Scheduling method and device of edge computing task, equipment and medium

Abstract

The invention relates to the field of financial science and technology, and provides an edge computing task scheduling method and device, equipment and a medium, and the method comprises the steps: obtaining a task type corresponding to a target task, and determining a target task model corresponding to the target task according to a preset mapping relation between the task type and a task model; the storage position of the target task model is obtained, and the storage position comprises an online model library or an offline model library; obtaining the waiting time corresponding to the target task, and determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage position of the target task model; and according to the processing priority of the target task, scheduling the target task model from the online model library to execute the target task, so that the performance of edge computing equipment can be improved, and the satisfaction degree of a user to financial services is improved.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and specifically to a scheduling method, device, equipment and medium for edge computing tasks. Background Art

[0002] In the fintech sector, when using cloud computing to transfer customer data to a central location, privacy and security guidelines must be adhered to to reduce the chances of data theft. However, the possibility of data loss always exists, and edge computing enables banks and financial institutions to deploy applications across local branches, reducing the need for cloud computing and the chances of data loss.

[0003] Edge computing enables data storage and processing close to where the data is generated or collected, rather than on cloud servers located thousands of kilometers away. It will deploy services seamlessly and reliably at the edge by maintaining flexibility. It is safer than cloud computing, because the original data used by edge computing does not have to be transmitted, so when moving the processing result data from the edge to the cloud, there is no need to worry about the loss of original data involving privacy.

[0004] Edge computing devices mostly need to use artificial intelligence models when performing tasks. For example, in financial payment scenarios, ID card recognition models are often needed to identify user ID card information, and face recognition models are often used to identify user living information. However, compared with cloud servers, edge computing devices are generally smaller in scale and have limited storage resources. Therefore, edge computing devices can only load some models. Frequent model loading and unloading will affect device performance, resulting in reduced user satisfaction with financial services. Summary of the invention

[0005] The embodiments of the present application provide a scheduling method, apparatus, device and medium for edge computing tasks, which can improve the performance of edge computing devices in executing tasks and enhance user satisfaction with financial services.

[0006] In a first aspect, an embodiment of the present application provides a method for scheduling edge computing tasks, the method comprising the following steps: Get the task type corresponding to the target task; Determine the target task model corresponding to the target task according to the mapping relationship between the preset task type and the task model; Obtaining a storage location of the target task model, wherein the storage location includes an online model library or an offline model library; Obtaining the waiting time corresponding to the target task, and determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; In a case where the storage location of the target task model is the offline model library, loading the target task model from the offline model library to the online model library; According to the processing priority of the target task, the target task model is scheduled from the online model library to execute the target task.

[0007] In some embodiments, the number of the target task models corresponding to the target task is one or more; Determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model includes: Determine whether the waiting time is greater than a preset time threshold, and if the waiting time is greater than the time threshold, determine that the processing priority of the target task is the first priority; If the waiting time is less than or equal to the time threshold, determine whether the storage locations of all the target task models are the online model library, and if the storage locations of all the target task models are the online model library, determine that the processing priority of the target task is the second priority; If the storage location of at least one of the target task models is the offline model library, determining that the processing priority of the target task is the third priority; The first priority, the second priority, and the third priority are arranged in descending order of processing priority.

[0008] In some embodiments, before obtaining the waiting time corresponding to the target task, the method further includes: Receiving the target task; The target task is added to the task queue, and the timestamp of the target task being added to the task queue is recorded, and the waiting time corresponding to the target task is determined according to the timestamp.

[0009] In some embodiments, after determining the waiting time corresponding to the target task according to the timestamp, the method further includes: In a preset first key value table, the first key value corresponding to the target task and the waiting time are recorded, wherein the first key value is generated when the target task is received; All key values ​​in the first key value table are sorted in descending order of the waiting time.

[0010] In some embodiments, scheduling the target task model from the online model library to execute the target task according to the processing priority of the target task includes: If the target task has the highest processing priority in the task queue and the waiting time is longer than other tasks with the same processing priority, the target task model is scheduled from the online model library to execute the target task.

[0011] In some embodiments, the method further comprises: Obtaining usage frequency information corresponding to each task model in the online model library; Determining the task model with low usage frequency according to the usage frequency information corresponding to each task model; The task model with low usage frequency is uninstalled from the online model library.

