Serverless processing method, device and network equipment

Abstract

The present invention discloses a serverless processing method, device and network equipment, which belongs to the field of serverless computing technology and solves the problem of low serverless processing efficiency of cloud computing platforms in related technologies. The method includes: receiving tasks uploaded by user devices; obtaining the current required container resource amount of the task obtained by analysis and the current available serverless container resource amount of the node device obtained by monitoring, wherein the node device is the mobile edge computing MEC node device closest to the user device; determining the comparison result of the current required container resource amount and the current available serverless container resource amount; selecting a task processing location between the server of the cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location. The method can improve the efficiency of serverless task processing.

Description

Technical Field

[0001] The present invention relates to the field of serverless computing technology, and in particular to a serverless processing method, device and network equipment. Background Art

[0002] Serverless computing is a cloud computing method that hosts applications on infrastructure that does not require end-user management. Serverless applications allow developers to focus on code rather than infrastructure configuration and management, which can speed up development and release cycles and allow better and more efficient expansion.

[0003] At present, after the developer submits the code program of the task, the serverless cloud computing platform will perform a function cold start when the function in the code is called for the first time. Function cold start is the process from obtaining the function execution engine to creating the function instance. Since the serverless function needs to open up all the network environments involved through the function cold start, the function can provide services. Therefore, the processing time of serverless task processing on the cloud computing platform is long and the processing efficiency is low. Summary of the invention

[0004] To this end, the present invention provides a serverless processing method, apparatus and network device to solve the problem of low efficiency of serverless task processing in related technologies.

[0005] In order to achieve the above-mentioned objectives, the first aspect of the present invention provides a serverless processing method, which includes: receiving a task uploaded by a user device; obtaining the current required container resource amount of the task obtained by analysis and the current available serverless container resource amount of the current node device obtained by monitoring, wherein the current node device is a mobile edge computing MEC node device closest to the user device; determining a comparison result of the current required container resource amount and the current available serverless container resource amount; and selecting a task processing location between a server in a cloud data center and the current node device according to the comparison result, so as to perform serverless processing on the task at the task processing location.

[0006] Among them, according to the comparison result, a task processing location is selected between the server of the cloud data center and the current node device to perform serverless processing on the task at the task processing location, including: if the currently required container resource amount is greater than or equal to the currently available serverless container resource amount, uploading the task to the cloud data center to perform serverless processing on the task in the cloud data center; if the currently required container resource amount is less than the currently available serverless container resource amount, performing serverless processing on the task on the current node device.

[0007] The serverless processing of the task at the task processing location includes: determining a resource pre-start requirement at the processing location according to a status of the task; wherein the status of the task is used to indicate a demand trend of the task for serverless container resources; and serverlessly processing the task at the processing location according to the resource pre-start requirement.

[0008] Among them, determining the resource pre-start requirement based on the status of the task includes: if the status of the task is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the status of the task is in a period of stable resource demand or a period of declining resource demand, then determining the container resource pre-start requirement is that container resource pre-start is not required.

[0009] Among them, if the task processing location is in this node device and the resource pre-start requirement is to need to execute container resource pre-start, then the task is serverlessly processed at the processing location according to the resource pre-start requirement, including: providing the task with server-free container resources with a container resource amount equal to the currently required container resource amount on the local node device, and processing the task; if it is monitored that the remaining resource amount of the server-free container resources of the local node device is greater than or equal to the specified number of container resources, then pre-starting the server-free container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the server-free container resources that have been pre-started by this node device are directly used to expand the container; wherein, the specified number of container resources is less than the current required amount of container resources for the task.

[0010] The method further includes: if it is monitored that the remaining resource amount of the serverless container resources of the node device is less than the specified container resource amount, a resource pre-start request is sent to the cloud data center; wherein the resource pre-start request is used to request the cloud data center to pre-start a serverless container resource with a resource amount equal to the container resource difference for the task; wherein the container resource difference is the difference between the specified container resource amount and the remaining resource amount; during the processing of the task, if it is monitored that the remaining resource amount is less than the demand for serverless container resources by the task, the task and the processing result of the completed part of the task are uploaded to the cloud data center; wherein the cloud data center is used to store the processing result of the completed part, and call the serverless container resource with a pre-started container resource amount equal to the container resource difference, continue to process the unfinished part of the task, and store the processing result of the unfinished part, so as to obtain the stored processing results of all the tasks according to the processing results of the completed part and the processing results of the unfinished part.

[0011] Among them, if the task processing location is in the local node device and the resource pre-start requirement is that no container resource pre-start is required, then the task is serverlessly processed at the processing location according to the resource pre-start requirement, including: providing the task with server-less container resources with a container resource amount equal to the currently required container resource amount in the local node device, and processing the task; uploading and storing the processing result of the task to the reserved storage space of the cloud data center.

[0012] Among them, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is to pre-start container resources, then the task is serverlessly processed at the processing location according to the resource pre-start requirement, including: allocating serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and processing the task; pre-starting serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the serverless container resources pre-started in the cloud data center can be directly used to expand the container.

[0013] Among them, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is that no container resource pre-start is required, then the task is serverlessly processed at the processing location according to the resource pre-start requirement, including: allocating serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and processing the task; storing the processing result of the task in a reserved storage space of the cloud data center.

[0014] A second aspect of the present invention provides a serverless processing method, which includes: receiving a task uploaded by a mobile edge computing MEC node device; wherein the MEC node device is the MEC node device that is closest to the user device to which the task belongs, and the task is uploaded under the condition of satisfying the task upload condition; the task upload condition includes: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device; and performing serverless processing on the task.

[0015] Among them, the serverless processing of the task includes: determining the resource pre-start requirement according to the state of the task; wherein the state of the task is used to indicate the demand trend of the task for serverless container resources; and serverlessly processing the task according to the resource pre-start requirement.

[0016] Among them, determining the resource pre-start requirement based on the status of the task includes: if the status of the task is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the status of the task is in a period of stable resource demand or a period of declining resource demand, then determining the container resource pre-start requirement is that container resource pre-start is not required.

