figure 1 As shown, at least one embodiment of the present application provides an algorithm scheduling method, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including:

[0102] Step 101, receiving an artificial intelligence AI algorithm order request sent by a network-wide service management device;

[0103] Optionally, the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; further, the AI ​​algorithm order request may also include but is not limited to at least one of the following: user information, camera information, start and stop time, processing frequency requirements, time requirements, and regional requirements.

[0104] Optionally, when the whole network business management device receives the AI ​​algorithm calling request sent by the visual network application, the whole network business management device sends an AI algorithm order request to the first algorithm management device to request to call the algorithm requested by the AI ​​algorithm calling request.

[0105] Optionally, the first algorithm management device in the embodiment of the present application is deployed in the first-level platform of the visual Internet platform to provide full-network algorithm capabilities.

[0106] Step 102: Send an AI analysis task request to the computing network calling device according to the AI ​​algorithm order request;

[0107] Optionally, the AI ​​analysis task request includes the algorithm identifier of the first algorithm; further, the AI ​​analysis task request may also include but is not limited to at least one of the following: platform identifier (used to identify the platform for algorithm scheduling), priority region requirements, start and stop time, processing frequency, and time requirements.

[0108] Step 103, receiving the AI ​​analysis task response sent by the computing network calling device;

[0109] Optionally, the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0110] Step 104: Send the AI ​​analysis task response to the whole network business management device.

[0111] It should be noted that in the embodiment of the present application, the whole network business management device sends an AI algorithm order request to the first algorithm management device, and the first algorithm management device sends an AI analysis task request to the computing network calling device based on the AI ​​algorithm order request to obtain the algorithm service address from the computing network calling device, and feeds back the algorithm service address to the whole network business management device, and the whole network business management device makes an AI analysis request based on the algorithm service address to implement algorithm call to obtain analysis results, thereby realizing cross-resource reuse and sharing of algorithms.

[0112] It should be noted that the first algorithm management device mentioned in the embodiment of the present application can be understood as an algorithm management device set in a first-level platform in the visual networking platform. For example, the first algorithm management device can also be called a first-level AI algorithm warehouse; the whole-network business management device can be understood as a whole-network business management device set in a first-level platform in the visual networking platform. For example, the whole-network business management device can also be called a whole-network business center; the computing network calling device can be understood as a computing network calling device set in a first-level platform in the visual networking platform. For example, the computing network calling device can also be called a computing network calling center.

[0113] Optionally, the architecture of the visual networking platform of the embodiment of the present application is as follows: Figure 2 As shown, it mainly includes:

[0114] The whole network business management equipment, first algorithm management equipment, and computing network calling equipment deployed in the first-level platform; the second algorithm management equipment (also called the second-level AI algorithm warehouse) and streaming media processing equipment deployed in the second-level platform.

[0115] Optionally, the first-level platform mentioned in the embodiment of the present application is a network-wide business center that supports business user and equipment management, operation data visualization and reporting, capability exposure and other functions; the nodes in the second-level platform are mostly deployed in various provinces / cities to provide video surveillance services for various provinces / cities.

[0116] Optionally, the whole network business management device is connected to the visual network application through the IF1 interface to realize the customer-initiated AI algorithm call request; the whole network business management device is connected to the first algorithm management device through the IF3 interface to realize algorithm information subscription, AI algorithm order request, AI real-time analysis request, and AI analysis result reporting; the first algorithm management device is connected to the internal and external algorithm platform through the IF2 interface to realize algorithm introduction; the first algorithm management device is connected to the computing network calling device through the IF4 interface to start AI analysis tasks, algorithm synchronization, and AI analysis result reporting; the computing network calling device is connected to the computing power device (also known as the computing network brain or computing power network) through the IF5 interface to realize Synchronize business analysis tasks and report task analysis results; the computing power device is connected to the streaming media processing device in the secondary platform through the IF7 interface to realize the download of pictures or video streams required for business analysis; the whole network business management device is connected to the second algorithm management device in the secondary platform through the IF6 interface for algorithm synchronization and operation data synchronization; optionally, the whole network business management device can also be directly connected to the computing power device through the IF8 interface to realize the startup of AI analysis tasks and reporting of AI analysis results; the first algorithm management device is connected to the computing power device through the IF9 interface to realize algorithm synchronization; the computing network calling device is connected to the whole network business management device through the IF4 interface to realize the reporting of AI analysis results.