[0012] In some embodiments, obtaining the storage location of the target task model includes: Searching for the target task model in the online model library; If the target task model is found in the online model library, determining that the storage location of the target task model is the online model library; If the target task model cannot be found in the online model library, the storage location of the target task model is determined to be the offline model library.

[0013] In a second aspect, an embodiment of the present application provides a scheduling device for edge computing tasks, the device comprising: A first acquisition module is used to acquire the task type corresponding to the target task; A first determination module is used to determine a target task model corresponding to the target task according to a preset mapping relationship between the task type and the task model; A second acquisition module is used to acquire the storage location of the target task model, wherein the storage location includes an online model library or an offline model library; A second determination module is used to obtain the waiting time corresponding to the target task, and determine the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; A loading module, used for loading the target task model from the offline model library to the online model library when the storage location of the target task model is the offline model library; An execution module is used to schedule the target task model from the online model library to execute the target task according to the processing priority of the target task.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program or instructions, and the processor implements the method provided in the first aspect of the embodiment of the present application when executing the computer program or instructions.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program or instructions are stored. When the computer program or instructions are executed by a processor, the method provided in the first aspect of the embodiment of the present application is implemented.

[0016] The solution provided by the embodiment of the present application first obtains the task type corresponding to the target task, and determines the target task model corresponding to the target task according to the mapping relationship between the preset task type and the task model; then obtains the storage location of the target task model, wherein the storage location includes an online model library or an offline model library; obtains the waiting time corresponding to the target task, and determines the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; when the storage location of the target task model is the offline model library, loads the target task model from the offline model library to the online model library; according to the processing priority of the target task, schedules the target task model from the online model library to execute the target task. According to the embodiment of the present application, the processing priority of the task to be executed by the edge computing device can be reasonably configured according to the waiting time of the task and the storage location of the task model required, thereby improving the performance of the edge computing device in executing tasks, and thus improving the user's satisfaction with financial services. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method for scheduling edge computing tasks provided in an embodiment of the present application; Figure 2 An optional schematic diagram of obtaining the storage location of the target task model provided in an embodiment of the present application; Figure 3 An optional schematic diagram for determining the processing priority of a target task provided in an embodiment of the present application; Figure 4 A schematic diagram of a scheduling device for edge computing tasks provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0019] It should be noted that, unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0020] First, some nouns involved in this application are analyzed: Artificial Intelligence (AI): It is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science. AI attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robots, language recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0021] Neural Networks (NN): A complex network system formed by a large number of simple processing units (called neurons) that are widely interconnected. It reflects many basic characteristics of human brain functions and is a highly complex nonlinear dynamic learning system. Neural networks have large-scale parallel, distributed storage and processing, self-organization, self-adaptation and self-learning capabilities, and are particularly suitable for processing imprecise and fuzzy information processing problems that require considering many factors and conditions at the same time.

[0022] Edge computing: Edge computing enables data storage and processing close to where the data is generated or collected, rather than on a server located thousands of kilometers away. It will deploy services seamlessly and reliably at the edge by maintaining flexibility. It is more secure than cloud computing because there is no need to transfer data. Therefore, there is no worry about data loss when moving data from the edge to the cloud. As a result, it provides faster insights and business benefits, reduces response time, and improves overall service.

[0023] Multi-model inference: refers to the use of multiple AI models to perform a task. For example, to identify the number of women in a picture, to complete this inference, the DNN model is first used to identify the location and number of faces, and then the gender recognition network is used to identify the gender of each face.

[0024] In the fintech sector, when using cloud computing to transfer customer data to a central location, privacy and security guidelines must be adhered to to reduce the chances of data theft. However, the possibility of data loss always exists, and edge computing enables banks and financial institutions to deploy applications across local branches, reducing the need for cloud computing and the chances of data loss.

[0025] Edge computing devices mostly need to use artificial intelligence models when performing tasks. For example, in financial payment scenarios, ID card recognition models are often needed to identify user ID card information, and face recognition models are often used to identify user living information. However, compared with cloud servers, edge computing devices are generally smaller in scale and have limited storage resources. Therefore, edge computing devices can only load some models. Frequent model loading and unloading will affect device performance, resulting in reduced user satisfaction with financial services.