[0017] Among them, if the resource pre-start requirement is to need to execute container resource pre-start, then the task is serverlessly processed according to the resource pre-start requirement, including: allocating serverless container resources to the task according to the currently required container resource quantity and processing the task; pre-starting serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the pre-started serverless container resources of this cloud data center can be directly used to expand the container.

[0018] Among them, if the resource pre-start requirement is that there is no need to perform container resource pre-start, then the task is serverlessly processed according to the resource pre-start requirement, including: if the resource pre-start requirement is that there is no need to pre-start container resources, serverless container resources are allocated to the task according to the currently required amount of container resources and the task is processed, and the processing result of the task is stored in the reserved storage space of this cloud data center.

[0019] The method further includes: receiving a resource pre-start request from the MEC node device; wherein the resource pre-start request is used to request to pre-start a serverless container resource for the task whose resource amount is equal to the container resource difference; wherein the container resource difference is the difference between the specified number of container resources and the remaining resource amount of the serverless container resource of the MEC node device; pre-starting the serverless container resource for the task according to the container resource difference; in response to the task uploaded by the MEC node device and the processing result of the completed part of the task, storing the processing result of the completed part, and using the pre-started serverless container resource whose resource amount is equal to the container resource difference to process the unfinished part of the task to obtain the processing result of the unfinished part; storing the processing result of the unfinished part, and obtaining the stored processing result of all the processing results of the task based on the processing result of the completed part and the processing result of the unfinished part.

[0020] The third aspect of the present invention provides a serverless processing device, which includes: a task receiving module, used to receive tasks uploaded by user devices; an information acquisition module, used to obtain the current required container resource amount of the task obtained by analysis and the current available serverless container resource amount of the current node device obtained by monitoring, wherein the current node device is a mobile edge computing MEC node device closest to the user device; a result determination module, used to determine the comparison result of the current required container resource amount and the current available serverless container resource amount; a processing location determination module, used to select a task processing location between the server of the cloud data center and the current node device according to the comparison result, so as to perform serverless processing on the task at the task processing location.

[0021] The fourth aspect of the present invention provides a serverless processing device, which includes: a receiving module for receiving tasks uploaded by a mobile edge computing MEC node device; wherein the MEC node device is the MEC node device that is closest to the user device to which the task belongs, and the task is uploaded when the task upload condition is met; the task upload condition includes: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device; a processing module for performing serverless processing on the task.

[0022] A fifth aspect of the present invention provides a network device, comprising: one or more processors; a memory on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-9, or the method according to any one of claims 10-15; one or more I / O interfaces, connected between the processor and the memory, and configured to implement information interaction between the processor and the memory.

[0023] The present invention has the following advantages: a user device can upload a task to a MEC node device that is closest to the user device. After receiving the task uploaded by the user device, the MEC node device compares the container resources currently required for the task with the currently available serverless container resources of the MEC node device, and selects a processing location for the task between a server in a cloud data center and the node device according to the comparison result, so that the task can be processed at different processing locations. By processing the task at the nearest MEC node, the task processing interaction process can be reduced, the task processing time can be reduced, and the task processing efficiency can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0025] Figure 1 A flowchart of a server-free processing method provided by an embodiment of the present invention;

[0026] Figure 2 A flowchart of a server-free processing method provided by another embodiment of the present invention;

[0027] Figure 3 A schematic diagram of the structure of a serverless processing device provided by an embodiment of the present invention;

[0028] Figure 4 A schematic diagram of the structure of a serverless processing device provided by another embodiment of the present invention;

[0029] Figure 5 A schematic diagram of a network architecture for serverless processing according to an exemplary embodiment of the present invention;

[0030] Figure 6 A schematic diagram of a network architecture of a network device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0032] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0033] The terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. As used in the present invention, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0034] When the terms “comprising” and / or “made of…” are used in the present invention, it specifies the existence of features, integers, steps, operations, elements and / or components, but does not exclude the existence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0035] Embodiments of the present invention may be described with reference to plan views and / or cross-sectional views by way of ideal schematic views of the present invention. Therefore, the exemplary illustrations may be modified according to manufacturing techniques and / or tolerances.

[0036] Unless otherwise defined, all terms (including technical and scientific terms) used in the present invention have the same meaning as those commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present invention, and will not be interpreted as having an idealized or overly formal meaning unless the present invention clearly defines it as such.

[0037] In a first aspect, an embodiment of the present invention provides a server-free processing method.

[0038] Figure 1 1 is a flow chart showing a serverless processing method according to an embodiment of the present invention. Figure 1 As shown, the server-free processing method in the embodiment of the present invention includes the following steps.

[0039] S110, receiving a task uploaded by a user device.

[0040] In this step, the tasks uploaded by the user device may include multiple types, for example, the task may include at least one of: computing offloading tasks, collaborative computing tasks and intranet data processing tasks. Among them, computing offloading tasks refer to the mobile edge computing environment, which allows user devices with insufficient processing power and limited resources to write computing tasks to the MEC server, and return them to the user device after the task is completed; collaborative computing tasks refer to multiple user devices can cooperate with each other on the MEC server to jointly perform computing tasks; intranet data processing tasks refer to the MEC server storing, interacting and processing data containing local network status information transmitted between user devices or between user devices and core network data, so as to achieve the purpose of dynamically optimizing and improving user experience.

[0041] It should be understood that the user devices in the embodiments of the present application may include but are not limited to: personal computers, smart phones, tablet computers, personal digital assistants, etc.

[0042] S120, obtaining the current required container resource amount of the task obtained through analysis and the current available serverless container resource amount of the node device obtained through monitoring; wherein the node device is the mobile edge computing MEC node device closest to the user device.

[0043] In the embodiments of the present application, a container is a technology for packaging and running operating system (Windows operating system and Linux operating system) applications across different environments locally and in the cloud. Container resources can be understood as the hardware resources required when using containers. In some embodiments, a container can be an independent software package, which contains the code, system tools, and system libraries required for software operation.

[0044] In some embodiments, the container is pre-configured with a variety of computing resources so that the container can perform tasks according to the configured computing resources, that is, the container resources include the computing resources required to perform the tasks. It should be understood that the present application does not specifically limit the type of computing resources, and the computing resources may include at least one of: a central processing unit (CPU), memory, disk I / O, and network throughput.