[0117] Optionally, the functions that can be realized by the first algorithm management device mainly include: being responsible for the unified operation of algorithms, for example, introducing ecological AI algorithms, providing management, access, and distribution of AI algorithm capabilities at the whole network level, realizing the management, release, update, and delisting of algorithms throughout their life cycle, supporting algorithm orchestration to form AI services, etc.; being able to connect to the computing network to call equipment to perform task scheduling, realizing rapid distribution of algorithms across the entire network, and flexible task-based distribution of algorithm services; being able to be responsible for collecting algorithm-related operation data from the second-level platforms of each province, and conducting comprehensive data management and analysis. Specifically, the first algorithm management device mainly includes the following functional modules:

[0118] The algorithm operation module is mainly used to provide functions such as algorithm storage, algorithm mall, ordering management, and billing settlement;

[0119] The algorithm management module is mainly used to implement functions such as image storage, algorithm list, version management, and algorithm release;

[0120] The algorithm operation module is mainly used to realize functions such as algorithm distribution, task processing, algorithm orchestration, and computing network scheduling.

[0121] Optionally, the functions that can be realized by the computing network calling device mainly include: realizing the call of computing network capabilities, being responsible for receiving AI algorithm order requests, creating tasks and sending them to the computing network brain, receiving the returned task results and feeding back to the requester.

[0122] Optionally, the functions that can be realized by the whole network business management device mainly include: business subscription management, user management, device management, support for operation data visualization and reporting, support for capability calling, etc. Optionally, the whole network business management device can also support algorithm information subscription, algorithm download, AI analysis order request, AI analysis result reception, and operation data reporting.

[0123] Optionally, the functions that can be realized by the second algorithm management device mainly include: algorithm image management, reasoning analysis, algorithm distribution to edge ends, etc. Optionally, the second algorithm management device can also support receiving algorithm synchronization and operation data reporting from the service management device of the entire network.

[0124] Optionally, the functions that can be realized by the streaming media processing device mainly include: media access, media storage, media forwarding, media distribution, real-time on-demand, video playback, audio and video intercom, etc.

[0125] Optionally, the functions that can be realized by the computing power device mainly include: downloading various algorithms required for business analysis from the first algorithm management device and the second algorithm management device, and completing algorithm deployment; receiving business tasks from the computing network calling device, calling the computing network resources of the entire network to complete the task, and then reporting the task results.

[0126] Optionally, in order to ensure that the computing device can smoothly execute the algorithm, the computing device usually needs to synchronize the algorithm used in the first algorithm management device. In one implementation, before sending the AI ​​analysis task request to the computing network calling device according to the AI ​​algorithm order request, it also includes:

[0127] Sending an algorithm synchronization request to a computing device, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm;

[0128] In a case where the computing power device needs to download the first algorithm, an algorithm file of the first algorithm is sent to the computing power device.

[0129] It should be noted that in order to ensure that the computing device can smoothly execute the algorithm, when the first algorithm is not stored on the computing device side, it is necessary to download the algorithm file from the first algorithm management device.

[0130] It should be noted that the above process can realize the AI ​​algorithm task scheduling based on the first algorithm management device. Specifically, Figure 3 As shown in the figure, the detailed implementation process of AI algorithm task scheduling includes:

[0131] S301, the visual networking application sends an AI algorithm call request to the network-wide service management device, where the AI ​​algorithm call request is used to request to call and execute a certain AI algorithm, for example, AI algorithm 1;

[0132] S302. The whole network business management device sends an AI algorithm order request to the first algorithm management device. The AI ​​algorithm order request includes: an algorithm identifier. Optionally, the algorithm identifier can be understood as a unique identifier of the algorithm in the first algorithm management device. Optionally, the AI ​​algorithm order request can also include one or more of the following: user information, camera information, start and stop time, processing frequency requirements, time requirements, and regional requirements.

[0133] Optionally, the whole network business management device can first request the second algorithm management device of the secondary platform to execute the AI ​​algorithm, and return the result if the execution is successful; if the execution fails due to insufficient resources or other reasons, or the algorithm is in the first algorithm management device, then execute step S302.

[0134] S303: The first algorithm management device sends an algorithm synchronization request to the computing power device;

[0135] Optionally, the first algorithm management device analyzes the order requirements and sends an algorithm synchronization request to the computing power device, where the algorithm synchronization request carries one or more of the following: platform identifier, algorithm identifier, algorithm configuration information, algorithm mirror address, resources required for the algorithm, and priority region requirements.

