Task processing method and system, storage medium and computer program product

Abstract

Provided in the embodiments of the present disclosure are a task processing method and system, and a storage medium. In the method, when it is determined that a task processing request is for a target-type service, a gateway determines a target computing resource from among computing resources having temporary leases, and directly sends the task processing request to the target computing resource for task processing, such that when processing the task processing request for the target-type service, the gateway does not need to access a scheduling component to determine a corresponding computing resource, thereby reducing a system side delay caused by accessing the scheduling component in a task processing process, and improving the processing efficiency of the target-type service. In addition, because the gateway does not need to send a corresponding predetermined request to the scheduling component in the present embodiment, the scheduling component also does not need to determine a computing resource and feed back a computing resource identifier, thereby saving server resources on the gateway and the scheduling component.

Description

[0001] METHOD, SYSTEM, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT The present disclosure claims priority to Chinese Patent Application No. 202410833486.0, filed on June 25, 2024 with the Chinese Patent Office, entitled “METHOD, SYSTEM, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT”, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD The present disclosure relates to the field of cloud computing technology, and more specifically, to a task processing method, system, storage medium and computer program product. BACKGROUND Serverless technology refers to that a user creates and runs a software application and service in a cloud server, and the user does not need to care about the content related to information technology (IT) facilities (such as management, upgrade, etc.), which is a product of further evolution of IT architecture, and mainly has the following characteristics: fine-grained allocation of computing resources is achieved; no need to pre-allocate resources, no need to configure and manage the operating system; truly high scalability and elasticity, supporting on-demand scaling; use on demand, pay on demand. Under the conventional architecture of Serverless, computing resources are often dynamically created. In order to provide the high elasticity capability required by the Serverless architecture and reduce the system cold start probability, a dynamic routing scheduling scheme is usually adopted, that is, the service gateway under the Serverless architecture usually requests the scheduling component to allocate computing resources for processing the task processing request when receiving the request, and requests the scheduling component to release the computing resources after the computing resources are executed. Since in the existing Serverless architecture, each request needs the service gateway to request the scheduling component to allocate the corresponding computing resources, the average call delay of the system side is high. This is a disaster and unacceptable for businesses with high queries-per-second (QPS) and low function execution time. SUMMARY The present disclosure provides a task processing method, system, storage medium and computer program product to solve the problem of high average call delay of the system side in the existing Serverless architecture.In a first aspect, the embodiments of the present disclosure provide a task processing method applied to a gateway, the method comprising: receiving a task processing request sent by a client; if the task processing request is a task processing request of a target type service, determining a target computing resource from computing resources with temporary leases; the computing resources with temporary leases are computing resources leased from a scheduling component; and sending the task processing request to the target computing resource for task processing. In a possible implementation manner, the determining the target computing resource from the computing resources comprises: obtaining a real-time concurrency degree of the computing resources; determining whether there is an available computing resource from the computing resources based on the real-time concurrency degree of the computing resources and a supported single-instance concurrency degree; and if there is an available computing resource, determining the target computing resource from the available computing resource. In a possible implementation manner, the determining whether there is an available computing resource from the computing resources based on the real-time concurrency degree of the computing resources and the supported single-instance concurrency degree comprises: determining whether the real-time concurrency degree reaches the single-instance concurrency degree; if the real-time concurrency degree does not reach the single-instance concurrency degree, determining that the computing resource is an available computing resource; and if the real-time concurrency degree reaches the single-instance concurrency degree, determining that the computing resource is an unavailable computing resource. In a possible implementation manner, the method further comprises: if there is no available computing resource, continuing to lease computing resources from the scheduling component, and taking the leased computing resource as the target computing resource. In a possible implementation manner, the leasing computing resources from the scheduling component comprises: sending a lease request to the scheduling component, the lease request being used to request the scheduling component to allocate a computing resource for processing the target type service; and receiving a lease response fed back by the scheduling component, the lease response being used to indicate the leased computing resource, and the indicated leased computing resource being determined from computing resources without temporary leases and used for processing the target type service. In a possible implementation manner, the method further comprises: recording a lease duration of the computing resource with the temporary lease; when the recorded lease duration reaches a maximum lease duration of the computing resource, sending a renewal request to the scheduling component, the renewal request being used to request the scheduling component to perform a renewal operation on the computing resource; receiving a renewal response fed back by the scheduling component, and re-recording the lease duration of the computing resource. In a possible implementation manner, the lease response is further used to indicate the maximum lease duration of the computing resource; or the method further comprises: negotiating the maximum lease duration of the computing resource with a system component.In a possible implementation, after the real-time concurrency of the computing resource is acquired, the method further includes: in a case where the real-time concurrency of the computing resource meets a preset condition, recording a duration for which the real-time concurrency of the computing resource meets the preset condition, the preset condition including: the real-time concurrency being less than or equal to a preset concurrency; and if the duration reaches a preset duration, sending a release request to the scheduling component, the release request being used to request the scheduling component to perform a release operation on the computing resource. In a possible implementation, before the task processing request sent by the client is received, the method further includes: receiving configuration information sent by a system component, the configuration information being used to indicate that a computing resource with a temporary lease is used to perform task processing on a target service type; and configuring a task processing scenario of the target type service according to the configuration information. In a second aspect, an embodiment of the present disclosure provides a task processing method, applied to a scheduling component, and the method includes: receiving a subscription request sent by a gateway, the subscription request being used to request allocation of a computing resource for processing a target type service; determining a computing resource for processing the target type service from computing resources without a temporary lease and used to process the target type service, and establishing a temporary lease between the computing resource and the gateway; and feeding back a subscription response to the gateway, the subscription response being used to indicate the computing resource with the temporary lease. In a possible implementation, the scheduling component is configured with a lease interface, and the method further includes: receiving a renewal request sent by the gateway through the lease interface, the renewal request being used to request a renewal operation on the computing resource; determining a computing resource corresponding to the renewal request, and performing a renewal processing on the temporary lease between the computing resource and the gateway; and feeding back a renewal response of the computing resource to the gateway. In a possible implementation, the method further includes: receiving a release request sent by the gateway, the release request being used to request a release operation on the computing resource; determining a computing resource corresponding to the release request, canceling the temporary lease between the computing resource and the gateway, and performing the release operation on the computing resource.In a third aspect, the embodiments of the present disclosure provide a task processing method applied to a cloud platform, the cloud platform comprising: a system component, and a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones, the method comprising: receiving, by the system component, a task processing request sent by a client, and forwarding the task processing request to at least two target gateways in the plurality of gateways; determining, by the target gateways, a target computing resource from computing resources with a temporary lease in a case where the task processing request is a task processing request of a target type service; the computing resource with a temporary lease is a computing resource leased from a scheduling component; and sending, by the target gateways, the task processing request to the target computing resource for task processing. In a fourth aspect, the present disclosure provides a task processing apparatus applied to a gateway, the apparatus comprising: a receiving module configured to receive a task processing request sent by a client; a processing module configured to determine a target computing resource from computing resources with a temporary lease in a case where the task processing request is a task processing request of a target type service; the computing resource with a temporary lease is a computing resource leased from a scheduling component; and a sending module configured to send the task processing request to the target computing resource for task processing. In a fifth aspect, the present disclosure provides a task processing apparatus applied to a scheduling component, the apparatus comprising: a receiving module configured to receive a lease request sent by a gateway, the lease request being used to request allocation of a computing resource for processing a target type service; a processing module configured to determine a computing resource for processing the target type service from computing resources without a temporary lease and used for processing the target type service, and establish a temporary lease between the computing resource and the gateway; and a sending module configured to feed back a lease response to the gateway, the lease response being used to indicate the computing resource with a temporary lease. In a sixth aspect, the present disclosure provides an electronic device comprising: a memory, a processor and a communication interface; wherein the memory stores computer execution instructions, the communication interface is configured to communicate with other devices through a transmission medium; and the processor executes the computer execution instructions stored in the memory to implement the task processing method according to any one of the possible implementation manners of any one of the first aspect, the second aspect and the third aspect.In a seventh aspect, the present disclosure provides a task processing system, comprising: a system component, and a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones, the system component is configured to receive a task processing request sent by a client, and forward the task processing request to at least two target gateways in the plurality of gateways; the target gateway is configured to determine a target computing resource from computing resources with temporary leases in the case of determining that the task processing request is a target type of business task processing request; the computing resource with temporary leases is a computing resource subscribed from a scheduling component, and the target computing resource is in the same availability zone as the target gateway; and the target gateway is further configured to send the task processing request to the target computing resource for task processing. In an eighth aspect, the present disclosure provides a computer readable storage medium, the computer readable storage medium stores a computer program, when the computer program is executed by a processor, the processor executes the task processing method as described in any one of the possible implementation manners of the first aspect, the second aspect and the third aspect. In a ninth aspect, the present disclosure provides a computer program product, comprising a computer program, when the computer program is executed by a processor, the task processing method as described in any one of the possible implementation manners of the first aspect, the second aspect and the third aspect is realized. The present disclosure provides a task processing method, system, storage medium and computer program product. The method determines a target computing resource from computing resources with temporary leases by the gateway in the case of determining that the task processing request is a target type of business, and directly sends the task processing request to the target computing resource for task processing, so that the gateway does not need to access the scheduling component to determine the corresponding computing resource when processing the target type of business task processing request, thereby reducing the system side delay caused by accessing the scheduling component in the task processing process, and improving the processing efficiency of the target type of business. At the same time, since the gateway does not need to send the corresponding predetermined request to the scheduling component, the scheduling component also does not need to determine the computing resource and feed back the computing resource identifier, thereby saving the server resources on the gateway and the scheduling component. BRIEF DESCRIPTION OF DRAWINGS In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present disclosure, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.FIG. 1 is a structural diagram of a possible task processing system according to an embodiment of the present disclosure; FIG. 2 is a flow diagram of a prior task scheduling scheme; FIG. 3 is a flow diagram of a possible gateway call scheduling component according to an embodiment of the present disclosure; FIG. 4 is a flow diagram of a possible task processing method according to an embodiment of the present disclosure; FIG. 5 is a flow diagram of a possible task processing method according to an embodiment of the present disclosure; FIG. 6 is an interaction diagram of a possible task processing method according to an embodiment of the present disclosure; FIG. 7 is an interaction diagram of a possible task processing method according to an embodiment of the present disclosure; FIG. 8 is a flow diagram of a possible task processing method according to an embodiment of the present disclosure; FIG. 9 is a structural diagram of a possible task processing apparatus according to an embodiment of the present disclosure; FIG. 10 is a structural diagram of a possible task processing apparatus according to an embodiment of the present disclosure; FIG. 11 is a structural diagram of an electronic device according to an embodiment of the present disclosure. In order to make the purposes, technical schemes and advantages of the embodiments of the present disclosure clearer, the technical schemes in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present disclosure. First, the terms related to the embodiments of the present disclosure are introduced.