[0026] Most of the target tasks performed by edge computing devices require the use of artificial intelligence models. However, compared with cloud servers, edge computing devices are generally smaller in scale and have limited storage resources. Therefore, edge computing devices can only load part of the models. Frequent model loading and unloading will affect the overall performance.

[0027] In view of this, embodiments of the present application provide a scheduling method, apparatus, device and medium for edge computing tasks, which can improve the performance of edge computing devices in executing tasks.

[0028] The scheduling method for edge computing tasks, the scheduling device for edge computing tasks, the electronic device and the storage medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the scheduling method for edge computing tasks in the embodiments of the present application is described.

[0029] The scheduling method of edge computing tasks provided in the embodiment of the present application is applied to edge computing devices, wherein the edge computing device can be a terminal or a server, specifically, it can be software running in the terminal or the server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server can be configured as an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms; the software can be an application of the scheduling method for edge computing tasks, etc., but is not limited to the above forms.

[0030] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0031] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on the attribute information or attribute information set of the object (such as a user, etc.) and other data related to the characteristics of the object, the permission or consent of the corresponding object will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. In addition, when the embodiment of the present application needs to obtain the attribute information of the object, the separate permission or separate consent of the corresponding object will be obtained through a pop-up window or jump to a confirmation page. After clearly obtaining the separate permission or separate consent of the corresponding object, the relevant data of the object necessary for the normal operation of the embodiment of the present application will be obtained.

[0032] See also Figure 1 , Figure 1A flow chart of a scheduling method for edge computing tasks provided in an embodiment of the present application is provided. The method includes the following steps S110-S160, and each step is introduced in turn below.

[0033] Step S110, obtaining the task type corresponding to the target task.

[0034] In some embodiments, a task type library is preconfigured. For example, the task type library has Types, this task type library is represented as follows: ; in, Represents the i-th task type.

[0035] After receiving the target task to be executed, the edge computing device searches for the task type corresponding to the target task in the task type library, so as to determine the task module to be used for the target task according to the task type in the subsequent steps.

[0036] Exemplarily, the task types in the task type library may include: intelligent financial customer service tasks, financial anti-fraud detection tasks, pre-loan review tasks, credit limit prediction tasks, etc. The embodiments of the present application do not impose too many restrictions on the task types.

[0037] Step S120: Determine the target task model corresponding to the target task according to the preset mapping relationship between the task type and the task model.

[0038] In some embodiments, the model library is preconfigured, for example, the number of models in the model library is , this model library is represented as follows:

[0039] in, represents the i-th model.

[0040] In specific implementation, the mapping relationship between the task type and the task model can be pre-configured on the edge computing device. According to the mapping relationship, the task model used when executing a task of a certain task type can be determined.

[0041] It is understandable that a task type can correspond to one or more task models, that is, one or more task types may be used when performing a task of a certain task type. For example, in financial anti-fraud detection, multiple models are used at the same time: face detection model, age recognition model, gender recognition model, expression recognition model, scene recognition model, etc.

[0042] It is understandable that according to the mapping relationship between the preset task type and the task model, the task type used by the target task can be found. These task types can be called target task models, and the number of target task types can be one or more.

[0043] Step S130, obtaining the storage location of the target task model, wherein the storage location includes an online model library or an offline model library.

[0044] It is understandable that the memory of edge computing devices is limited and cannot load all models, so the model library can be divided into an "online model library" loaded in the memory. ” and the “offline model library” stored on disk ". In this embodiment of the present application, for each target task model corresponding to the target task, it is determined whether the target task model is stored in an online model library or an offline model library.

[0045] It can be understood that the models in the online model library can be called online models, and the models in the offline model library can be called offline models.

[0046] See also Figure 2 , Figure 2 An optional schematic diagram of obtaining the storage location of the target task model provided in the embodiment of the present application. Figure 2 As shown, obtaining the storage location of the target task model includes the following steps S131-S133: Step S131, searching for a target task model in an online model library; Step S132: if the target task model is found in the online model library, then the storage location of the target task model is determined to be the online model library; Step S133: If the target task model cannot be found in the online model library, the storage location of the target task model is determined to be the offline model library.