[0136] S304: The computing power device downloads the algorithm file from the first algorithm management device, that is, the computing power device downloads the algorithm image from the first algorithm management device as needed.

[0137] S305. The first algorithm management device sends an AI analysis task request to the computing network calling device. The AI ​​analysis task request includes an algorithm identifier. Optionally, the AI ​​analysis task request may also carry a platform identifier, a priority region requirement, a start and stop time, a processing frequency requirement, and a time requirement.

[0138] S306, the computing network calling device sends an AI analysis task request to the computing power device;

[0139] S307, the computing power device dispatches computing network resources to start the AI ​​analysis task;

[0140] S308. The computing power device sends an AI analysis task response to the computing network calling device. The AI ​​analysis task response carries the algorithm service address.

[0141] S309, the computing network calling device returns the AI ​​analysis task response to the whole network service management device through the first algorithm management device;

[0142] S310. The network-wide service management device sends an AI analysis request to the computing device, where the AI ​​analysis request includes an algorithm service address.

[0143] Optionally, the AI ​​analysis request may also carry algorithm input parameters such as image / video addresses.

[0144] S311, computing equipment downloads pictures / videos on demand;

[0145] In this step, the computing device downloads the image / video used for algorithm execution from the streaming media processing device according to the image / video address.

[0146] S312, the computing device performs AI analysis and processing on the image / video to obtain analysis results;

[0147] S313. The computing power device calls the device through the computing network and returns the analysis results to the business management device of the entire network.

[0148] Optionally, the embodiment of the present application may also implement synchronization of algorithms in the first algorithm management device and the second algorithm management device. Specifically, in one implementation, the method further includes:

[0149] The first algorithm management device receives the algorithm application sent by the whole network service management device, wherein the algorithm application carries the algorithm identifier of the second algorithm;

[0150] Determining whether the network-wide service management device has download authority for the second algorithm;

[0151] In a case where the whole-network service management device has the download authority for the second algorithm, the algorithm encapsulation file download address of the second algorithm is sent to the whole-network service management device.

[0152] It should be noted that after the first algorithm management device sends the algorithm encapsulation file download address of the second algorithm to the whole network business management device, the whole network business management device can send the algorithm encapsulation file download address and algorithm attribute information of the second algorithm to the second algorithm management device, and the second algorithm management device can download the relevant algorithm file from the first algorithm management device based on the algorithm encapsulation file download address and algorithm attribute information.

[0153] Optionally, in one implementation, the method further includes:

[0154] Receiving algorithm subscription information sent by the network-wide service management device, wherein the algorithm subscription information includes an algorithm information receiving address;

[0155] According to the algorithm subscription information, an algorithm information notification is sent to the network-wide service management device.

[0156] It should be noted that by subscribing to the algorithm to the first algorithm management device, when the algorithm in the first algorithm management device changes, the business management device of the entire network can be notified in time, thereby ensuring that the business management device of the entire network can determine the usable algorithm in real time and ensure the smooth execution of the algorithm business.

[0157] like Figure 4 As shown, the detailed process of the algorithm synchronization process of the embodiment of the present application specifically includes:

[0158] S401. The whole network service management device sends algorithm subscription information to the first algorithm management device, where the algorithm subscription information includes an algorithm information receiving address;

[0159] S402. The first algorithm management device sends an algorithm information notification to the whole network service management device. Optionally, the algorithm information notification may include: an algorithm list and algorithm attribute information of each type of algorithm (for example, the algorithm attribute information may include an algorithm description, computing power required for the algorithm, input parameters, algorithm identifier, etc.).

[0160] S403, the whole network service management device sends an algorithm application to the first algorithm management device, where the algorithm application carries an algorithm identifier of the algorithm to be applied for;

[0161] Optionally, the whole network service management device may send an algorithm application to the first algorithm management device when a certain algorithm is needed.

[0162] S404, the first algorithm management device determines whether the whole network service management device has the download authority of the applied algorithm;

[0163] S405, when the network-wide service management device has the download authority for the applied algorithm, the algorithm encapsulation file download address of the applied algorithm is sent to the network-wide service management device;

[0164] S406. The whole network service management device synchronizes the algorithm encapsulation file download address and algorithm attribute information of the algorithm to the second algorithm management device of each secondary platform as needed;

[0165] S407, the second algorithm management device synchronizes the algorithm from the first algorithm management device as needed according to the business needs of each province;

[0166] Optionally, the synchronization may be understood as downloading the algorithm file from the first algorithm management device.