[0002] Serverless, a cloud computing mode, in which users do not need to manage servers, and resources are automatically allocated and services are provided by the cloud platform. The Serverless architecture has high elasticity and pay-as-you-go characteristics. Computing resources generally refer to the central processing unit (CPU) resources, memory resources, hard disk resources, and network resources required by computer programs during runtime. The computing resources in this embodiment may, for example, be function instances in the Serverless architecture. Function instance: a basic execution unit in the Serverless architecture, responsible for executing specific business logic of a user function. Scheduling component: a component in the Serverless architecture responsible for dynamically allocating function instances to handle requests according to the priority of the requests and the resource usage. Single-instance concurrency degree: the number of concurrent requests that can be handled by a single function instance. When the real-time concurrency degree on a single instance reaches the single-instance concurrency degree, new requests are sent to other function instances that have not reached the single-instance concurrency degree. Queries-per-second (QPS): the amount of request data processed by a unit of computing resources per unit of time. Disaster recovery (DR): the ability of an IT system to quickly and effectively recover business operations in the event of a major natural disaster, human sabotage, cyber attack, or other sudden event.

[0003] RTA (Real Time Application Programming Interface) service is used to meet the real-time personalized delivery requirements of advertisers and reduce waste of funds in bidding. Embodiments shown in the disclosure can be applied to a Serverless architecture of a cloud service, for example. First, a task processing system under a Serverless architecture of a cloud service is explained. FIG. 1 is a structural schematic diagram of a possible task processing system provided by an embodiment of the disclosure. The task processing system can be entirely deployed in a cloud environment. The cloud environment is an entity that provides cloud services to users by utilizing underlying resources in a cloud computing mode. The cloud environment includes a large number of underlying resources (including computing resources, storage resources, and network resources) owned by a cloud service provider. The computing resources included in the cloud environment can be a large number of computing devices (such as servers). The cloud environment can be deployed in different regions, and each region contains at least one availability zone, and the computing resources can be deployed in the at least one availability zone. As shown in FIG. 1, the task processing system includes system components, gateways, computing resources, and scheduling components. The system components are communicatively connected to a plurality of different availability zones, and at least one gateway, computing resource, and scheduling component can be deployed in each availability zone. For example, at least gateway 11, scheduling component 12, and computing resource 13o are deployed in availability zone 1. The gateway can communicate with the scheduling component or computing resource in the same availability zone or the scheduling component or computing resource in a different availability zone. In the task processing process, considering the disaster recovery capability of the service, computing resources in at least two availability zones are usually required to execute the corresponding service. The task processing system can be provided to users by the cloud service provider as a task processing service in a cloud platform abstraction, and after the user purchases the cloud service (for example, pre-recharge and then settle according to the use of the final resources), the cloud platform provides the user with a task processing service by using the task processing system. It can be understood that the task processing system can be logically divided into multiple parts, each part having different functions. The deployment of the task processing system is relatively flexible, and the entire task processing system can be deployed in a cloud data center, or each part with different functions can be deployed in any two or three of a client, an edge environment, and a cloud environment. The client can include at least one of the following, for example: a terminal server, a smart phone, a notebook computer, a tablet computer, a personal desktop computer, and the like. The edge environment is an environment including a set of edge computing devices close to the client, and the edge computing devices include: an edge server, an edge small station with computing power, and the like. The parts of the task processing system deployed in different environments or devices cooperate to provide the user with a task processing function.It should be understood that the embodiments of the present disclosure do not limit the division of which parts of the task processing system are specifically deployed in which environment, and in actual application, adaptive deployment can be performed according to the computing capability of the client, the resource occupation of the edge environment and the cloud environment, or specific application requirements. The following explains and describes the existing task scheduling scheme under the Serverless architecture of the cloud service. In the Serverless architecture, function instances are often dynamically created. In order to provide the high elasticity capability required by the Serverless architecture and reduce the system cold start probability, a dynamic routing task scheduling scheme is usually adopted. FIG. 2 is a flowchart of the existing task scheduling scheme. Under the Serverless architecture, the gateway shown in FIG. 1 may, for example, be an Application Programming Interface (API) service gateway in FIG. 2, the scheduling component may, for example, be a routing scheduling component in FIG. 2, and the computing resource may, for example, be a function instance in FIG. 2. The Reserve interface and the Ide interface are deployed on the scheduling component. As shown in FIG. 2, the specific process of the existing task scheduling scheme includes:

[0004] The API service gateway, after obtaining the task processing request, prebooks a function instance for executing the task processing request through a Reserve interface to a routing scheduling component; the routing scheduling component confirms and returns a corresponding function instance identifier; the API service gateway determines and calls the function instance for task processing according to the function instance identifier. After the function instance completes task processing, the API service gateway feeds back processing results to the API service gateway; after receiving the processing results fed back by the function instance, the API service gateway requests the routing scheduling component to release the function instance; the routing scheduling component releases the function instance through an Ide interface based on the release request. It can be understood that, in the above process, since each request needs the API service gateway to request the routing scheduling component to allocate a function instance for processing the request, and release the function instance after the function instance completes task processing, so that in the execution process of each task processing request, the system side delay includes at least twice network delay of the API service gateway accessing the routing scheduling component. The above task scheduling scheme is applicable in most scenarios, but in the business scenario with high QPS and low function execution time, the above task scheduling scheme will affect the performance of such business. The business with high QPS and low function execution time in the embodiment of the present disclosure may include, for example, a class RTA business. The following takes the class RTA business as an example for description. Specifically, the QPS of the class RTA business can reach more than 100,000 QPS, and the function execution time of the class RTA business is usually only about 10 milliseconds (milliseconds, ms). Since the QPS of such business is very high, when processing the class RTA business, considering the disaster recovery capability, the API service gateway usually prebooks function instances from at least two routing scheduling components distributed in different availability zones. FIG. 3 is a flowchart of a possible gateway calling a scheduling component provided by an embodiment of the present disclosure. As shown in FIG. 3, the gateway A and the scheduling component B are arranged in the availability zone A, and the gateway B and the scheduling component Bo are arranged in the availability zone B. In the case that the gateway A in the availability zone A receives a task processing request sent by a client, the gateway A will prebook function instances from the scheduling component A in the same availability zone and the scheduling component B in a different availability zone; the scheduling component A and the scheduling component B will respectively feed back function instance identifiers to the gateway A. Among them, the scheduling component usually determines function instances from the same availability zone and feeds back corresponding function instance identifiers. Similarly, the process of the gateway B in the availability zone B calling the scheduling component is similar to the calling process of the gateway A, which will not be described herein.Based on the content shown in FIG. 3, it can be known that the gateway accesses at least one cross availability zone scheduling component in the execution process of the task processing request, which causes the system side delay to at least include twice cross availability zone access. Since the network delay of the same region cross availability zone and the network delay of the availability zone can be different by 10 times at most. Therefore, in the case that each request needs to pass through the scheduling component, the cross availability zone call causes further growth of the system side scheduling delay. In actual scenarios, the above case can bring an additional 10ms+ system side average scheduling delay. And the business execution time of such business is very short, so that the existing task scheduling scheme cannot meet the needs of the business with high QPS and low function execution time, and at the same time affects the execution efficiency of such business. In view of the above problems, the embodiment of the present disclosure provides a task processing method, which pre-allocates computing resources for the gateway, so that the gateway can directly call the computing resources for task processing, solves the problem that each request needs to pass through the scheduling component in the prior art, causes the network delay to be large, and at the same time, since the gateway does not need to access the scheduling component in the process of executing the request, the scheduling component also does not need to feed back the call response to the gateway, thereby saving the server resources of the gateway and the scheduling component. The technical scheme of the present disclosure and how the technical scheme of the present disclosure solves the above technical problems will be described in detail in the specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present disclosure will be described below with reference to the drawings. First, the task processing system shown in the embodiment of the present disclosure is explained. It can be known in combination with the above FIG. 1 that the functions of the gateway and the scheduling component of the task processing system are adjusted in the present disclosure to support pre-allocation of computing resources for the gateway. The functions of the gateway and the scheduling component of the task processing system are adjusted so as to be implemented under the current existing system architecture without the need to change and adjust the hardware architecture. The task processing system shown in the embodiment includes a system component, and a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones. The system component is used to receive the task processing request sent by the client, and forward the task processing request to the gateway. It can be understood that when forwarding, the system component will forward the task processing request to at least one gateway considering the business type of the task processing request. For example: when the business has high QPS and low function execution time, the system component forwards the task processing request to at least two gateways deployed in different availability zones to improve the disaster recovery capability of such business.In an actual scenario, at least one gateway, a scheduling component, and a computing resource can be deployed in one availability zone, and the specific deployment quantity is determined according to actual conditions. In addition to the existing Reserve / Ide interface, the scheduling component in this embodiment is also configured with a lease interface. It can be understood that the lease interface can be a software interface or a hardware interface. The specific implementation is determined according to actual conditions. For example, the implementation of the software interface in the prior art can be used to implement the lease interface. The lease interface is used to receive a renewal request sent by the gateway. The computing resource in this embodiment can be, for example, a function instance in a Serverless architecture, which is mainly responsible for executing specific business logic of a user function. The gateway can be, for example, an API service gateway. The gateway can receive a task processing request sent by a client or a system component, and determine a corresponding computing resource to perform task processing based on the type of the task processing request. It can be understood that the computing resource determined by the gateway can be in the same availability zone as the gateway or in a different availability zone from the gateway, which is not limited in the present disclosure, and is determined according to actual conditions. In this embodiment, a temporary lease mechanism is introduced between the gateway and the scheduling component. The following takes a task processing scenario of a class RTA business as an example to explain and describe. Specifically, the system component can configure a temporary lease mechanism for the gateway and the scheduling component, and the temporary lease mechanism is used to indicate that the gateway signs up with the scheduling component for a mechanism for processing computing resources of a corresponding type of business. Under the temporary lease mechanism, the scheduling component can establish a temporary lease between the computing resource and the gateway, so that the gateway can directly call the computing resource with the temporary lease to perform task processing within a maximum signing duration. It can be understood that the scheduling component signs up the gateway for the computing resource, and the computing resource with the temporary lease will not be released when it is not processing a task, at which time the gateway has the ability to directly call the computing resource within the maximum signing duration. In one possible implementation, the gateway can record the real-time concurrency degree of the computing resource with the temporary lease. That is, the real-time concurrency degree of the computing resource with the temporary lease is controlled by the gateway. The real-time concurrency degree refers to the number of currently concurrently processed task processing requests of the computing resource. The real-time concurrency degree is not greater than the single-instance concurrency degree supported by the computing resource. If the real-time concurrency degree of the computing resource is less than the single-instance concurrency degree supported by the computing resource, it indicates that the computing resource is currently in an unsaturated state, and the computing resource can currently accept new task processing requests. If the real-time concurrency degree of the computing resource reaches the single-instance concurrency degree supported by the computing resource, it indicates that the computing resource is currently in a saturated state, and the computing resource cannot currently accept more task processing requests.It can be understood that the gateway can determine the real-time concurrency of the computing resource with the temporary lease, and allocate the task processing request to the computing resource based on the real-time concurrency of the computing resource. In a possible implementation, there is a single lease relationship between the computing resource and the gateway. The single lease relationship means that, for any one computing resource, the scheduling component leases it to one gateway. After leasing the computing resource to the gateway, the scheduling component will not lease the computing resource to other gateways as long as the computing resource is not released. It can be understood that the gateway can lease the computing resource to the scheduling component through the Reserve interface of the scheduling component, or lease the computing resource to the scheduling component through the lease interface of the scheduling component, and the present disclosure does not make special limitations on this. The system component shown in the embodiment can send configuration information for indicating that the target type of business is processed by the computing resource with the temporary lease to the gateway and the scheduling component, and the gateway and the scheduling component receive the configuration information sent by the system component, and configure the task processing scene of the target type of business according to the configuration information. The gateway can send a signing request to the scheduling component, and the signing request is used to request the scheduling component to allocate a computing resource for processing the target type of business; the scheduling component determines the corresponding computing resource based on the signing request, and establishes a temporary lease between the computing resource and the gateway, and the temporary lease is used to indicate that the gateway processes the task processing request of the target type of business through the computing resource within the signing duration. In the case of determining that the received task processing request is the task processing request of the target type of business, the gateway can determine the target computing resource from the computing resource with the temporary lease, and directly send the task processing request to the target computing resource for task processing. It can be understood that, due to the introduction of the temporary lease mechanism, the gateway does not need to access the scheduling component to determine the corresponding computing resource when processing the task processing request of the target type of business, thereby reducing the system side delay caused by accessing the scheduling component in the entire task processing process. At the same time, in the entire task processing process, the gateway does not need to send the corresponding predetermined request to the scheduling component, and the scheduling component does not need to determine the computing resource and feed back the computing resource identifier, thereby saving the server resources on the gateway and the scheduling component. FIG. 4 is a flow diagram of a possible task processing method provided by an embodiment of the present disclosure. The execution subject of the embodiment may, for example, be the gateway in the task processing system shown in the above embodiments. As shown in FIG. 4, the task processing method comprises the following steps.