[0047] When it is necessary to obtain the storage location of the target task model, first search the online model library to determine whether the target task model is stored in the online model library. If not, the storage location of the target task model is determined to be the offline model library.

[0048] Step S140, obtaining the waiting time corresponding to the target task, and determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model.

[0049] It is understandable that if the task models involved in the target task are all online models, then this target task can be processed first, which can minimize the number of times the model is loaded and unloaded. At the same time, in order to ensure the service quality of the target task, it should also be avoided that the target task waits for too long for processing, so the embodiment of the present application also includes the waiting time into the consideration factor for determining the processing priority of the target task.

[0050] That is to say, in the embodiment of the present application, the processing priority of the target task is determined based on the waiting time corresponding to the target task and the storage location of the target task model used by the target task.

[0051] In some embodiments, the method for setting the processing priority of the target task is as follows:

[0052] in, Indicates the target task There are three processing priorities: "3", "2" and "1". The higher the processing priority of a task, the faster it will be processed.

[0053] If the target task Waiting time Exceeds a pre-set threshold , then set the target task The processing priority is the highest priority "3"; If the target task Type of task Required sub-model library All models are online model library In the model, then set the target task The processing priority is medium priority "2"; Otherwise, set the target task The processing priority is the lowest priority "1".

[0054] In some embodiments, the number of target task models corresponding to the target task is one or more; correspondingly, see Figure 3 , Figure 3 An optional schematic diagram for determining the processing priority of a target task provided in an embodiment of the present application is as follows: Figure 3 As shown, according to the waiting time corresponding to the target task and the storage location of the target task model, the processing priority of the target task is determined, including the following steps S141-S145: Step S141, determining whether the waiting time is greater than a preset time threshold; Step S142, if the waiting time is greater than the time threshold, determining the processing priority of the target task as the first priority; Step S143, if the waiting time is less than or equal to the time threshold, determine whether the storage locations of all target task models are online model libraries; Step S144, if the storage locations of all target task models are online model libraries, determine that the processing priority of the target task is the second priority; Step S145: if the storage location of at least one target task model is an offline model library, determine that the processing priority of the target task is the third priority.

[0055] Among them, the first priority, the second priority, and the third priority are arranged in order from high to low processing priority.

[0056] In an embodiment of the present application, the waiting time of the target task is taken as the primary consideration for determining the processing priority. If the waiting time of the target task has exceeded the preset time threshold, the target task is taken as the first priority task to avoid the waiting time of the target task exceeding the preset time threshold too much, thereby ensuring the service quality of the tasks to be executed in the task queue; secondly, the processing priority is determined according to the storage location of the target task model involved in the target task. If all target task models are online models, the target task can be taken as the second priority task. For target tasks with at least one target task model as an offline model, it is processed as a third priority task and deferred. This can optimize the processing efficiency of multiple tasks to be executed in the task queue and improve the task processing performance of the edge computing device.

[0057] Step S150: when the storage location of the target task model is an offline model library, the target task model is loaded from the offline model library to the online model library.

[0058] In an embodiment of the present application, if the task model used for the current target task is an offline model, the offline model is loaded into the memory, that is, the target task model is loaded from the offline model library to the online model library. In this way, when executing the target task, the corresponding target task model can be directly obtained from the online model library, and the target task model can be used to complete the processing of the target task.

[0059] Step S160, according to the processing priority of the target task, scheduling the target task model from the online model library to execute the target task.

[0060] In an embodiment of the present application, the edge computing device determines the currently executed task based on the processing priority of each task to be executed in the task queue. If it is determined that the currently executed task is the above-mentioned target task, the target task model is scheduled from the online model library to execute the target task.

[0061] In some embodiments, according to the processing priority of the target task, the target task model is scheduled from the online model library to execute the target task, including: if the target task has the highest processing priority in the task queue and the waiting time is longer than other tasks with the same processing priority, then the target task model is scheduled from the online model library to execute the target task.

[0062] For example, the target task The processing priority of the task is the highest priority "3", and the priorities of other tasks in the task queue are lower than "3", then it is determined that the currently executed task is the above target task , schedule target tasks from the online model library The target task model involved is used to perform the target task.