[0045] In the embodiment of the present application, various tasks are executed by containers, so that the execution of different tasks can be effectively isolated from resources.

[0046] S130, determining a comparison result between a currently required amount of container resources and a currently available amount of serverless container resources.

[0047] In this step, the comparison result may be that the currently required container resource amount is greater than or equal to the currently available serverless container resource amount, or the comparison result may be that the currently required container resource amount is less than the currently available serverless container resource amount.

[0048] S140, selecting a task processing location between a server in the cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location.

[0049] In this step, the cloud data center is the cloud computing data center. Cloud computing is a internet Through cloud computing, shared hardware and software resources and information can be provided to computers and other devices on demand. Cloud computing data centers are data centers based on cloud computing architecture. Cloud computing data centers can highly virtualize various IT equipment, turning them into corresponding computing resources, network resources, storage resources, etc., and use corresponding technologies to enable these resources to be automatically allocated to the outside world according to user needs on the Internet.

[0050] In the embodiment of the present invention, the cloud data center may also be referred to as a cloud computing data center or a cloud computing platform. The cloud data center can be accessed through software such as a browser or other Web Services To access, the software and data involved in cloud computing in a cloud data center can be stored on servers in the cloud data center.

[0051] According to the serverless processing method of an embodiment of the present invention, as the MEC node device closest to the user device to which the task belongs, after receiving the task uploaded by the user device, the container resources currently required for the task are compared with the currently available serverless container resources of the MEC node device, and the processing location of the task is selected from the server in the cloud data center and the node device according to the comparison result, so that the task can be processed at different processing locations. By processing the task at the nearest MEC node, the task processing interaction process can be reduced, the task processing time can be reduced, and the task processing efficiency can be improved.

[0052] In some embodiments, step S140 may specifically include the following steps.

[0053] S11, if the currently required container resource amount is greater than or equal to the currently available serverless container resource amount, the task is uploaded to the cloud data center to be processed by the serverless processing of the cloud data center.

[0054] S12: If the currently required container resource amount is less than the currently available serverless container resource amount, the serverless processing task is performed by the node device.

[0055] In this embodiment, if the amount of container resources currently required for the task is less than the amount of serverless container resources currently available on the MEC node device, the task is directly processed serverlessly on the MEC node device; if the amount of container resources currently required for the task is greater than or equal to the amount of serverless container resources currently available on the MEC node device, the task is uploaded to the serverless of the cloud data center for processing; thereby, tasks are processed at different processing locations. By processing tasks at the nearest MEC node, the task processing interaction process can be reduced, the task processing time can be reduced, and the task processing efficiency can be improved.

[0056] In some embodiments, the serverless processing of the task at the task processing location in step S140 may specifically include: S21, determining the resource pre-start requirement at the processing location based on the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; S22, performing serverless processing of the task at the processing location based on the resource pre-start requirement.

[0057] In some embodiments, the container resource demand trend may be, for example, one of: a demand increase period, a demand decay period, and a demand stability period. The demand increase period is used to indicate a trend that the demand for serverless container resources by a task gradually increases over time, the demand decay period is used to indicate a trend that the demand for serverless container resources by a task gradually decreases over time, and the demand stability period is used to indicate a trend that the demand for serverless container resources by a task remains unchanged over time.

[0058] In this embodiment, based on the processing of tasks at different processing locations, the container resource pre-start strategy can be further determined according to the container resource demand trend represented by the task status, that is, whether to pre-start the container resources, thereby realizing flexible execution of the container resource pre-start strategy according to the demand trend of the task for serverless container resources.

[0059] In some embodiments, the above step S22 may specifically include: if the task status is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the task status is in a period of stable resource demand or a period of declining resource demand, the container resource pre-start requirement is determined to be that container resource pre-start is not required.

[0060] In this embodiment, if the task status is in a period of increasing container resource demand, it is determined that container resource pre-start is required; if the task status is in a period of stable container resource demand or a period of declining demand, it is determined that container resource pre-start is not required, so as to flexibly execute the container resource pre-start strategy according to the specific content of the strategy in the future.

[0061] In some embodiments, if the task processing location is in this node device and the resource pre-start requirement is to execute container resource pre-start, the above-mentioned step S22 may specifically include: S31, providing serverless container resources with a container resource amount equal to the currently required container resource amount for the task in this node device, and processing the task; S32, if it is monitored that the remaining resource amount of the serverless container resources of this node device is greater than or equal to the specified number of container resources, then pre-starting the serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the serverless container resources that have been pre-started by this node device can be directly used to expand the container.

[0062] In some embodiments, the specified number of container resources is less than the current number of container resources required by the task. For example, the specified number of container resources may be a predetermined ratio of the number of container resources currently required by the task. The predetermined ratio may be 20%, or other ratio values, such as 10%, 30%, etc., which may be obtained according to the requirements in the actual application scenario, and is not specifically limited in the embodiments of the present invention.

[0063] In this embodiment, the task is handed over to the MEC serverless server for processing, and at the same time, it is monitored whether the remaining serverless container quantity is greater than the specified number of container resources; if so, the serverless container resources are pre-started for standby according to the specified number of container resources, and the time for starting the container during subsequent capacity expansion is reduced by pre-starting the container resources, the container startup time is reduced, and the task processing efficiency is improved.

[0064] In some embodiments, if the task processing location is in the local node device and the resource pre-start requirement is to execute container resource pre-start, the above step S22 may further include the following steps.

[0065] S33, if it is monitored that the remaining resource amount of the server-less container resources of this node device is less than the specified number of container resources, a resource pre-start request is sent to the cloud data center; wherein the resource pre-start request is used to request the cloud data center to pre-start server-less container resources for the task whose resource amount is equal to the container resource difference; wherein the container resource difference is the difference between the specified number of container resources and the remaining resource amount.

[0066] S34, during the processing of the task, if it is monitored that the remaining resources are less than the demand for serverless container resources of the task, the processing results of the task and the completed part of the task are uploaded to the cloud data center.

[0067] Among them, the cloud data center is used to store the processing results of the completed part, and call the serverless container resources whose pre-started container resources are equal to the container resource difference to continue processing the unfinished part of the task and store the processing results of the unfinished part, so as to obtain the complete processing results of the stored task based on the processing results of the completed part and the processing results of the unfinished part.