[0167] S408. The second algorithm management device synchronizes operation data with the whole network service management device.

[0168] In summary, the embodiments of the present application combine the AI ​​needs of the visual Internet of Things with the computing network capabilities. Based on the business characteristics of the visual Internet of Things, it can support the unified introduction and distribution of algorithms for the entire network, and realize the operation and sharing of algorithms for the entire network. It can also build a unified AI algorithm scheduling mechanism for the entire network, provide one-stop AI task-based services, and realize efficient scheduling and utilization of AI computing resources.

[0169] like Figure 5 As shown, the embodiment of the present application also provides an algorithm scheduling method, which is applied to a computing network calling device deployed in a primary platform of a visual network platform, including:

[0170] Step 501: receiving an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0171] Step 502: Send the AI ​​analysis task request to the computing device;

[0172] Step 503: receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0173] Step 504: Send the AI ​​analysis task response to the first algorithm management device.

[0174] Optionally, the method further includes:

[0175] Receiving the analysis result sent by the computing power device;

[0176] The analysis result is sent to the service management device of the whole network.

[0177] It should be noted that all descriptions about the computing network calling device side in the above embodiments are applicable to the embodiments of the algorithm scheduling method applied to the computing network calling device side, and can achieve the same technical effects, so they will not be repeated here.

[0178] like Figure 6 As shown, the embodiment of the present application also provides an algorithm scheduling method, which is applied to a network-wide service management device, including:

[0179] Step 601: Send an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0180] Step 602: receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0181] Step 603: Send an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0182] Step 604, receiving the analysis result fed back by the computing power device.

[0183] Optionally, the receiving the analysis result fed back by the computing power device includes:

[0184] Receive the analysis results fed back by the computing power device through the computing network calling device.

[0185] Optionally, the method further includes:

[0186] Sending an algorithm application to the first algorithm management device, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0187] Receive the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device;

[0188] The algorithm encapsulation file download address and algorithm attribute information of the second algorithm are sent to the second algorithm management device.

[0189] It should be noted that all descriptions about the whole network business management device side in the above embodiments are applicable to the embodiments of the algorithm scheduling method applied to the whole network business management device side, and can achieve the same technical effect, which will not be repeated here.

[0190] The present application also provides a visual networking platform, including:

[0191] First algorithm management equipment, whole network service management equipment, and network calculation calling equipment;

[0192] Wherein, the first algorithm management device, the whole network service management device and the computing network calling device are deployed on a first-level platform;

[0193] The first algorithm management device receives an artificial intelligence (AI) algorithm order request sent by the whole network service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; sends an AI analysis task request to the computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0194] The computing network calling device sends the AI ​​analysis task request to the computing power device; receives the AI ​​analysis task response fed back by the computing power device, wherein the AI ​​analysis task response carries the algorithm service address of the first algorithm; and sends the AI ​​analysis task response to the first algorithm management device;

[0195] The first algorithm management device sends the AI ​​analysis task response to the whole network service management device;

[0196] The whole network business management device sends an AI analysis request to the computing power device, wherein the AI ​​analysis request includes the algorithm service address of the first algorithm; and receives the analysis result fed back by the computing power device.

[0197] Optionally, the visual networking platform further includes:

[0198] A second algorithm management device deployed on the secondary platform;

[0199] Among them, the whole network business management device sends an algorithm application to the first algorithm management device, and the algorithm application carries the algorithm identifier of the second algorithm; receives the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device; sends the algorithm encapsulation file download address and algorithm attribute information of the second algorithm to the second algorithm management device; the second algorithm management device downloads the algorithm file from the first algorithm management device according to the algorithm encapsulation file download address and the algorithm attribute information.

[0200] It should be noted that all descriptions about the visual networking platform in the above embodiments are applicable to this embodiment and can achieve the same technical effects, and will not be repeated here.

[0201] like Figure 7 As shown, at least one embodiment of the present application further provides an algorithm scheduling device 700, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including:

[0202] The first receiving module 701 is used to receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0203] A first sending module 702 is used to send an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0204] The second receiving module 703 is used to receive the AI ​​analysis task response sent by the computing network calling device, where the AI ​​analysis task response carries the algorithm service address of the first algorithm;

[0205] The second sending module 704 is used to send the AI ​​analysis task response to the whole network business management device.

