Chain function scheduling system and method for server-free computing

Abstract

The invention provides a chained function scheduling system and method for server-free computing. The system comprises a function reflector, a function composer and a multi-stage function router, the method comprises the following steps: acquiring a function routing table of a multi-level function router, and constructing a full function routing table; generating a function chain calling request according to the business logic of the target task and the full function routing table; the business logic is determined according to the task request of the target task; in a heterogeneous server-free environment, according to the function chain calling request, calling a plurality of target functions in sequence; the objective functions are used for executing business logic. According to the method, the function routing table of the multi-level function router is converged, the full function routing table is constructed, the function chain calling request is generated on the basis, and the multiple target functions are called in sequence, so that the function calling process is effectively simplified, the efficiency and reliability of function calling are improved, the expandability is high, and the method is suitable for a complex server-free computing scene.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing, and in particular to a chain function scheduling system and method for serverless computing. Background Art

[0002] With the rapid development of cloud computing technology, cloud computing has gradually evolved from a virtual machine-based computing paradigm to a serverless and Function as a Service (FaaS) paradigm. The core of this new computing paradigm is function computing, which orchestrates and calls multiple functions together according to business processes through function chaining. Existing technologies primarily use two approaches to invoking function computing: client-side orchestration or cloud-hosted orchestration. Client-side orchestration is time-consuming and labor-intensive, resulting in low efficiency. Cloud-hosted orchestration typically only supports function computing on a single cloud provider, resulting in vendor lock-in, poor reliability, and difficulty adapting to complex serverless computing scenarios. Furthermore, with the development of edge computing and the Internet of Things, function computing scenarios are becoming increasingly complex and heterogeneous. Functions may be deployed in diverse computing environments, including public clouds, private clouds, and edge devices. In such complex, heterogeneous environments, efficiently and reliably chaining functions has become a pressing technical challenge. Summary of the Invention

[0003] The purpose of the present invention is to provide a chain function scheduling system and method for serverless computing, so as to solve the technical problems in the prior art of low function call efficiency, poor reliability, and difficulty in adapting to complex heterogeneous serverless computing scenarios.

[0004] The present invention provides a chain function scheduling system for serverless computing, comprising: Function reflector, used to obtain the function routing table of multi-level function routers and build the full function routing table; A function arranger, configured to generate a function chain call request according to the business logic of the target task and the full function routing table; the business logic is determined according to the task request corresponding to the target task; A multi-level function router is used to call multiple target functions in sequence according to the function chain call request in a heterogeneous serverless computing environment; the multiple target functions are used to execute the business logic.

[0005] In some embodiments, the multi-level function router includes: A head-end function router, configured to call the first target function according to the function chain call request, and forward the function chain call request to an intermediate function router; an intermediate function router, configured to receive the function chain call request sent by the head-end function router, call the second target function according to the function chain call request, and forward the function chain call request to the end function router; The end function router is used to receive the function chain call request sent by the intermediate function router, call the third target function according to the function chain call request, obtain the function chain call result, and send the function chain call result to the function orchestrator.

[0006] In some embodiments, the head-end function router is also used to send a function routing request to the intermediate function reflector and receive the address information of the intermediate function router sent by the function reflector; the intermediate function router is also used to send a function routing request to the function reflector and receive the address information of the end function router sent by the function reflector.

[0007] In some embodiments, the function routing table includes a function name, a function access credential, and function metadata; the function chain call request includes a function call chain and a corresponding request access credential.

[0008] In some embodiments, calling multiple target functions according to the function chain call request includes: Determining, according to the function chain call request, request access credentials for each target function corresponding to the multi-level function router; Determining, according to the function routing table, function access credentials of each target function corresponding to the multi-level function router; The request access credential of each target function is compared with the function access credential, and the request access credential of each target function is verified according to the comparison result. If the verification passes, each target function is called.

[0009] The present invention also provides a chained function scheduling method for serverless computing, characterized in that it is applied to any of the chained function scheduling systems for serverless computing described above, and the method includes: Obtain the function routing table of the multi-level function router and build the full function routing table; Generate a function chain call request according to the business logic of the target task and the full function routing table; the business logic is determined according to the task request corresponding to the target task; In a heterogeneous serverless computing environment, multiple target functions are called in sequence according to the function chain call request; the multiple target functions are used to execute the business logic.

[0010] In some embodiments, calling multiple target functions in sequence according to the function chain call request includes: The head-end function router calls the first target function according to the function chain call request, and forwards the function chain call request to the intermediate function router; The intermediate function router receives the function chain call request sent by the head-end function router, calls the second target function according to the function chain call request, and forwards the function chain call request to the end function router; The end function router receives the function chain call request sent by the intermediate function router, calls the third target function according to the function chain call request, obtains a function chain call result, and sends the function chain call result to the function orchestrator.