[0068] In this embodiment, when the serverless processing task of the MEC node device is in a period of increasing resource demand, but the remaining resource amount of the serverless container resources of the MEC node device is less than the specified number of container resources, the serverless container resources of the pre-start container resource difference can be requested from the cloud data center according to the container resource difference. When the MEC cannot meet the task growth demand, the serverless container resources of the cloud data center are directly called, which reduces the time of restarting the container and increases the processing efficiency.

[0069] In some embodiments, if the task processing location is in the local node device and the resource pre-start requirement is that no container resource pre-start is required, the above step S22 may specifically include: S41, providing the task with server-less container resources with a container resource amount equal to the currently required container resource amount in the local node device, and processing the task; S42, uploading and storing the processing results of the task to the reserved storage space of the cloud data center.

[0070] In this embodiment, the task is processed by the MEC node device, and the task processing result is uploaded to the cloud data center for storage after the processing is completed. Processing on the edge side reduces the task interaction process, reduces the task processing time, and improves the task processing efficiency.

[0071] In some embodiments, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is to pre-start container resources, the above step S22 may specifically include: S51, allocating serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and processing the task; S52, pre-starting serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the serverless container resources pre-started in the cloud data center can be directly used to expand the container.

[0072] In this embodiment, the task is uploaded to the serverless server of the cloud data center for processing. When the task is in a period of increasing resource demand, the cloud data center can provide serverless resources with a container resource amount equal to the currently required container resource amount to process the task. At the same time, according to the specified number of container resources, the serverless resources are pre-started for backup. By pre-starting container resources, the time for starting the container during subsequent capacity expansion is reduced, thereby improving the efficiency of task expansion.

[0073] In some embodiments, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is that no container resource pre-start is required, then the above step S22 may specifically include: S53, allocating serverless container resources to the task in the cloud data center according to the current required container resource amount, and processing the task; S54, storing the processing result of the task in a reserved storage space of the cloud data center.

[0074] In this embodiment, the task is uploaded to the serverless server of the cloud data center for processing. When the task status is in a demand decay period or a stable period, serverless resources with a container resource amount equal to the currently required container resource amount are directly provided to process the task, and the processing results are stored in a predetermined storage location in the cloud data center.

[0075] According to the serverless processing method of an embodiment of the present invention, a user device can upload a task to the nearest MEC node. After receiving the task uploaded by the user device, the MEC node compares the container resources currently required for the task with the amount of serverless container resources currently available on the MEC node, and determines the processing location of the task according to the comparison result; and when processing tasks at different processing locations, the container is further pre-started during the period when task demand increases, and the serverless container resources of the pre-start container resource difference are requested from the cloud data center according to the container resource difference, thereby reducing the startup time of the container and increasing processing efficiency.

[0076] The step division of the above methods is only for the purpose of clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this patent.

[0077] In a second aspect, an embodiment of the present invention provides a server-free processing method.

[0078] Figure 2 FIG. 1 is a flow chart showing a serverless processing method according to another embodiment of the present invention. Figure 2 As shown, the serverless processing method in the embodiment of the present invention includes the following steps. S210, receiving a task uploaded by a mobile edge computing MEC node device; wherein the MEC node device is the MEC node device closest to the user device to which the task belongs, and the task is uploaded when the task upload condition is met; the task upload condition includes: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device.

[0079] S220, performing serverless processing on the task.

[0080] According to the serverless processing method of the embodiment of the present invention, the server corresponding to the cloud data center can receive the task uploaded by the mobile edge computing MEC node device. The task is that after the user device uploads the task to the MEC node device that is closest to the user device, the MEC node device uploads the task to the cloud data center when it determines that the current container resource required for the task is greater than or equal to the current available serverless container resource of the MEC node device; that is, in the serverless processing method of the embodiment of the present invention, the task uploaded by the user device can determine the specific processing location for processing the task based on the result of comparing the current container resource required for the task with the current available serverless container resource of the MEC node device, thereby realizing flexible selection of serverless processing of the task at different processing locations based on the comparison result, effectively reducing the burden of serverless processing tasks of the cloud computing center and improving task processing efficiency.

[0081] In some embodiments, step S220 may specifically include: S61, determining resource pre-start requirements based on the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; S62, performing serverless processing on the task based on the resource pre-start requirements.

[0082] In this embodiment, based on the processing of tasks in the cloud data center, the container resource pre-start strategy can be further determined according to the container resource demand trend represented by the task status, that is, whether to pre-start the container resources, thereby realizing the flexible execution of the container resource pre-start strategy according to the demand trend of the task for serverless container resources.

[0083] In some embodiments, step S61 may specifically include: if the task status is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the task status is in a period of stable resource demand or a period of declining resource demand, the container resource pre-start requirement is determined to be that container resource pre-start is not required.

[0084] In this embodiment, if the task status is in a period of increasing container resource demand, it is determined that container resource pre-start is required; if the task status is in a period of stable container resource demand or a period of declining demand, it is determined that container resource pre-start is not required, so as to flexibly execute the container resource pre-start strategy according to the specific content of the strategy in the future.

[0085] In some embodiments, if the resource pre-start requirement is to execute container resource pre-start, step S62 may specifically include: S71, allocating serverless container resources to the task and processing the task according to the current required container resource quantity; S72, pre-starting serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the pre-started serverless container resources of this cloud data center can be directly used to expand the container capacity.

[0086] In this embodiment, task processing is performed serverlessly in the cloud data center. When the task is in a period of increasing resource demand, the cloud data center can provide serverless resources with a container resource amount equal to the currently required container resource amount to process the task. At the same time, according to the specified number of container resources, serverless resources are pre-started for backup. By pre-starting container resources, the time for starting containers during subsequent capacity expansion is reduced, thereby improving the efficiency of task expansion.

[0087] In some embodiments, if the resource pre-start requirement is that there is no need to perform container resource pre-start, step S62 may specifically include: S81, if the resource pre-start requirement is that there is no need to pre-start container resources, allocate serverless container resources to the task and process the task according to the current required container resource amount, and, S82, store the processing result of the task in the reserved storage space of this cloud data center.