[0206] Optionally, before the first sending module 702 sends the AI ​​analysis task request to the computing network calling device according to the AI ​​algorithm order request, it also includes:

[0207] A seventh sending module, configured to send an algorithm synchronization request to the computing power device, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm;

[0208] An eighth sending module is used to send the algorithm file of the first algorithm to the computing power device when the computing power device needs to download the first algorithm.

[0209] Optionally, the device further comprises:

[0210] a seventh receiving module, configured to receive an algorithm application sent by the network-wide service management device, wherein the algorithm application carries an algorithm identifier of a second algorithm;

[0211] A judgment module, used to judge whether the whole network service management device has the download authority of the second algorithm;

[0212] The ninth sending module is used to send the algorithm encapsulation file download address of the second algorithm to the whole network business management device when the whole network business management device has the download authority of the second algorithm.

[0213] Optionally, the device further comprises:

[0214] An eighth receiving module, configured to receive the algorithm subscription information sent by the network-wide service management device, wherein the algorithm subscription information includes an algorithm information receiving address;

[0215] The tenth sending module is used to send an algorithm information notification to the whole network service management device according to the algorithm subscription information.

[0216] It should be noted that the device provided in at least one embodiment of the present application is a device capable of executing the above-mentioned algorithm scheduling method, then all embodiments of the above-mentioned algorithm scheduling method are applicable to the device and can achieve the same or similar beneficial effects.

[0217] At least one embodiment of the present application further provides an algorithm scheduling device, which is applied to a first algorithm management device deployed in a primary platform of a visual networking platform, including a transceiver and a processor;

[0218] The transceiver is used to: receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm;

[0219] Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0220] Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0221] Send the AI ​​analysis task response to the whole network business management device.

[0222] Optionally, the transceiver is further used for:

[0223] Sending an algorithm synchronization request to a computing device, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm;

[0224] In a case where the computing power device needs to download the first algorithm, an algorithm file of the first algorithm is sent to the computing power device.

[0225] Optionally, the transceiver is further used for:

[0226] Receiving an algorithm application sent by the network-wide service management device, wherein the algorithm application carries an algorithm identifier of a second algorithm;

[0227] Determining whether the network-wide service management device has download authority for the second algorithm;

[0228] In a case where the whole-network service management device has the download authority for the second algorithm, the algorithm encapsulation file download address of the second algorithm is sent to the whole-network service management device.

[0229] Optionally, the transceiver is further used for:

[0230] Receiving algorithm subscription information sent by the network-wide service management device, wherein the algorithm subscription information includes an algorithm information receiving address;

[0231] According to the algorithm subscription information, an algorithm information notification is sent to the network-wide service management device.

[0232] like Figure 8 As shown, an embodiment of the present invention further provides an algorithm management device, which is a first algorithm management device, including a processor 800, a transceiver 810, a memory 820, and a program stored in the memory 820 and executable on the processor 800; wherein the transceiver 810 is connected to the processor 800 and the memory 820 through a bus interface, wherein the processor 800 is used to read the program in the memory and execute the following process:

[0233] Receive, through the transceiver 810, an artificial intelligence AI algorithm order request sent by the network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0234] Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0235] Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0236] Send the AI ​​analysis task response to the whole network business management device.

[0237] The transceiver 810 is configured to receive and send data under the control of the processor 800 .

[0238] Among them, Figure 8 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 800 and memory represented by memory 820. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 810 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, including transmission media such as wireless channels, wired channels, optical cables, and the like.

[0239] The processor 800 is responsible for managing the bus architecture and general processing, and the memory 820 can store data used by the processor 800 when performing operations.

[0240] Optionally, the processor 800 may be a CPU (central processing unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or a CPLD (Complex Programmable Logic Device), and the processor may also adopt a multi-core architecture.

[0241] The processor calls the computer program stored in the memory to execute any of the methods provided in the embodiments of the present application according to the obtained executable instructions. The processor and the memory can also be arranged physically separately.

[0242] Optionally, the processor 800 is used to read the program in the memory and further perform the following process:

[0243] Sending an algorithm synchronization request to the computing device through the transceiver 810, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm;

[0244] In a case where the computing power device needs to download the first algorithm, an algorithm file of the first algorithm is sent to the computing power device.