[0011] In some embodiments, before forwarding the function chain call request to the intermediate function router, the method further includes: Sending a function routing request to the function reflector, and receiving address information of the intermediate function router sent by the function reflector; Before forwarding the function chain call request to the terminal function router, the method further includes: Send a function routing request to the function reflector, and receive the address information of the terminal function router sent by the function reflector.

[0012] In some embodiments, the function routing table includes a function name, a function access credential, and function metadata; the function chain call request includes a function call chain and a corresponding request access credential.

[0013] In some embodiments, calling multiple target functions in sequence according to the function chain call request includes: Determining, according to the function chain call request, request access credentials for each target function corresponding to the multi-level function router; Determining, according to the function routing table of the multi-level function router, the function access credentials of each target function corresponding to the multi-level function router; The request access credential of each target function is compared with the function access credential, and the request access credential of each target function is verified according to the comparison result. If the verification passes, each target function is called.

[0014] The chained function scheduling system and method for serverless computing provided by the present invention have the following beneficial effects: (1) The function routing tables of multi-level function routers are aggregated through the function reflector to build a full function routing table, providing a rich and comprehensive data foundation for function chain calls, so that the function orchestrator can flexibly select the appropriate function combination according to business needs; (2) The multi-level function router architecture supports cross-environment function calls, effectively solving the problem of function calls in heterogeneous serverless computing environments and avoiding cloud vendor lock-in; (3) By collaborating with the function orchestrator and the multi-level function router, the delay caused by multiple function calls and responses is reduced, thereby significantly improving the overall performance of chained function calls. (4) Introduce a function access credential verification mechanism to ensure the security and controllability of function calls and prevent unauthorized function access; (5) The system architecture has good scalability and can support more complex function call topologies by adding function router nodes. It is suitable for various complex serverless computing scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 It is a structural diagram of a chain function scheduling system for serverless computing provided by an embodiment of the present invention.

[0017] Figure 2 This is one of the flow charts of the chain function scheduling method for serverless computing provided by an embodiment of the present invention.

[0018] Figure 3 This is the second flow chart of the chain function scheduling method for serverless computing provided by an embodiment of the present invention.

[0019] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0021] Figure 1 This is a schematic diagram of the structure of the chain function scheduling system for serverless computing provided by an embodiment of the present invention. Figure 1As shown, the chain function scheduling system for serverless computing includes: Function reflector, used to obtain the function routing table of multi-level function routers and build the full function routing table; The function orchestrator is used to generate function chain call requests based on the business logic of the target task and the full function routing table. The business logic is determined by the task request corresponding to the target task. A multi-level function router is used to call multiple target functions in sequence according to function chain call requests in a heterogeneous serverless computing environment; multiple target functions are used to execute business logic.

[0022] The function reflector is used to generate a full function routing table based on the function routing table periodically advertised by the function router, and the function orchestrator is used to subscribe to the full function routing table from the function reflector.

[0023] The function orchestrator is the initiator of the entire function call chain. It's typically deployed near the user or business side, coordinating and managing the calling sequence and dependencies between different functions. The function orchestrator is responsible for orchestrating and generating appropriate function call chains based on business logic and the list of available functions, and obtaining the results of the function chain calls after execution.

[0024] Optionally, the core functions of the function orchestrator include but are not limited to the following two aspects: 1) Workflow definition and management: Allows users to define complex workflow logic, including function sequence, branching logic, loops, etc., through a visual interface or code; 2) State management: Track the execution status of each function to ensure that tasks proceed as planned. Even if a function fails, it can be retried to ensure the continued execution of the task.

[0025] For example, in the field of data processing, a function orchestrator can combine multiple data processing functions in a certain order and logic to implement a series of operations such as data collection, cleaning, conversion, and storage.

[0026] Optionally, the function orchestrator is also used to obtain the full function routing table through the subscription interface provided by the function reflector. The first update is a full update, and subsequent updates are incremental push updates.

[0027] Optionally, the multi-level function router includes multiple function routers (such as a head-end function router, an intermediate function router, and an end-end function router), and different function routers are deployed in different serverless computing environments.

[0028] Among them, the function router is a device specifically used to coordinate and manage the communication and execution processes between multiple functions or services.

[0029] The working principle of the function router is as follows: 1) Message Receiving and Parsing: The function router first receives request messages from external or internal sources. These messages may contain key information such as the function name and parameters to be executed. The function router parses the received messages to understand the content and intent of the request.

[0030] 2) Routing Decision: Based on pre-defined rules, algorithms, or configuration information, the function router determines which specific function or service to forward the request to. These rules can be based on various factors, such as the function name, parameter types, source, and destination.

[0031] 3) Message forwarding: Once the target function or service is determined, the function router forwards the request message to the corresponding target. This typically involves network communication, such as sending the message to the server, virtual machine, or container where the target function resides via protocols such as HTTP / HTTPS or message queues.

[0032] 4) Execution Coordination: In some cases, the function router also needs to coordinate the execution order and dependencies of multiple functions or services. For example, the execution of some functions may need to wait for the completion results of other functions as input parameters.