[0005] S401: Receive the task processing request sent by the client. When the client needs the cloud environment to process the task, the client can send the task processing request to the cloud service platform. Different task types can correspond to different task processing requests. After receiving the task processing request sent by the client, the gateway needs to judge the task type corresponding to the task processing request to determine the corresponding computing resource for task processing.

[0006] S402: If the task processing request is a task processing request of a target type service, determine a target computing resource from the computing resources with temporary leases. The target type service may include, for example, a service with high QPS and low function execution time, and the performance form of the target type service is related to the architecture of the task processing system. For example, in a Serverless architecture, the target type service may be an RTA-like service. The computing resources include multiple types, and different types of computing resources are used to process task processing requests of corresponding types of services. In this step, the computing resources with temporary leases refer to the computing resources that the gateway signs from the scheduling component for processing the target type service, and the performance form of the computing resources is related to the architecture of the task processing system. For example, in a Serverless architecture, the computing resources may be function instances. The number of computing resources with temporary leases may be one or more. The gateway may determine the target computing resource from one or more computing resources with temporary leases. It can be understood that the gateway may determine the target computing resource from the computing resources with temporary leases by using a preset determination strategy. The determination strategy may include, for example, determining a computing resource in the same availability zone as the gateway as the target computing resource, or determining a computing resource with a network delay between the gateway and the computing resource less than a preset network delay as the target computing resource, or determining a computing resource with a real-time concurrency degree less than a supported single-instance concurrency degree as the target computing resource. The present disclosure does not limit this, and the actual situation is specific. S403: Send the task processing request to the target computing resource for task processing. The task processing method provided in this embodiment can determine the target computing resource from the computing resources with temporary leases when the task processing request is of the target type service, and send the task processing request directly to the target computing resource for task processing, so that the gateway does not need to access the scheduling component to determine the corresponding computing resource when processing the task processing request of the target type service, thereby reducing the system-side delay caused by accessing the scheduling component in the task processing process, and improving the processing efficiency of the target type service. At the same time, since the gateway of this embodiment does not need to send a corresponding predetermined request to the scheduling component, the scheduling component also does not need to determine the computing resource and feed back the computing resource identifier, thereby saving the server resources on the gateway and the scheduling component. FIG. 5 is a flowchart of a possible task processing method provided in an embodiment of the present disclosure. This embodiment is a detailed description of a possible implementation manner of the gateway determining the target computing resource from the computing resources with temporary leases based on the above-mentioned embodiment of FIG. 4.As shown in FIG. 5, the method comprises the following steps.

[0007] S501: Obtain the real-time concurrency of the computing resource. The real-time concurrency refers to the number of task processing requests currently processed by the computing resource in parallel. Since the computing resource is the computing resource contracted by the gateway from the scheduling component, that is, the scheduling resource leases the computing resource to the gateway for task processing within the contract duration. Therefore, in the case where the gateway determines that the task processing request is a task processing request of the target type service, the gateway needs to allocate the corresponding computing resource for the task processing of the task processing request, that is, the real-time concurrency of the computing resource is controlled by the gateway. Based on this, the gateway can count the real-time concurrency of the computing resource.

[0008] S502: determining whether there is available computing resource from the computing resource based on the real-time concurrency of the computing resource and the supported single-instance concurrency, wherein the single-instance concurrency can be used to represent the concurrent processing capability of the corresponding computing resource, and the single-instance concurrency supported by different computing resources can be the same or different. The available computing resource can indicate, for example, the computing resource that is not currently in a saturated state. After obtaining the real-time concurrency of the computing resource, the computing resource that is not currently in a saturated state can be determined according to the real-time concurrency of the computing resource and the single-instance concurrency supported by the computing resource. For example, the available computing resource can be determined in a case where the real-time concurrency does not reach the single-instance concurrency; the available computing resource can also be determined in a case where the ratio of the real-time concurrency to the single-instance concurrency does not reach a preset ratio; or the effective real-time concurrency quantity can be determined according to the single-instance concurrency, and the available computing resource can be determined in a case where the real-time concurrency is less than the effective real-time concurrency quantity. The specific implementation manner of determining whether there is available computing resource is not specially limited in the present disclosure. In a possible implementation manner, the step of determining whether there is available computing resource can include, for example: determining whether the real-time concurrency of the computing resource reaches the supported single-instance concurrency; if the real-time concurrency of the computing resource does not reach the single-instance concurrency supported by the computing resource, it indicates that the computing resource can currently undertake and process a new task processing request, and the computing resource is determined as available computing resource. If the real-time concurrency of the computing resource reaches the single-instance concurrency supported by the computing resource, it indicates that the computing resource cannot currently undertake and process a new task processing request, and the computing resource is determined as unavailable computing resource. It can be understood that, in a case where the number of computing resources is one, if the real-time concurrency of the computing resource reaches the supported single-instance concurrency, it is determined that there is no available computing resource in the computing resource. In a case where the number of computing resources is multiple, if the real-time concurrency of each computing resource reaches the supported single-instance concurrency, it is determined that there is no available computing resource in the computing resource.