[0245] Optionally, the processor 800 is used to read the program in the memory and further perform the following process:

[0246] Receiving, through the transceiver 810, an algorithm application sent by the network-wide service management device, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0247] Determining whether the network-wide service management device has download authority for the second algorithm;

[0248] In a case where the whole-network service management device has the download authority for the second algorithm, the algorithm encapsulation file download address of the second algorithm is sent to the whole-network service management device.

[0249] Optionally, the processor 800 is used to read the program in the memory and further perform the following process:

[0250] Receiving the algorithm subscription information sent by the network-wide service management device through the transceiver 810, wherein the algorithm subscription information includes the algorithm information receiving address;

[0251] According to the algorithm subscription information, an algorithm information notification is sent to the network-wide service management device.

[0252] At least one embodiment of the present application also provides an algorithm management device, which is a first algorithm management device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, each process in the algorithm scheduling method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0253] like Fig. 9 As shown, at least one embodiment of the present application further provides an algorithm scheduling device 900, which is applied to a computing network calling device deployed in a primary platform of a visual network platform, including:

[0254] A third receiving module 901 is configured to receive an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0255] The third sending module 902 is used to send the AI ​​analysis task request to the computing power device;

[0256] The fourth receiving module 903 is used to receive the AI ​​analysis task response fed back by the computing power device, where the AI ​​analysis task response carries the algorithm service address of the first algorithm;

[0257] The fourth sending module 904 is used to send the AI ​​analysis task response to the first algorithm management device.

[0258] Optionally, the device further comprises:

[0259] A ninth receiving module, used to receive the analysis result sent by the computing power device;

[0260] The eleventh sending module is used to send the analysis result to the service management device of the whole network.

[0261] It should be noted that the device provided in at least one embodiment of the present application is a device capable of executing the above-mentioned algorithm scheduling method, then all embodiments of the above-mentioned algorithm scheduling method are applicable to the device and can achieve the same or similar beneficial effects.

[0262] At least one embodiment of the present application further provides an algorithm scheduling device, which is applied to a computing network calling device deployed in a primary platform of a visual networking platform, including a transceiver and a processor;

[0263] The transceiver is used to: receive an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0264] Sending the AI ​​analysis task request to the computing device;

[0265] receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0266] The AI ​​analysis task response is sent to the first algorithm management device.

[0267] Optionally, the transceiver is further used for:

[0268] Receiving the analysis result sent by the computing power device;

[0269] The analysis result is sent to the service management device of the whole network.

[0270] At least one embodiment of the present application further provides a computing network calling device. The specific structure of the computing network calling device can be found in Figure 8 As shown, no further description is given here.

[0271] The processor is used to read the program in the memory and execute the following processes:

[0272] Receiving, by a transceiver, an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm;

[0273] Sending the AI ​​analysis task request to the computing device;

[0274] receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0275] The AI ​​analysis task response is sent to the first algorithm management device.

[0276] Optionally, the processor is configured to read a program in the memory and further perform the following process:

[0277] Receiving the analysis result sent by the computing power device;

[0278] The analysis result is sent to the service management device of the whole network.

[0279] like Fig.10 As shown, at least one embodiment of the present application further provides an algorithm scheduling device 1000, which is applied to a network-wide service management device, including:

[0280] The fifth sending module 1001 is used to send an AI algorithm order request to the first algorithm management device, where the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0281] A fifth receiving module 1002 is configured to receive an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0282] The sixth sending module 1003 is used to send an AI analysis request to the computing power device, where the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0283] The sixth receiving module 1004 is used to receive the analysis results fed back by the computing power device.

[0284] Optionally, the sixth receiving module 1004 is configured to:

[0285] Receive the analysis results fed back by the computing power device through the computing network calling device.

[0286] Optionally, the device further comprises:

[0287] A twelfth sending module, configured to send an algorithm application to the first algorithm management device, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0288] a tenth receiving module, configured to receive a download address of an algorithm encapsulation file of the second algorithm fed back by the first algorithm management device;

[0289] The thirteenth sending module is used to send the algorithm encapsulation file download address and algorithm attribute information of the second algorithm to the second algorithm management device.