[0063] For example, the target task ,Task The processing priority of the target task is "2", while the priority of other tasks in the task queue is "1". The waiting time is longer than the task The waiting time is determined to be the target task. , schedule target tasks from the online model library The target task model involved is used to perform the target task.

[0064] In some embodiments, before obtaining the waiting time corresponding to the target task, the following steps S101-S102 are also included: Step S101, receiving a target task; Step S102, adding the target task to the task queue, recording the timestamp of when the target task is added to the task queue, and determining the waiting time corresponding to the target task according to the timestamp.

[0065] It is understandable that after receiving the task to be executed, the edge computing device adds the task to be executed to the task queue for caching, and records the timestamp when the task is added to the task queue, so as to determine the waiting time of the task according to the timestamp in the subsequent steps.

[0066] In some embodiments, after determining the waiting time corresponding to the target task according to the timestamp, the following steps S201-S202 are also included: Step S201, recording the first key value and waiting time corresponding to the target task in a preset first key value table, wherein the first key value is generated when receiving the target task; Step S202: sort all key values ​​in the first key value table in descending order of waiting time.

[0067] In an embodiment of the present application, the first key value table is used to record the waiting time of the task to be executed, which is determined according to the timestamp when the task is added to the task queue and is continuously updated. The first key value table has multiple key values ​​corresponding to the tasks to be executed, and the key value is used to identify the tasks to be executed. It should be understood that the key value corresponding to the target task can be referred to as the first key value here, and the first key value can be a unique identifier generated when the edge computing device receives the target task.

[0068] In the embodiment of the present application, the key values ​​in the first key value table are arranged in descending order according to the waiting time, so that it is possible to quickly know which tasks to be executed have a waiting time exceeding a preset time threshold.

[0069] In some embodiments, the method of the embodiment of the present application further includes the following steps S301-S303: Step S301, obtaining usage frequency information corresponding to each task model in the online model library; Step S302, determining a task model with low usage frequency according to usage frequency information corresponding to each task model; Step S303: Uninstalling the task models with low usage frequency from the online model library.

[0070] In the embodiment of the present application, considering that the memory of the edge computing device is very limited, it is necessary to eliminate some infrequently used online models in a timely manner. In order to minimize the number of times the model is loaded and unloaded, the embodiment of the present application gives priority to eliminating online models with low model usage frequency.

[0071] In some embodiments, a visit count record table is set to count the number of visits to all models. Assuming that the number of models in the model library is , then the access count record table can be expressed as follows:

[0072] in Represents the cumulative number of visits to the i-th model.

[0073] In this embodiment, the usage frequency information corresponding to the task model is the cumulative number of accesses. In specific implementation, the cumulative number of accesses corresponding to each task model in the online model library is obtained, a cumulative number of accesses lower limit is set, and the task model whose cumulative number of accesses is lower than the cumulative number of accesses lower limit is uninstalled from the online model library, thus saving the memory space of the edge computing device.

[0074] In other embodiments, the usage frequency information corresponding to the task model is the number of accesses to the model in a specific time period. In specific implementation, the number of accesses to each model in the model library in a specific time period such as the last month or the last week can be counted, a lower limit of the number of accesses can be set, and the task models whose number of accesses in the specific time period is lower than the lower limit of the number of accesses can be uninstalled from the online model library.

[0075] The solution provided by the embodiment of the present application first obtains the task type corresponding to the target task, and determines the target task model corresponding to the target task according to the mapping relationship between the preset task type and the task model; then obtains the storage location of the target task model, wherein the storage location includes an online model library or an offline model library; obtains the waiting time corresponding to the target task, and determines the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; when the storage location of the target task model is an offline model library, loads the target task model from the offline model library to the online model library; according to the processing priority of the target task, schedules the target task model from the online model library to execute the target task. According to the embodiment of the present application, the processing priority of the task to be executed by the edge computing device can be reasonably configured according to the waiting time of the task and the storage location of the required task model, thereby improving the performance of the edge computing device in executing tasks.

[0076] The following describes the scheduling method for edge computing tasks provided in the embodiment of the present application in conjunction with specific application scenarios.