[0009] S503: determining the target computing resource from the available computing resources. In a case where the number of available computing resources is one, the available computing resource is determined as the target computing resource. In a case where the number of available computing resources is multiple, the target computing resource can be determined from the multiple available computing resources by using the determination strategy in step S402 described above. For details, refer to the description of step S402 above, which will not be repeated here.

[0010] S504: continuing to sign a computing resource from the scheduling component, and taking the signed computing resource as the target computing resource. In a case where it is determined that there is no available computing resource, that is, the computing resources with temporary tenancy signed by the gateway from the scheduling component cannot currently accept and process new task processing requests, the gateway needs to continue to sign a new computing resource from the scheduling component for calling. For details of the signing process, refer to the following embodiment. After the gateway signs a new computing resource from the scheduling component, the newly signed computing resource can be taken as the target computing resource. Since the gateway immediately signs a new computing resource from the scheduling component when it is determined that the computing resource signed by the gateway currently has no available computing resource for accepting and processing new task processing requests, the deficiency of insufficient elasticity in the prior art is overcome, the high elasticity of the Serverless system architecture under the conventional mechanism is retained, and the effect that the computing resource required for expansion can be quickly expanded when facing impulse traffic is achieved. It should be understood that in a case where there is no target computing resource, the gateway can also request a computing resource for executing a task processing request from the scheduling component in the existing manner, and perform the signing process of the computing resource from the scheduling component asynchronously. In this signing scenario, the computing resource requested from the scheduling component in the existing manner can be within the signing range or not within the signing range. The present disclosure does not limit this. The task processing method provided in this embodiment is because the real-time concurrency of the computing resource is managed by the gateway, so that the real-time concurrency of the computing resource can be counted by the gateway, and the counting result is still accurate. At the same time, the gateway can determine the available computing resource according to the real-time concurrency of the computing resource and the supported single-instance concurrency, so as to ensure that the task processing request can be executed by the computing resource. FIG. 6 is an interaction schematic diagram I of a possible task processing method provided in an embodiment of the present disclosure. This embodiment is a supplementary description of a possible implementation manner of signing a computing resource by the gateway based on the above-mentioned embodiment of FIG. 4 or the embodiment of FIG. 5. As shown in FIG. 6, the method comprises:

[0011] S601: The gateway sends a lease request to the scheduling component. The lease request is used to request the scheduling component to allocate a computing resource for processing a target type of service. It can be understood that the timing of the gateway sending the lease request to the scheduling component may, for example, include: all computing resources with temporary leases are unavailable computing resources, or the number of available computing resources among the computing resources with temporary leases is less than a preset threshold, or the gateway currently does not have a leased computing resource. The case where the gateway currently does not have a leased computing resource may, for example, include: before the gateway enables the temporary lease mechanism, or the gateway currently does not have a computing resource with a temporary lease, etc. The gateway can send the lease request to the scheduling component through a lease interface deployed on the scheduling component, or through a Reserve / Ide interface deployed on the scheduling component. The present disclosure does not limit this. In the case where the gateway leases a computing resource from the scheduling component through the Reserve interface of the scheduling component, the task processing system may, for example, continue to use the existing task processing system, so that the task processing method provided in the embodiment has good adaptability, and the cost of using the existing task processing system is also low. In the case where the gateway leases a computing resource from the scheduling component through the lease interface of the scheduling component, the pressure on the Reserve interface of the scheduling component can be relieved, thereby improving the data processing efficiency of the service.

[0012] S602: The scheduling component determines a computing resource for processing the target type service from the computing resources without temporary tenancy and for processing the target type service, and establishes a temporary tenancy between the computing resource and the gateway. Different service types of services correspond to different computing resources. When the scheduling component receives the subscription request sent by the gateway, the scheduling component determines at least one computing resource corresponding to the target type service according to the subscription request, and determines a computing resource without temporary tenancy from the at least one computing resource; and then establishes a temporary tenancy between the computing resource and the gateway. It can be understood that when the scheduling component establishes a temporary tenancy between the computing resource and the gateway, that is, the scheduling component leases the computing resource to the gateway, a single tenancy relationship is formed between the computing resource and the gateway. For example: Gateway 1 sends a subscription request to the scheduling component, and the scheduling component establishes a temporary tenancy between computing resource 1 and gateway 1. Computing resource 1 can be directly called by gateway 1 within the subscription duration, and other gateways cannot use computing resource 1 to process tasks within the subscription duration. The purpose of forming a single tenancy relationship between the computing resource and the gateway is to facilitate the gateway to manage the real-time concurrency of the subscribed computing resources, so as to reserve the accurate concurrency statistical ability under the Serverless system architecture.

[0013] S603: The scheduling component feeds back a subscription response to the gateway. The subscription response is used to indicate the computing resource subscribed by the gateway, the indicated computing resource for subscription is determined from the computing resource without a temporary lease, and is used to process the target type of service. After the scheduling component establishes a temporary lease between the computing resource and the gateway, the scheduling component feeds back the subscription response to the gateway, which carries the identifier of the computing resource, so that the gateway determines and invokes the corresponding computing resource for task processing according to the identifier of the computing resource. It can be understood that in a single subscription process, the number of computing resources subscribed by the scheduling component to the gateway can be one or multiple. The number of computing resources for subscription can be determined by the scheduling component itself or indicated by the gateway. In a possible implementation, the scheduling component can determine the number of computing resources for subscription to the gateway according to a preset subscription strategy. The subscription strategy may, for example, include: determining the number of computing resources for subscription according to the usage of the resources of the scheduling component, or determining the number of computing resources for subscription according to the type of service corresponding to the subscription request, or determining the number of computing resources for subscription according to the current usage of at least one computing resource corresponding to the target type of service, or determining the number of computing resources for subscription according to the number of computing resources in the availability zone where the gateway is located, and the like. The present disclosure does not limit the subscription strategy, and the actual situation is used for reference. In a possible implementation, the subscription request may, for example, include: the number of computing resources to be subscribed. At this time, the scheduling component may, for example, subscribe to the gateway the number of computing resources to be subscribed based on the indication of the gateway. Meanwhile, in the case where the number of computing resources to be subscribed is indicated by the gateway, the scheduling component may still determine the number of computing resources for subscription to the gateway according to the preset subscription strategy. The task processing method provided in this embodiment, by subscribing to the scheduling component the computing resource used to process the target type of service, so that the gateway can directly invoke the computing resource for task processing when processing the task processing request of the target type of service, without the need to access the scheduling component to determine the corresponding computing resource, thereby reducing the system side delay caused by accessing the scheduling component in the task processing process, and improving the processing efficiency of the target type of service. Meanwhile, since the gateway of this embodiment does not need to send a corresponding predetermined request to the scheduling component, and the scheduling component does not need to determine the computing resource and feed back the identifier of the computing resource, the server resources on the gateway and the scheduling component are saved. FIG. 7 is an interaction schematic diagram II of a possible task processing method provided in this embodiment of the present disclosure. This embodiment is a supplementary description of a possible implementation of the gateway renewing the computing resource on the basis of any of the embodiments of FIGS. 4-6.As shown in FIG. 7, the method comprises the following steps.