[0290] It should be noted that the device provided in at least one embodiment of the present application is a device capable of executing the above-mentioned algorithm scheduling method, then all embodiments of the above-mentioned algorithm scheduling method are applicable to the device and can achieve the same or similar beneficial effects.

[0291] At least one embodiment of the present application further provides an algorithm scheduling device, which is applied to a network-wide service management device, including a transceiver and a processor;

[0292] The transceiver is used to: send an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0293] receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0294] Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0295] Receive the analysis result fed back by the computing power device.

[0296] Optionally, the transceiver is used to:

[0297] Receive the analysis results fed back by the computing power device through the computing network calling device.

[0298] Optionally, the transceiver is further used for:

[0299] Sending an algorithm application to the first algorithm management device, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0300] Receive the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device;

[0301] The algorithm encapsulation file download address and algorithm attribute information of the second algorithm are sent to the second algorithm management device.

[0302] At least one embodiment of the present application further provides a whole network service management device. The specific structure of the whole network service management device can be found in Figure 8 As shown, no further description is given here.

[0303] The processor is used to read the program in the memory and execute the following processes:

[0304] Sending an AI algorithm order request to a first algorithm management device through a transceiver, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm;

[0305] receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm;

[0306] Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm;

[0307] Receive the analysis result fed back by the computing power device.

[0308] Optionally, the processor is used to read a program in a memory and execute the following process:

[0309] The analysis results fed back by the computing power device through the computing network calling device are received through the transceiver.

[0310] Optionally, the processor is used to read the program in the memory and further perform the following process:

[0311] Sending an algorithm application to the first algorithm management device through a transceiver, wherein the algorithm application carries an algorithm identifier of the second algorithm;

[0312] Receive the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device;

[0313] The algorithm encapsulation file download address and algorithm attribute information of the second algorithm are sent to the second algorithm management device.

[0314] At least one embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, each process in the algorithm scheduling method embodiment described above is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0315] The embodiment of the present application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the various processes of the above-mentioned method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, they are not repeated here.

[0316] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0317] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0318] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. An algorithm scheduling method, applied to a first algorithm management device deployed in a primary platform of a visual networking platform, characterized in that: include: Receive an artificial intelligence (AI) algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm; Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; Send the AI ​​analysis task response to the whole network business management device.

2. The method according to claim 1, characterized in that: Before sending the AI ​​analysis task request to the computing network calling device according to the AI ​​algorithm order request, it also includes: Sending an algorithm synchronization request to a computing device, wherein the algorithm synchronization request includes: an algorithm identifier and / or an algorithm mirror address of the first algorithm; In a case where the computing power device needs to download the first algorithm, an algorithm file of the first algorithm is sent to the computing power device.

3. The method according to claim 1, characterized in that Also includes: Receiving an algorithm application sent by the network-wide service management device, wherein the algorithm application carries an algorithm identifier of a second algorithm; Determining whether the network-wide service management device has download authority for the second algorithm; In a case where the whole-network service management device has the download authority for the second algorithm, the algorithm encapsulation file download address of the second algorithm is sent to the whole-network service management device.

4. The method according to claim 1 or 3, characterized in that: Also includes: Receiving algorithm subscription information sent by the network-wide service management device, wherein the algorithm subscription information includes an algorithm information receiving address; According to the algorithm subscription information, an algorithm information notification is sent to the network-wide service management device.

5. An algorithm scheduling method, applied to a computing network calling device deployed in a primary platform of a visual networking platform, characterized in that: include: Receiving an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; Sending the AI ​​analysis task request to the computing device; receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; The AI ​​analysis task response is sent to the first algorithm management device.

6. The method according to claim 5, characterized in that Also includes: Receiving the analysis result sent by the computing power device; The analysis result is sent to the service management device of the whole network.

7. An algorithm scheduling method, applied to a network-wide service management device, characterized in that: include: Sending an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm; Receive the analysis result fed back by the computing power device.

8. The method according to claim 7, characterized in that The receiving of the analysis result fed back by the computing power device includes: Receive the analysis results fed back by the computing power device through the computing network calling device.

9. The method according to claim 7, characterized in that: Also includes: Sending an algorithm application to the first algorithm management device, wherein the algorithm application carries an algorithm identifier of the second algorithm; Receive the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device; The algorithm encapsulation file download address and algorithm attribute information of the second algorithm are sent to the second algorithm management device.