[0088] In this embodiment, the task is uploaded to the serverless server of the cloud data center for processing. When the task status is in a demand decay period or a stable period, serverless resources with a container resource amount equal to the currently required container resource amount are directly provided to process the task, and the processing results are stored in a predetermined storage location in the cloud data center.

[0089] In some embodiments, the serverless processing method also includes the following steps.

[0090] S91, receiving a resource pre-start request from an MEC node device; wherein the resource pre-start request is used to request a serverless container resource whose pre-start resource amount for a task is equal to the container resource difference; wherein the container resource difference is the difference between the specified number of container resources and the remaining amount of serverless container resources of the MEC node device.

[0091] S92: Pre-start serverless container resources for the task according to the container resource difference.

[0092] S93, in response to the tasks uploaded by the MEC node device and the processing results of the completed part of the tasks, stores the processing results of the completed part, and uses the serverless container resources whose pre-started resource amount is equal to the container resource difference to process the unfinished part of the task to obtain the processing results of the unfinished part.

[0093] S94, storing the processing results of the unfinished part, and obtaining all the processing results of the stored tasks according to the processing results of the completed part and the processing results of the unfinished part.

[0094] Through the above steps S91-S94, when the serverless processing task of the MEC node device is in a period of increasing resource demand, but the remaining resource amount of the serverless container resources of the MEC node device is less than the specified number of container resources, the serverless container resources of the pre-started container resource difference can be requested from the cloud data center according to the container resource difference. When the MEC cannot meet the task growth demand, the task and the processing results of the completed part of the task are uploaded to the server corresponding to the cloud data center. The cloud data center determines the unfinished part of the task based on the received task and the processing results of the completed part of the task, and uses the serverless container resources of the pre-started container resource difference to continue task processing; reducing the time for MEC to restart the container and increasing the task processing efficiency.

[0095] According to the serverless processing method of an embodiment of the present invention, the server corresponding to the cloud data center can receive a task uploaded by a MEC node device, where the task is that the user device uploads the task to the nearest MEC node, and the MEC node compares the container resources currently required for the task with the amount of serverless container resources currently available on the MEC node, and determines the processing location of the task based on the comparison result; and when the server corresponding to the cloud data center performs serverless processing on the task, it can further pre-start the container during the period when the task demand increases, and request the cloud data center to pre-start the serverless container resources of the container resource difference based on the container resource difference, thereby reducing the startup time of the container and increasing processing efficiency.

[0096] The step division of the above methods is only for the purpose of clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this patent.

[0097] In a third aspect, an embodiment of the present invention provides a serverless processing device.

[0098] Figure 3 FIG. 2 shows a schematic diagram of the structure of a serverless processing device provided according to an embodiment of the present invention. Figure 3 As shown, the serverless processing device 300 includes:

[0099] The task receiving module 310 is used to receive tasks uploaded by user equipment.

[0100] The information acquisition module 320 is used to obtain the current required container resources of the task obtained through analysis and the current available serverless container resources of the node device obtained through monitoring; wherein, the node device is the mobile edge computing MEC node device closest to the user device.

[0101] The result determination module 330 is used to determine the comparison result between the currently required container resource amount and the currently available serverless container resource amount.

[0102] The processing location determination module 340 is used to select a task processing location between the server in the cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location.

[0103] In some embodiments, the processing location determination module 340 is specifically used to: if the currently required amount of container resources is greater than or equal to the currently available amount of serverless container resources, upload the task to the cloud data center to process the task through the serverless of the cloud data center; if the currently required amount of container resources is less than the currently available amount of serverless container resources, process the task through the serverless of this node device.

[0104] In some embodiments, the serverless processing device 300 also includes: a pre-start requirement determination module, which is used to determine the resource pre-start requirement at the processing location according to the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; and a task processing module, which is used to perform serverless processing on the task at the processing location according to the resource pre-start requirement.

[0105] In some embodiments, the pre-start requirement determination module can be specifically used to: if the task status is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the task status is in a period of stable resource demand or a period of declining resource demand, the container resource pre-start requirement is determined to be that no container resource pre-start is required.

[0106] In some embodiments, if the task processing location is on the local node device and the resource pre-start requirement is to execute container resource pre-start, the task processing module can be specifically used to: provide server-less container resources with a container resource amount equal to the currently required container resource amount for the task on the local node device, and process the task; if it is monitored that the remaining resource amount of the server-less container resources of the local node device is greater than or equal to the specified number of container resources, then pre-start the server-less container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the server-less container resources that have been pre-started on the local node device can be directly used to expand the container.

[0107] In some embodiments, the serverless processing device 300 also includes: a request sending module, which is used to send a resource pre-start request to the cloud data center if it is monitored that the remaining resource amount of the serverless container resources of the node device is less than the specified container resource amount; wherein the resource pre-start request is used to request the cloud data center to pre-start serverless container resources for the task with a resource amount equal to the container resource difference; wherein the container resource difference is the difference between the specified container resource amount and the remaining resource amount; a task uploading module, which is used to upload the task and the processing results of the completed part of the task to the cloud data center during the processing of the task if it is monitored that the remaining resource amount is less than the demand for serverless container resources by the task; wherein the cloud data center is used to store the processing results of the completed part, and call the serverless container resources whose pre-started container resource amount is equal to the container resource difference, continue to process the unfinished part of the task, and store the processing results of the unfinished part, so as to obtain all the processing results of the stored task based on the processing results of the completed part and the processing results of the unfinished part.

[0108] In some embodiments, if the task processing location is in the local node device and the resource pre-start requirement is that no container resource pre-start is required, the task processing module can be specifically used to: provide server-less container resources with a container resource amount equal to the currently required container resource amount for the task in the local node device, and process the task; upload and store the processing results of the task to the reserved storage space of the cloud data center.

[0109] In some embodiments, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is to pre-start container resources, the task processing module can be specifically used to: allocate serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and process the task; pre-start serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the serverless container resources pre-started in the cloud data center can be directly used to expand the container.

[0110] In some embodiments, if the task processing location is on a server in a cloud data center and the resource pre-start requirement is that no container resource pre-start is required, the task processing module can be specifically used to: allocate serverless container resources to the task in the cloud data center according to the current required container resource amount, and process the task; store the processing results of the task in a reserved storage space of the cloud data center.