[0077] The embodiment of the present application can be applied to the loan review scenario in the field of financial technology. In specific implementation, the user's loan application request is obtained through the business terminal, and the loan application request includes the user's live collection information, ID card image and user basic information; the business terminal requests the loan review task from the edge computing device, and uploads the user's live collection information, ID card image and personal information form to the edge computing device; the edge computing device responds to receiving the loan review task and adds the loan review task to the task queue; the edge computing device obtains the storage location of the face recognition model, ID card image text recognition model and credit assessment model required for the loan review task, wherein the face recognition model and ID card image text recognition model are online models, and the credit assessment model is an offline model; the edge computing device sets the above loan review task to the second priority, and loads the offline credit assessment model into the online model library; when it is the turn to execute the above loan review task, the face recognition model, ID card image text recognition model and credit assessment model are used to execute the above user's loan review task, and return the loan review result to the business terminal.

[0078] It is to be understood that, although the steps in the above-mentioned various flow charts are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in the present embodiment, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the above-mentioned flow charts can include a plurality of steps or a plurality of stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps.

[0079] See also Figure 4 , Figure 4 A schematic diagram of a scheduling device for edge computing tasks provided in an embodiment of the present application, the device comprising: The first acquisition module 1100 is used to acquire the task type corresponding to the target task; A first determination module 1200 is used to determine a target task model corresponding to a target task according to a preset mapping relationship between a task type and a task model; A second acquisition module 1300 is used to acquire a storage location of a target task model, wherein the storage location includes an online model library or an offline model library; The second determination module 1400 is used to obtain the waiting time corresponding to the target task, and determine the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; A loading module 1500 is used to load the target task model from the offline model library to the online model library when the storage location of the target task model is the offline model library; The execution module 1600 is used to schedule the target task model from the online model library to execute the target task according to the processing priority of the target task.

[0080] In some embodiments, the number of target task models corresponding to the target task is one or more; the second determination module 1400 is specifically used to: Obtain the waiting time corresponding to the target task, and determine whether the waiting time is greater than a preset time threshold. If the waiting time is greater than the time threshold, determine that the processing priority of the target task is the first priority; If the waiting time is less than or equal to the time threshold, determine whether the storage locations of all target task models are online model libraries. If the storage locations of all target task models are online model libraries, determine that the processing priority of the target task is the second priority. If the storage location of at least one target task model is an offline model library, determining that the processing priority of the target task is the third priority; Among them, the first priority, the second priority, and the third priority are arranged in order from high to low processing priority.

[0081] In some embodiments, the above device further includes a task cache module, and the task cache module is used to: Receive target tasks; Add the target task to the task queue, record the timestamp when the target task is added to the task queue, and determine the waiting time corresponding to the target task based on the timestamp.

[0082] In some embodiments, the task cache module is further configured to: In the preset first key value table, record the first key value and waiting time corresponding to the target task, wherein the first key value is generated when receiving the target task; Sort all key values ​​in the first key value table in descending order of waiting time.

[0083] In some embodiments, the execution module 1600 is specifically used to: If the target task has the highest processing priority in the task queue and its waiting time is longer than other tasks with the same processing priority, the target task model is scheduled from the online model library to execute the target task.

[0084] In some embodiments, the above device further includes an unloading module, and the unloading module is used to: Obtain usage frequency information corresponding to each task model in the online model library; According to the usage frequency information corresponding to each task model, determine the task model with low usage frequency; Unload the task models with low usage frequency from the online model library.

[0085] In some embodiments, the second acquisition module 1300 is specifically used for: Find the target task model in the online model library; If the target task model is found in the online model library, the storage location of the target task model is determined to be the online model library; If the target task model cannot be found in the online model library, the storage location of the target task model is determined to be the offline model library.

[0086] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] The specific implementation of the scheduling device for the edge computing task is basically the same as the specific implementation of the scheduling method for the above-mentioned edge computing task. The two have the same technical principles and the same beneficial effects. In order to avoid repetition of content, they will not be repeated here.

[0088] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program or instruction, and the processor implements the scheduling method of the edge computing task when executing the computer program or instruction. The electronic device can be a server or a smart terminal.