10. A visual networking platform, characterized in that: include: First algorithm management equipment, whole network service management equipment, and network calculation calling equipment; Wherein, the first algorithm management device, the whole network service management device and the computing network calling device are deployed on a first-level platform; The first algorithm management device receives an artificial intelligence (AI) algorithm order request sent by the whole network service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; sends an AI analysis task request to the computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; The computing network calling device sends the AI ​​analysis task request to the computing power device; receives the AI ​​analysis task response fed back by the computing power device, wherein the AI ​​analysis task response carries the algorithm service address of the first algorithm; and sends the AI ​​analysis task response to the first algorithm management device; The first algorithm management device sends the AI ​​analysis task response to the whole network service management device; The whole network business management device sends an AI analysis request to the computing power device, wherein the AI ​​analysis request includes the algorithm service address of the first algorithm; and receives the analysis result fed back by the computing power device.

11. The visual networking platform according to claim 10, characterized in that: Also includes: A second algorithm management device deployed on the secondary platform; Among them, the whole network business management device sends an algorithm application to the first algorithm management device, and the algorithm application carries the algorithm identifier of the second algorithm; receives the algorithm encapsulation file download address of the second algorithm fed back by the first algorithm management device; sends the algorithm encapsulation file download address and algorithm attribute information of the second algorithm to the second algorithm management device; the second algorithm management device downloads the algorithm file from the first algorithm management device according to the algorithm encapsulation file download address and the algorithm attribute information.

12. An algorithm scheduling device, applied to a first algorithm management device deployed in a primary platform of a visual networking platform, characterized in that: include: A first receiving module is used to receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm; A first sending module, configured to send an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; A second receiving module is used to receive an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; The second sending module is used to send the AI ​​analysis task response to the whole network business management device.

13. An algorithm scheduling device, applied to a first algorithm management device deployed in a primary platform of a visual networking platform, characterized in that: including a transceiver and a processor; The transceiver is used to: receive an artificial intelligence AI algorithm order request sent by a network-wide service management device, wherein the AI ​​algorithm order request includes an algorithm identifier of a first algorithm; Sending an AI analysis task request to a computing network calling device according to the AI ​​algorithm order request, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; Receiving an AI analysis task response sent by the computing network calling device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; Send the AI ​​analysis task response to the whole network business management device.

14. An algorithm management device, the algorithm management device being a first algorithm management device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the algorithm scheduling method according to any one of claims 1 to 4 are implemented.

15. An algorithm scheduling device, applied to a computing network calling device deployed in a primary platform of a visual networking platform, characterized in that: include: A third receiving module is used to receive an AI analysis task request sent by the first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; A third sending module is used to send the AI ​​analysis task request to the computing power device; A fourth receiving module, configured to receive an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; The fourth sending module is used to send the AI ​​analysis task response to the first algorithm management device.

16. An algorithm scheduling device, applied to a computing network calling device deployed in a primary platform of a visual networking platform, characterized in that: including a transceiver and a processor; The transceiver is used to: receive an AI analysis task request sent by a first algorithm management device, wherein the AI ​​analysis task request includes an algorithm identifier of the first algorithm; Sending the AI ​​analysis task request to the computing device; receiving an AI analysis task response fed back by the computing device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; The AI ​​analysis task response is sent to the first algorithm management device.

17. A computing network calling device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the algorithm scheduling method as claimed in claim 5 or 6 when executing the program.

18. An algorithm scheduling device, applied to a network-wide service management device, characterized in that: include: A fifth sending module, configured to send an AI algorithm order request to the first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; a fifth receiving module, configured to receive an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; a sixth sending module, configured to send an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm; The sixth receiving module is used to receive the analysis results fed back by the computing power device.

19. An algorithm scheduling device, applied to a network-wide service management device, characterized in that: including a transceiver and a processor; The transceiver is used to: send an AI algorithm order request to a first algorithm management device, wherein the AI ​​algorithm order request includes an algorithm identifier of the first algorithm; receiving an AI analysis task response sent by the first algorithm management device, wherein the AI ​​analysis task response carries an algorithm service address of the first algorithm; Sending an AI analysis request to a computing device, wherein the AI ​​analysis request includes an algorithm service address of the first algorithm; Receive the analysis result fed back by the computing power device.

20. A network-wide service management device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the steps of the algorithm scheduling method as described in any one of claims 7 to 9 are implemented.

21. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method according to any one of claims 1 to 9 are implemented.

22. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 9.

Images