[0111] According to the serverless processing device of an embodiment of the present invention, a user device can upload a task to the nearest MEC node. After receiving the task uploaded by the user device, the MEC node compares the container resources currently required for the task with the amount of serverless container resources currently available on the MEC node, and determines the processing location of the task according to the comparison result; and when processing tasks at different processing locations, the container is further pre-started during the period when task demand increases, and the serverless container resources of the pre-start container resource difference are requested from the cloud data center according to the container resource difference, thereby reducing the startup time of the container and increasing processing efficiency.

[0112] In a fourth aspect, an embodiment of the present invention provides a serverless processing device.

[0113] Figure 4 FIG. 2 shows a schematic diagram of the structure of a serverless processing device provided according to another embodiment of the present invention. Figure 4 As shown, the serverless processing device 400 includes the following modules.

[0114] The receiving module 410 is used to receive tasks uploaded by a mobile edge computing MEC node device; wherein the MEC node device is the MEC node device that is closest to the user device to which the task belongs, and the task is uploaded when the task upload conditions are met; the task upload conditions include: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device; the processing module 420 is used to perform serverless processing on the task.

[0115] In some embodiments, the processing module 420 may specifically include: a demand determination unit, used to determine the resource pre-start requirement based on the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; the processing module 420 is also used to perform serverless processing on the task based on the resource pre-start requirement.

[0116] In some embodiments, the demand determination unit can be specifically used to: if the task status is in a period of rising resource demand, the resource pre-start requirement is to execute container resource pre-start; if the task status is in a period of stable resource demand or a period of declining resource demand, the container resource pre-start requirement is determined as not requiring container resource pre-start.

[0117] In some embodiments, if the resource pre-start requirement is to execute container resource pre-start, the processing module 420 can be specifically used to: allocate serverless container resources to the task and process the task according to the current required container resource quantity; pre-start serverless container resources for the task according to the specified number of container resources, so that when the task needs to be expanded, the pre-started serverless container resources of this cloud data center can be directly used to expand the container capacity.

[0118] In some embodiments, if the resource pre-start requirement is that there is no need to perform container resource pre-start, the processing module 420 can be specifically used for: if the resource pre-start requirement is that there is no need to pre-start container resources, allocate serverless container resources to the task and process the task according to the current required amount of container resources, and store the processing results of the task in the reserved storage space of this cloud data center.

[0119] In some embodiments, the serverless processing device 400 also includes: a request receiving module, which is used to receive a resource pre-start request from an MEC node device; wherein the resource pre-start request is used to request to pre-start a serverless container resource for a task whose resource amount is equal to the container resource difference; wherein the container resource difference is the difference between the specified number of container resources and the remaining resource amount of the serverless container resources of the MEC node device; a pre-start module, which is used to pre-start the serverless container resource for the task according to the container resource difference; the processing module 420 is also used to respond to the tasks uploaded by the MEC node device and the processing results of the completed part of the tasks, store the processing results of the completed part, and use the serverless container resources with the pre-started resource amount equal to the container resource difference to process the unfinished part of the task to obtain the processing results of the unfinished part; the storage module is used to store the processing results of the unfinished part, and obtain all the processing results of the stored tasks based on the processing results of the completed part and the processing results of the unfinished part.

[0120] According to the serverless processing device of an embodiment of the present invention, the server corresponding to the cloud data center can receive a task uploaded by a MEC node device, where the task is that the user device uploads the task to the nearest MEC node, and the MEC node compares the container resources currently required for the task with the amount of serverless container resources currently available on the MEC node, and determines the processing location of the task based on the comparison result; and when the server corresponding to the cloud data center performs serverless processing on the task, it can further pre-start the container during the period when the task demand increases, and request the cloud data center to pre-start the serverless container resources of the container resource difference based on the container resource difference, thereby reducing the startup time of the container and increasing processing efficiency.

[0121] It should be clear that the present invention is not limited to the specific configurations and processes described in the above embodiments and shown in the figures. For the convenience and brevity of description, a detailed description of the known methods is omitted here, and the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, which will not be repeated here.

[0122] Figure 5 A schematic diagram of a network architecture for serverless processing according to an exemplary embodiment of the present invention is shown. Figure 5 As shown, the network architecture includes: user equipment 510, MEC node equipment 520 and cloud data center 530.

[0123] Among them, the MEC node device 520 may include: a task analysis module 521, a resource supervision module 522, a decision module 523 and a MEC serverless 524; the MEC serverless 524 can perform task processing through the container resources of at least one container contained therein.

[0124] The cloud data center 530 may include: a cloud data center serverless 531 and a storage module 532; wherein the cloud data center serverless 531 may perform task processing through container resources of at least one container.

[0125] In some embodiments, the MEC node device 520 may be the MEC node device that is closest to the user device 510; that is, the user device 510 may upload the task to be processed to the MEC node device 520 that is closest to it, so as to perform further analysis, decision-making, and task processing at the MEC node device 520.

[0126] Specifically, the MEC node device 520 can perform analysis, decision-making and task processing through the following modules.

[0127] The task analysis module 521 is used to analyze the relevant characteristics of the task, such as the container resources currently required to process the task (which can be represented as m below), and the resource demand trend of the task status (for example, whether it is in the stage of increasing demand or decreasing demand).

[0128] The resource supervision module 522 is used to monitor the current available serverless container resource amount of the MEC node device 520 (which can be represented as M below).

[0129] The decision-making module 523 is used to determine the task processing location according to the information obtained by analysis through the task analysis module 521 and the information obtained by monitoring through the resource supervision module 522.

[0130] In some embodiments, the decision-making module 523 is specifically used to, when it is determined that m >= M, that is, when it is determined that the container resources currently required to process the task are greater than or equal to the current available serverless container resource amount of the MEC node device 520, upload the task to the serverless in the cloud data center 531 of the cloud data center 530 for processing; the storage module 532 of the cloud data center 530 is used to store the task processing result in a predetermined storage space.

[0131] The cloud data center 530 is specifically used for: when the task status is in the rising period of container resource demand, providing serverless resources with a container resource amount of m to process the task, and pre-starting serverless resources with a specified resource amount as a standby at the same time, so as to reduce the time for starting containers during subsequent expansion through the form of pre-starting container resources and improve the task expansion efficiency.