[0089] See also Figure 5 , Figure 5 The hardware structure of the electronic device is shown in FIG. 1 , and the electronic device includes: The processor 1010 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 1020 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1020 can store an operating system and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1020, and the processor 1010 calls the scheduling method for executing the edge computing task of the embodiment of this application; Input / output interface 1030, used to implement information input and output; The communication interface 1040 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.); Bus 1050 , which transmits information between various components of the device (e.g., processor 1010 , memory 1020 , input / output interface 1030 , and communication interface 1040 ); The processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 are connected to each other in communication within the device via the bus 1050 .

[0090] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program or instruction, and when the computer program or instruction is executed by a processor, the scheduling method of the above-mentioned edge computing task is implemented.

[0091] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0092] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0093] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0094] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0095] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0096] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0097] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0098] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0099] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0100] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0102] The preferred embodiments of the present application are described above with reference to the accompanying drawings, but the scope of the rights of the present application is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present application should be within the protection scope of the present application.

Claims

1. A scheduling method for edge computing tasks, characterized in that: The method comprises the following steps: Get the task type corresponding to the target task; Determine the target task model corresponding to the target task according to the mapping relationship between the preset task type and the task model; Obtaining a storage location of the target task model, wherein the storage location includes an online model library or an offline model library; Obtaining the waiting time corresponding to the target task, and determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; In a case where the storage location of the target task model is the offline model library, loading the target task model from the offline model library to the online model library; According to the processing priority of the target task, the target task model is scheduled from the online model library to execute the target task.

2. The method according to claim 1, characterized in that The number of the target task models corresponding to the target task is one or more; Determining the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model includes: Determine whether the waiting time is greater than a preset time threshold, and if the waiting time is greater than the time threshold, determine that the processing priority of the target task is the first priority; If the waiting time is less than or equal to the time threshold, determine whether the storage locations of all the target task models are the online model library, and if the storage locations of all the target task models are the online model library, determine that the processing priority of the target task is the second priority; If the storage location of at least one of the target task models is the offline model library, determining that the processing priority of the target task is the third priority; The first priority, the second priority, and the third priority are arranged in descending order of processing priority.

3. The method according to claim 1, characterized in that Before obtaining the waiting time corresponding to the target task, the method further includes: Receiving the target task; The target task is added to the task queue, and the timestamp of the target task being added to the task queue is recorded, and the waiting time corresponding to the target task is determined according to the timestamp.

4. The method according to claim 3, characterized in that After determining the waiting time corresponding to the target task according to the timestamp, the method further includes: In a preset first key value table, the first key value corresponding to the target task and the waiting time are recorded, wherein the first key value is generated when the target task is received; All key values ​​in the first key value table are sorted in descending order of the waiting time.

5. The method according to claim 1, characterized in that The step of scheduling the target task model from the online model library to execute the target task according to the processing priority of the target task includes: If the target task has the highest processing priority in the task queue and the waiting time is longer than other tasks with the same processing priority, the target task model is scheduled from the online model library to execute the target task.

6. The method according to claim 1, characterized in that The method further comprises: Obtaining usage frequency information corresponding to each task model in the online model library; Determining the task model with low usage frequency according to the usage frequency information corresponding to each task model; The task model with low usage frequency is uninstalled from the online model library.

7. The method according to claim 1, characterized in that The obtaining the storage location of the target task model includes: Searching for the target task model in the online model library; If the target task model is found in the online model library, determining that the storage location of the target task model is the online model library; If the target task model cannot be found in the online model library, the storage location of the target task model is determined to be the offline model library.

8. A scheduling device for edge computing tasks, characterized in that: The device comprises: A first acquisition module is used to acquire the task type corresponding to the target task; A first determination module is used to determine a target task model corresponding to the target task according to a preset mapping relationship between the task type and the task model; A second acquisition module is used to acquire the storage location of the target task model, wherein the storage location includes an online model library or an offline model library; A second determination module is used to obtain the waiting time corresponding to the target task, and determine the processing priority of the target task according to the waiting time corresponding to the target task and the storage location of the target task model; A loading module, used for loading the target task model from the offline model library to the online model library when the storage location of the target task model is the offline model library; An execution module is used to schedule the target task model from the online model library to execute the target task according to the processing priority of the target task.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program or instructions, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program or instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program or instructions, and when the computer program or instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images