[0014] S701: The gateway records the signing duration of the computing resource with the temporary lease. The signing duration can be used to indicate the real-time signing duration of the computing resource, and the gateway can record the real-time signing duration of the computing resource signed by itself. The starting time of the signing duration can be calculated from the time when the gateway receives the signing response or the renewal response sent by the scheduling component, or can be determined by the signing start time carried in the signing response or the renewal response, and the present disclosure does not make any limitation thereon.

[0015] S702: In the case where the recorded signing duration reaches the maximum signing duration of the computing resource, the gateway sends a renewal request to the scheduling component. The maximum signing duration is used to indicate the maximum duration that the gateway can directly schedule the computing resource to process tasks in a single signing. The maximum signing durations of different computing resources can be the same or different. The renewal request can be used to request the scheduling component to perform a renewal operation on the computing resource. If the gateway determines that the recorded signing duration reaches the maximum signing duration of the computing resource, it means that the computing resource has reached the maximum duration that can be directly scheduled, and the gateway cannot continue to directly schedule the computing resource to process tasks in the subsequent duration. At this time, the gateway can send a renewal request to the scheduling component to perform a renewal operation on the computing resource. The renewal request can include the computing resource identifier, for example.

[0016] S703: The scheduling component determines the computing resource corresponding to the renewal request, and performs a renewal operation on the temporary lease between the computing resource and the gateway. After receiving the renewal request sent by the gateway, the scheduling component determines the computing resource to be renewed, and performs a renewal operation on the temporary lease between the computing resource to be renewed and the gateway. The renewal operation can be, for example, re-establishing the temporary lease between the computing resource and the gateway, or updating the signing duration of the temporary lease based on the established temporary lease, and the present disclosure does not make any special limitation on the specific steps of the renewal operation, as long as the renewal operation is performed. In one possible implementation, the scheduling component side can also record the signing duration of the computing resource with the temporary lease. When the scheduling component determines the computing resource to be renewed, the scheduling component can perform a zeroing operation on the signing duration of the computing resource recorded by itself, or can synchronize the signing duration of the computing resource recorded by itself to the gateway when the signing duration recorded by itself does not match the computing resource identifier in the renewal request. As long as the signing durations between the gateway and the scheduling component are matched, it is acceptable.

[0017] S704: The scheduling component feeds back a renewal response of the computing resource to the gateway. S705: The gateway re-records the lease duration of the computing resource. Wherein, the renewal response is used to indicate that the computing resource has been renewed. After the gateway receives the renewal response fed back by the scheduling component, the gateway needs to re-record the lease duration of the computing resource, so as to perform the renewal operation again when the subsequent lease duration reaches the maximum lease duration. In a possible implementation, the maximum lease duration may be set by the scheduling component or determined by the gateway and the system component. In the case where the maximum lease duration is set by the scheduling component, the renewal response fed back by the scheduling component to the gateway may also be used to indicate the maximum lease duration of the computing resource, and the gateway may determine the maximum lease duration of the computing resource according to the received renewal response. In the case where the maximum lease duration is determined by the gateway and the system component, the gateway may determine the maximum lease duration through the maximum lease duration setting parameter sent by the system component. The task processing method provided in this embodiment records the lease duration of the computing resource with a temporary lease, and performs the renewal operation to the scheduling component when the lease duration reaches the maximum lease duration, thereby ensuring that the gateway has computing resources that can be directly invoked at any time. Meanwhile, in the task processing process, the gateway only needs to perform the renewal operation once when there is no available computing resource, and can directly invoke the computing resource for task processing within the maximum lease duration, thereby reducing the system side delay caused by accessing the scheduling component in the task processing process, and improving the processing efficiency of the target type business. In a possible implementation, when the number of task processing requests processed by the gateway is small and the number of computing resources with a temporary lease is large, the gateway may also request the scheduling component to release the idle computing resource. The specific process of releasing the computing resource is as follows: in the case where the gateway determines that the real-time concurrency degree of the computing resource is less than or equal to the preset concurrency degree, the gateway may record the duration that the real-time concurrency degree of the computing resource is less than or equal to the preset concurrency degree; if the duration reaches the preset duration, the gateway sends a release request to the scheduling component, so that the scheduling component releases the computing resource. Wherein, the preset concurrency degree may be preset by the system component or set by the gateway according to the actual situation. The release request is used to request the scheduling component to perform the release operation on the computing resource. FIG. 8 is a flowchart of a possible task processing method provided in an embodiment of the present disclosure. This embodiment is applied to a cloud platform, and the cloud platform includes a system component, a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones. As shown in FIG. 8, the task processing method provided in this embodiment includes:

[0018] S801: The system component receives a task processing request sent by a client;

[0019] S802: The system component forwards the task processing request to at least two target gateways in a plurality of gateways located in different availability zones.

[0020] S803: In the case where the target gateways determine that the task processing request is a task processing request of a target type of business, the target gateways determine a target computing resource from a computing resource with a temporary lease; the computing resource with a temporary lease is a computing resource subscribed from a scheduling component.

[0021] S804: The target gateway A sends the task processing request to the target computing resource A for task processing.