[0132] Specifically, the specified resource amount can be pre-started according to a predetermined ratio of the current required container resource quantity for processing the task. Exemplarily, the predetermined ratio can be 20%, or it can be other values, such as 10%, 30%, etc., which can be obtained according to the requirements in the actual application scenario, and the embodiments of the present invention do not make specific limitations.

[0133] The cloud data center 530 is specifically further used for: when the task status is in the declining period of container resource demand or the stable period of container resource demand, providing serverless resources with a container resource amount of m to process the task.

[0134] In some embodiments, the decision-making module 523 is specifically used to, when m < M, that is, when the container resources currently required to process the task are less than the current available serverless container resource amount of the MEC node device 520, send the task to the MEC serverless 524 for processing.

[0135] The MEC serverless 524 is used to process the task and upload the processing result of the task to the storage module 532 of the cloud data center 530 after the task processing is completed; the storage module 532 is used to store the processing result of the task.

[0136] In this embodiment, performing task processing on the edge-side MEC node device can reduce the task interaction process, thereby reducing the task processing time and improving the task processing efficiency.

[0137] In some embodiments, when the MEC serverless 524 is used to process the task, it is specifically used to: directly process the task when the task is in a stable period of container resource demand or in a period of attenuation of container resource demand.

[0138] In some embodiments, the decision module 523 is used to hand over the task to the MEC serverless 524 for processing if the task is in a period of increasing container resource demand.

[0139] The resource monitoring module 522 can be used to simultaneously monitor whether the remaining serverless container quantity is greater than the specified number of container resources. If the remaining serverless container quantity is greater than or equal to the specified number of container resources (for example, the remaining serverless container quantity is greater than or equal to 20% of the number of container resources currently required for the task), then according to the specified number of container resources, pre-start the serverless resources as backup serverless container resources; thereby reducing the time for starting the container during subsequent capacity expansion by pre-starting the container resources.

[0140] In some embodiments, the resource monitoring module 522, when used to monitor whether the remaining serverless container quantity is greater than the specified number of container resources, if the remaining serverless container quantity is less than the specified number of container resources (for example: the remaining serverless container quantity is less than 20% of the number of container resources currently required to process the task), the decision module 523 can be used to request the cloud data center 530 to pre-start the serverless container resources of the container resource difference based on the container resource difference (the container resource difference is equal to the difference between the specified number of container resources and the remaining serverless container quantity). When the MEC node device 520 cannot meet the growing demand for serverless resources for task processing, the processing results of the task and the completed part of the task are uploaded to the cloud data center 530, thereby directly calling the container resources of the cloud data center serverless 531 of the cloud data center 530, reducing the time to restart the container and increasing processing efficiency.

[0141] According to the serverless processing system of the embodiment of the present invention, by processing user tasks at the nearest MEC node, it is possible to reduce the task processing interaction process and the task processing time. At the same time, by pre-starting the container during the period of increased task demand, the container startup time is reduced, the startup time when the task is expanded is reduced, and the task processing efficiency is improved.

[0142] Figure 6 A schematic diagram of a network architecture of a network device according to an embodiment of the present invention is shown. Figure 6 , an embodiment of the present invention provides a network device, comprising:

[0143] One or more processors 601; a memory 602, on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors 601 implement any of the above methods; one or more I / O interfaces 603, connected between the processor 601 and the memory 602, configured to implement information exchange between the processor 601 and the memory 602.

[0144] Among them, the processor 601 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 602 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) 603 is connected between the processor 601 and the memory 602, and can realize information interaction between the processor 601 and the memory 602, including but not limited to a data bus (Bus), etc.

[0145] In some embodiments, the processor 601 , the memory 602 , and the I / O interface 603 are connected to each other through a bus, and further connected to other components of the computing device.

[0146] This embodiment further provides a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the method provided by this embodiment is implemented. To avoid repeated description, the specific steps of the method of this embodiment are not repeated here.

[0147] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the invented method described above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0148] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0149] Those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present embodiment and to form different embodiments.

[0150] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A serverless processing method, characterized in that: Applied to this node device, the method includes: Receive tasks uploaded by user devices; Acquire the currently required container resource amount of the task obtained through analysis and the currently available serverless container resource amount of the node device obtained through monitoring; wherein the node device is a mobile edge computing MEC node device that is closest to the user device; Determine a comparison result of the currently required amount of container resources and the currently available amount of serverless container resources; Selecting a task processing location between a server in a cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location; Wherein, performing serverless processing on the task at the task processing location includes: Determining resource pre-start requirements at the task processing location according to the state of the task; wherein the state of the task is used to indicate the demand trend of the task for serverless container resources; The task is serverlessly processed at the task processing location according to the resource pre-start requirement.

2. The method according to claim 1, wherein: The selecting a task processing location between a server in a cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location, includes: If the currently required container resource amount is greater than or equal to the currently available serverless container resource amount, uploading the task to the cloud data center to process the task through the serverless process of the cloud data center; If the currently required container resource amount is less than the currently available serverless container resource amount, the task is processed by the serverless method of the node device.

3. The method according to claim 1, wherein: The determining of the resource pre-start requirement according to the status of the task includes: If the task is in a resource demand increasing period, the resource pre-start requirement is to execute container resource pre-start; If the state of the task is in a stable resource demand period or a declining resource demand period, it is determined that the container resource pre-start requirement is that no container resource pre-start is required.

4. The method according to claim 3, wherein: If the task processing location is at the local node device and the resource pre-start requirement is to execute container resource pre-start, the task is serverlessly processed at the processing location according to the resource pre-start requirement, including: The node device provides serverless container resources with a container resource amount equal to the currently required container resource amount for the task, and processes the task; If it is monitored that the remaining resource amount of the serverless container resources of this node device is greater than or equal to the specified number of container resources, the serverless container resources are pre-started for the task according to the specified number of container resources, so that when the task needs to be expanded, the serverless container resources pre-started by this node device are directly used to expand the container capacity; wherein, the specified number of container resources is less than the current required container resources for the task.