[0022] S805: The target gateway B sends the task processing request to the target computing resource B for task processing. There is no explicit time sequence relationship between step S804 and step S805. Step S804 can be performed first, and then step S805 can be performed. Step S805 can be performed first, and then step S804 can be performed. Step S804 and step S805 can be performed at the same time. The present disclosure does not limit this. The method of the embodiment of the present disclosure is applied to the task processing system in the foregoing embodiments. The implementation principle and technical effects are described in the foregoing embodiments, and will not be described here. FIG. 9 is a structural schematic diagram one of a possible task processing apparatus provided by the embodiment of the present disclosure. The task processing apparatus is applied to a gateway. As shown in FIG. 9, the apparatus 900 includes: a receiving module 901 configured to receive a task processing request sent by a client; a processing module 902 configured to, in a case where the task processing request is a target type business task processing request, determine a target computing resource from computing resources with a temporary lease; the computing resource with the temporary lease is a computing resource signed from a scheduling component; and a sending module 903 configured to send the task processing request to the target computing resource for task processing. In a possible implementation manner, the apparatus further includes: an obtaining module 904 and a judging module 905; the obtaining module 904 is configured to obtain a real-time concurrency degree of the computing resource; the judging module 905 is configured to determine, based on the real-time concurrency degree of the computing resource and a supported single-instance concurrency degree, whether there is an available computing resource in the computing resource; and the processing module 902 is further configured to, in a case where there is a computing resource, determine the target computing resource from the available computing resource. In a possible implementation manner, the judging module 905 is specifically configured to judge whether the real-time concurrency degree reaches the single-instance concurrency degree; the processing module 902 is further configured to, in a case where the real-time concurrency degree does not reach the single-instance concurrency degree, determine that the computing resource is an available computing resource; and in a case where the real-time concurrency degree reaches the single-instance concurrency degree, determine that the computing resource is an unavailable computing resource. In a possible implementation manner, the processing module 902 is further configured to, in a case where there is no available computing resource, continue to sign a computing resource from the scheduling component, and take the signed computing resource as the target computing resource.In a possible implementation, the sending module 903 is further configured to send a subscription request to the scheduling component, where the subscription request is used to request the scheduling component to allocate a computing resource for processing the target type of service; and the receiving module 901 is further configured to receive a subscription response fed back by the scheduling component, where the subscription response is used to indicate a subscribed computing resource, and the indicated subscribed computing resource is determined from a computing resource without a temporary lease and a computing resource for processing the target type of service. In a possible implementation, the processing module 902 is further configured to record a subscription duration of the computing resource with the temporary lease; the sending module 903 is further configured to send a renewal request to the scheduling component when the recorded subscription duration reaches a maximum subscription duration of the computing resource, where the renewal request is used to request the scheduling component to perform a renewal operation on the computing resource; the receiving module 901 is further configured to receive a renewal response fed back by the scheduling component; and the processing module 902 is further configured to record the subscription duration of the computing resource again. In a possible implementation, the subscription response is further used to indicate the maximum subscription duration of the computing resource; or the processing module 902 is further configured to negotiate the maximum subscription duration of the computing resource with a system component. In a possible implementation, the processing module 902 is further configured to, when it is determined that real-time concurrency of the computing resource meets a preset condition, record a duration during which the real-time concurrency of the computing resource meets the preset condition, where the preset condition includes that the real-time concurrency is less than or equal to a preset concurrency; and the sending module 903 is further configured to, when the duration reaches a preset duration, send a release request to the scheduling component, where the release request is used to request the scheduling component to perform a release operation on the computing resource. In a possible implementation, the receiving module 901 is further configured to receive configuration information sent by a system component, where the configuration information is used to indicate that a computing resource with a temporary lease is used to process a target type of task; and the processing module 902 is further configured to configure a task processing scenario of the target type of service according to the configuration information. FIG. 10 is a structural schematic diagram two of a possible task processing apparatus provided by an embodiment of the present disclosure.The task processing apparatus is applied to a scheduling component, as shown in FIG. 10, the apparatus 1000 comprises: a receiving module 1001, configured to receive a subscription request sent by a gateway, the subscription request is used for requesting to allocate a computing resource for processing a target type service; a processing module 1002, configured to determine a computing resource for processing the target type service from a computing resource without a temporary lease and used for processing the target type service, and establish a temporary lease of the computing resource and the gateway; and a sending module 1003, configured to feed back a subscription response to the gateway, the subscription response is used for indicating the computing resource with the temporary lease. In a possible implementation manner, the receiving module 1001 is further configured to receive a renewal request sent by the gateway through the lease interface, the renewal request is used for requesting a renewal operation on the computing resource; the processing module 1002 is further configured to determine the computing resource corresponding to the renewal request, and perform a renewal processing on the temporary lease of the computing resource and the gateway; and the sending module 1003 is further configured to feed back a renewal response of the computing resource to the gateway. In a possible implementation manner, the receiving module 1001 is further configured to receive a release request sent by the gateway, the release request is used for requesting a release operation on the computing resource; the processing module 1002 is further configured to determine the computing resource corresponding to the release request, release the temporary lease of the computing resource and the gateway, and perform the release operation on the computing resource. FIG. 11 is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. As shown in FIG. 11, the electronic device 1100 can comprise at least one processor 1101, at least one storage medium 1102, and a communication interface 1103. The at least one processor 1101 is configured to implement the task processing method provided by the above-mentioned embodiments of the present disclosure. The at least one storage medium 1102 is configured to store program instructions and / or data. The storage medium 1102 and the processor 1101 are coupled. The coupling in the embodiment of the present disclosure is indirect coupling or communication connection between devices, units or modules, which can be electrical, mechanical or other forms, and is used for information interaction between devices, units or modules. The processor 1101 can operate in cooperation with the storage medium 1102. The processor 1101 can execute the program instructions stored in the storage medium 1102. At least one of the at least one storage medium can be included in the processor. The communication interface 1103 is configured to communicate through a transmission medium and other devices, so as to enable the electronic device 1100 to communicate with other devices. The communication interface 1103 can be, for example, a transceiver, an interface, a bus, a circuit or a device capable of realizing the transceiving function.The processor 1101 can transceive data and / or information by using the communication interface 1103, and be used for implementing the method provided by the embodiments of the present disclosure. For details, refer to the detailed description in the foregoing embodiments, which will not be repeated here. The specific connection medium between the processor 1101, the memory 1102 and the communication interface 1103 is not limited in the embodiments of the present disclosure. In FIG. 11, the processor 1101, the memory 1102 and the communication interface 1103 are connected through a bus 1104. The connection mode between other components is only illustrative and is not limited. The bus can be divided into an address bus, a data bus, a control bus and the like. For convenience of representation, only one thick line is used in FIG. 11, but it does not mean that there is only one bus or only one type of bus. It should be understood that the processor in the embodiments of the present disclosure can be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above method embodiments can be completed by integrated logic circuits or software form instructions in the processor. The processor mentioned above can be a general purpose processor, a digital signal processor (digital signal processor, DSP), an application specific integrated circuit (application specific integrated circuit, ASIC), a field programmable gate array (field programmable gate array, FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block disclosed in the embodiments of the present disclosure can be implemented or executed. The general purpose processor can be a microprocessor or the processor can be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present disclosure can be directly embodied in hardware code processor execution, or executed by a combination of hardware and software modules in the code processor. The software modules can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register or other mature storage medium in the art. The storage medium is located in the memory, and the processor reads the information in the memory, and combines the hardware to complete the steps of the above method. It should also be understood that the memory in the embodiments of the present disclosure can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.The non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache memory. By way of example, and not limitation, a RAM can be a static RAM (SRAM), a dynamic RAM (DRAM), a synchronous DRAM (SDRAM), a double data rate SDRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a synchlink DRAM (SLDRAM), or a direct rambus RAM (DR RAM). It should be noted, however, that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory. The present disclosure also provides a computer-readable storage medium having computer program (which can also be referred to as code, or instructions) stored therein. The computer program, when executed by a processor, causes a computer to perform the task processing method in any of the preceding embodiments. The present disclosure also provides a computer program product comprising the computer program, which, when executed by a processor, implements the task processing method in any of the preceding embodiments. Those skilled in the art can realize that the various illustrative logical blocks and steps described in connection with the embodiments disclosed herein can be implemented by electronic hardware, or by computer software and electronic hardware in combination. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present disclosure.The steps in the embodiments provided by the present disclosure can be implemented by different execution units or execution modules, or by different devices. It should be understood that the disclosed apparatuses, devices and methods can be implemented in other manners. For example, the embodiments of the method described above can be implemented by the corresponding apparatuses. The division of the units in the apparatuses is merely a logical division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units can be indirect couplings or communication connections through some interfaces, and can be electrical, mechanical or other forms. The units shown as separate components can or can not be physically separate, and components shown as units can or can not be physical units. That is, they can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments. In addition, the various function units in the embodiments of the present disclosure can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. In the above embodiments, the functions of each unit can be fully or partially implemented by software, hardware, firmware, or any combination thereof. When implemented by software, the whole or part of the computer program product can be realized in the form of a computer program product. The computer program product includes one or more computer instructions (program). When the computer program instructions (program) are loaded and executed on the computer, the whole or part of the processes or functions according to the embodiments of the present disclosure are produced. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired (such as a coaxial cable, optical fiber, digital subscriber line (DSL) or wireless (such as infrared, wireless, microwave, etc.) manner.The computer-readable storage medium can be any available medium or data storage that is accessible by a computer, or a data storage device such as a server, data center, etc. integrated with one or more sets of such available media. The medium can be a magnetic medium, (e.g., a floppy diskette, a hard disk drive, a magnetic tape), an optical medium, (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)), etc. If implemented in the form of a software function 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 solutions of the present disclosure essentially or the part that makes a contribution to the prior art or the part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the embodiments of the present disclosure. The aforementioned storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, etc. Various media that can store program codes. The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present disclosure are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant national and regional laws, regulations and standards, and provide corresponding operation entrances for users to choose authorization or refusal. The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present disclosure, which should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

CLAIM 1. A task processing method in which, The method applied to a gateway comprises: receiving a task processing request sent by a client; if the task processing request is a task processing request of a target type service, determining a target computing resource from computing resources with temporary leases; the computing resource with temporary leases is a computing resource leased from a scheduling component; and sending the task processing request to the target computing resource for task processing.

2. The method of claim 1, wherein, The determining of the target computing resource from the computing resources comprises: obtaining real-time concurrency of the computing resources; determining whether there is an available computing resource from the computing resources based on the real-time concurrency of the computing resources and supported single-instance concurrency; and if there is an available computing resource, determining the target computing resource from the available computing resource.

3. The method according to claim 2, wherein, The determining of whether there is an available computing resource from the computing resources based on the real-time concurrency of the computing resources and supported single-instance concurrency comprises: judging whether the real-time concurrency reaches the single-instance concurrency; if the real-time concurrency does not reach the single-instance concurrency, determining that the computing resource is an available computing resource; and if the real-time concurrency reaches the single-instance concurrency, determining that the computing resource is an unavailable computing resource.

4. The method according to claim 2 or 3, wherein, The method further comprises: if there is no available computing resource, continuing to lease a computing resource from the scheduling component, and taking the leased computing resource as the target computing resource.

5. The method according to any one of claims 1 to 4, wherein, The leasing of the computing resource from the scheduling component comprises: sending a lease request to the scheduling component, the lease request being used to request the scheduling component to allocate a computing resource for processing the target type service; and receiving a lease response fed back by the scheduling component, the lease response being used to indicate a leased computing resource, the indicated leased computing resource being determined from computing resources without temporary leases and used for processing the target type service.

6. The method according to any one of claims 1 to 5, wherein, The method further comprises: recording a lease duration of the computing resource with temporary leases; when the recorded lease duration reaches a maximum lease duration of the computing resource, sending a renewal request to the scheduling component, the renewal request being used to request the scheduling component to perform a renewal operation on the computing resource; receiving a renewal response fed back by the scheduling component, and re-recording the lease duration of the computing resource. The lease response is further used to indicate the maximum lease duration of the computing resource.

7. The method according to claim 5, wherein, Alternatively, the method further comprises: negotiating the maximum lease duration of the computing resource with a system component. ​ 8. The method according to any one of claims 2-4, wherein, After the real-time concurrency of the computing resource is acquired, the method further includes: in a case where the real-time concurrency of the computing resource meets a preset condition, recording a duration for which the real-time concurrency of the computing resource meets the preset condition, the preset condition including: the real-time concurrency being less than or equal to a preset concurrency; and in a case where the duration reaches a preset duration, sending a release request to the scheduling component, the release request being used to request the scheduling component to perform a release operation on the computing resource.

9. The method according to any one of claims 1-8, wherein, Before the task processing request sent by the client is received, the method further includes: receiving configuration information sent by a system component, the configuration information being used to indicate that a computing resource with a temporary lease is used to perform task processing on a target service type; and configuring a task processing scenario of the target type service according to the configuration information.

10. A task processing method in which, The method applied to the scheduling component includes: receiving a subscription request sent by a gateway, the subscription request being used to request an allocation of a computing resource used to process a target type service; determining a computing resource used to process the target type service from computing resources without a temporary lease and used to process the target type service, and establishing a temporary lease of the computing resource and the gateway; and feeding back a subscription response to the gateway, the subscription response being used to indicate the computing resource with the temporary lease.

11. The method according to claim 10, wherein, The scheduling component is configured with a lease interface, and the method further includes: receiving a renewal request sent by the gateway through the lease interface, the renewal request being used to request a renewal operation on a computing resource; determining the computing resource corresponding to the renewal request, and performing renewal processing on the temporary lease of the computing resource and the gateway; and feeding back a renewal response of the computing resource to the gateway.

12. The method according to claim 10 or 11, wherein, The method further includes: receiving a release request sent by the gateway, the release request being used to request a release operation on a computing resource; determining the computing resource corresponding to the release request, canceling the temporary lease of the computing resource and the gateway, and performing the release operation on the computing resource. The method applied to a cloud platform includes: a system component, and a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones, and the method includes: the system component receiving a task processing request sent by a client, and forwarding the task processing request to at least two target gateways in the plurality of gateways; in a case where the target gateways determine that the task processing request is a task processing request of a target type service, the target gateways determine a target computing resource from computing resources with a temporary lease; the computing resource with the temporary lease is a computing resource subscribed from a scheduling component; and the target gateways send the task processing request to the target computing resource for task processing.

13. A task processing method in which, The method includes:

14. An electronic device, wherein, ​ A memory, a processor and a communication interface; wherein the memory stores computer-executable instructions, and the communication interface is configured to communicate with other devices through a transmission medium; The processor executes the computer-executable instructions stored in the memory to implement the task processing method according to any one of claims 1-13.

15. A task processing system, wherein, The task processing system comprises: a system component, and a plurality of gateways, a plurality of computing resources and a plurality of scheduling components located in different availability zones, the system component is configured to receive a task processing request sent by a client, and forward the task processing request to at least two target gateways in the plurality of gateways; the target gateway is configured to, in a case where it is determined that the task processing request is a task processing request of a target type of business, determine a target computing resource from computing resources with a temporary lease; the computing resources with the temporary lease are computing resources subscribed from a scheduling component, and the target computing resource is in the same availability zone as the target gateway; and the target gateway is further configured to send the task processing request to the target computing resource for task processing.

16. A computer-readable storage medium, wherein, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the task processing method according to any one of claims 1-13.

17. A computer program product comprising a computer program, which, when executed by a processor, implements the task processing method according to any one of claims 1-13.

Images