5. The method according to claim 3, further comprising: If it is monitored that the remaining resource amount of the serverless container resources of the node device is less than the specified container resource amount, a resource pre-start request is sent to the cloud data center; The resource pre-start request is used to request the cloud data center to pre-start serverless container resources for the task with a resource amount equal to the container resource difference; wherein the container resource difference is the difference between the specified container resource quantity and the remaining resource quantity; During the processing of the task, if it is monitored that the remaining amount of resources is less than the demand of the task for serverless container resources, uploading the processing result of the task and the completed part of the task to the cloud data center; Among them, the cloud data center is used to store the processing results of the completed part, and call the serverless container resources whose pre-started container resources are equal to the container resource difference, continue to process the unfinished part of the task, and store the processing results of the unfinished part, so as to obtain all the stored processing results of the task based on the processing results of the completed part and the processing results of the unfinished part.

6. The method according to claim 3, wherein: If the task processing location is at the local node device and the resource pre-start requirement is that container resource pre-start is not required, server-free processing of the task is performed at the processing location according to the resource pre-start requirement, including: The node device provides serverless container resources with a container resource amount equal to the currently required container resource amount for the task, and processes the task; The processing result of the task is uploaded and stored in a predetermined storage space of the cloud data center.

7. The method according to claim 3, wherein: If the task processing location is on a server in a cloud data center and the resource pre-start requirement is to require container resource pre-start, the task is processed serverlessly at the processing location according to the resource pre-start requirement, including: Allocating serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and processing the task; According to the specified number of container resources, serverless container resources are pre-started for the task, so that when the task needs to be expanded, the serverless container resources pre-started in the cloud data center are directly used to expand the container capacity.

8. The method according to claim 3, wherein: If the task processing location is on a server in a cloud data center and the resource pre-start requirement is that container resource pre-start is not required, the task is processed serverlessly at the processing location according to the resource pre-start requirement, including: Allocating serverless container resources to the task in the cloud data center according to the currently required amount of container resources, and processing the task; The processing result of the task is stored in a predetermined storage space of the cloud data center.

9. A serverless processing method, characterized in that: Applied to a server in a cloud data center, the method comprises: Receive tasks uploaded by mobile edge computing MEC node devices; The MEC node device is the MEC node device that is closest to the user device to which the task belongs, and the task is a task that is uploaded when a task upload condition is met; the task upload condition includes: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device; performing serverless processing on the tasks; The serverless processing of the task includes: Determining resource pre-start requirements according to the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; The task is serverlessly processed according to the resource pre-start requirement.

10. The method according to claim 9, characterized in that The determining of the resource pre-start requirement according to the status of the task includes: If the task is in a resource demand increasing period, the resource pre-start requirement is to execute container resource pre-start; If the state of the task is in a stable resource demand period or a declining resource demand period, it is determined that the container resource pre-start requirement is that no container resource pre-start is required.

11. The method according to claim 10, characterized in that If the resource pre-start requirement is to require container resource pre-start, the serverless processing of the task according to the resource pre-start requirement includes: Allocate serverless container resources to the task according to the currently required container resource amount and process the task; According to the specified number of container resources, serverless container resources are pre-started for the task, so that when the task needs to be expanded, the pre-started serverless container resources of the cloud data center are directly used to expand the container capacity.

12. The method according to claim 10, characterized in that If the resource pre-start requirement is that container resource pre-start does not need to be performed, the serverless processing of the task according to the resource pre-start requirement includes: If the resource pre-start requirement is that no container resource pre-start is required, serverless container resources are allocated to the task according to the currently required container resource amount and the task is processed, and the processing result of the task is stored in the reserved storage space of this cloud data center.

13. The method according to claim 9, further comprising: Receive a resource pre-start request from the MEC node device; wherein the resource pre-start request is used to request to pre-start a serverless container resource for the task whose resource amount is equal to the container resource difference; wherein the container resource difference is the difference between the specified number of container resources and the remaining resource amount of the serverless container resource of the MEC node device; pre-starting serverless container resources for the task according to the container resource difference; In response to the task uploaded by the MEC node device and the processing result of the completed part of the task, the processing result of the completed part is stored, and the serverless container resources whose pre-started resource amount is equal to the container resource difference are used to process the unfinished part of the task to obtain the processing result of the unfinished part; The processing result of the unfinished part is stored, and all the processing results of the task that have been stored are obtained according to the processing result of the completed part and the processing result of the unfinished part.

14. A serverless processing device, characterized in that: include: A task receiving module, used to receive tasks uploaded by user devices; An information acquisition module, used to obtain the currently required container resource amount of the task obtained through analysis and the currently available serverless container resource amount of the node device obtained through monitoring, wherein the node device is a mobile edge computing MEC node device that is closest to the user device; A result determination module, used to determine a comparison result between the currently required container resource amount and the currently available serverless container resource amount; a processing location determination module, configured to select a task processing location from between a server in a cloud data center and the node device according to the comparison result, so as to perform serverless processing on the task at the task processing location; Wherein, performing serverless processing on the task at the task processing location includes: Determining resource pre-start requirements at the task processing location according to the state of the task; wherein the state of the task is used to indicate the demand trend of the task for serverless container resources; The task is serverlessly processed at the task processing location according to the resource pre-start requirement.

15. A serverless processing device, characterized in that: include: A receiving module is used to receive tasks uploaded by mobile edge computing MEC node devices; The MEC node device is the MEC node device that is closest to the user device to which the task belongs, and the task is a task that is uploaded when a task upload condition is met; the task upload condition includes: the current required container resource amount of the task is greater than or equal to the current available serverless container resource amount of the MEC node device; A processing module, used for performing serverless processing on the task; The serverless processing of the task includes: Determining resource pre-start requirements according to the status of the task; wherein the status of the task is used to indicate the demand trend of the task for serverless container resources; The task is serverlessly processed according to the resource pre-start requirement.

16. A network device, characterized in that: include: one or more processors; A memory having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8, or the method according to any one of claims 9 to 13; One or more I / O interfaces are connected between the processor and the memory and are configured to implement information interaction between the processor and the memory.

17. A computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method according to any one of claims 1 to 8 or the method according to any one of claims 9 to 13 is implemented.

Images