[0033] Optionally, the multi-level function router is further configured to parse the function chain call request, obtain relevant information of the function chain to be called, and determine multiple target functions based on the relevant information of the function chain to be called.

[0034] Among them, a heterogeneous serverless computing environment refers to a serverless architecture environment that contains multiple different types of computing resources. The computing resources may be distributed on different cloud service providers, private cloud environments or edge computing nodes. These computing resources can be dynamically allocated and scheduled according to task requirements to improve computing efficiency and resource utilization.

[0035] Heterogeneous computing refers to the use of multiple different types of computing units (such as CPUs and GPUs) within a computing system to collaboratively complete computing tasks. These different types of computing units have unique architectures and advantages, making them suitable for handling different types of computing tasks. For example, CPUs excel at general-purpose computing and control logic, while GPUs excel at parallel computing and graphics processing. By combining different types of computing units, heterogeneous computing can fully leverage the characteristics of various computing resources, achieving more efficient computing performance and lower energy consumption.

[0036] Serverless computing is a cloud computing service model that allows developers to build and run applications without having to worry about server management and operations. In a serverless computing environment, the cloud service provider manages the server infrastructure, including configuration, maintenance, and scaling. Developers only need to focus on their business logic and code implementation. This computing model offers high efficiency and flexibility, reducing development costs and operational complexity.

[0037] In an embodiment of the present invention, by adopting a function reflector, the function routing tables of multi-level function routers are aggregated to construct a full function routing table, providing a rich and comprehensive data foundation for function chain calls; through the collaborative work of the function orchestrator and the multi-level function routers, the delay caused by multiple function calls and responses is reduced, thereby significantly improving the overall performance of chained function calls, improving the scalability and reliability of the system, and being suitable for complex serverless computing scenarios.

[0038] In some embodiments, the multi-level function router includes: The head-end function router is configured to call the first target function according to the function chain call request and forward the function chain call request to the intermediate function router; The intermediate function router is configured to receive the function chain call request sent by the head-end function router, call the second target function according to the function chain call request, and forward the function chain call request to the end function router; The end function router is used to receive the function chain call request sent by the intermediate function router, call the third target function according to the function chain call request, obtain the function chain call result, and send the function chain call result to the function orchestrator.

[0039] Optionally, the multiple objective functions include at least a first objective function, a second objective function and a third objective function.

[0040] Among them, the first target function is the available function in the function routing table of the head-end function router; the second target function is the available function in the function routing table of the intermediate function router; and the third target function is the available function in the function routing table of the end function router.

[0041] Among them, the head-end function router is the entry function router of the function call chain, which is usually deployed on the entry side of the serverless computing environment. Its main function is to parse the function chain call request, obtain the function list of this area, and be responsible for the function chain call of this area. After completing the function call of this area, it forwards the function chain call request to the downstream function router.

[0042] The intermediate function router is a routing node in the middle of a function call chain, typically deployed at the ingress of a serverless computing environment. It transmits function call requests within the call chain and dynamically selects the target function to call or routes to the next routing node based on the function call chain's configuration rules. The intermediate function router maintains a list of functions in its region and periodically reports the function routing table to the function reflector. The collaborative operation of multiple function routers enables the construction of complex call topologies to meet diverse functional requirements.

[0043] Among them, the terminal function router is the terminal routing node of the function call chain, which is mainly responsible for passing the final function chain call request to the target function and returning the execution result to the function orchestrator after the call is completed.

[0044] In some embodiments, the head-end function router is also used to send a function routing request to the intermediate function reflector and receive the address information of the intermediate function router sent by the function reflector; the intermediate function router is also used to send a function routing request to the function reflector and receive the address information of the end function router sent by the function reflector.

[0045] Optionally, the function reflector is also used to receive the function routing request sent by the head-end function router, analyze the function routing request, compare it with the current full function routing table, select the required intermediate function router from multiple intermediate function routers (such as selecting an intermediate function router with a lighter load), determine the address information of the selected intermediate function router, and send the address information of the intermediate function router to the head-end function router.

[0046] Optionally, the function reflector is also used to receive the function routing request sent by the intermediate function router, analyze the function routing request, compare it with the current full function routing table, select the required end function router from the multi-level function router, determine the address information of the selected end function router, and send the address information of the end function router to the intermediate function router.

[0047] It can be understood that the function reflector is responsible for connecting function routers in different regions, coordinating function routers in different serverless computing environments, optimizing the overall function routing path, reducing latency, and improving system efficiency.

[0048] In some embodiments, a function routing table entry includes a function name, a function access credential, and function metadata; a function chain call request includes a function call chain and a corresponding request access credential.

[0049] For example, a routing entry in a function routing table includes the function name F (Function), access credentials T (Token), and function metadata M (Metadata).

[0050] Optionally, the function reflector aggregates the route push (F, T, M) of each function router R (Router) to establish a full function routing table. The format of the full function routing table entry is (R, F, T, M).

[0051] It should be noted that the function router is deployed at the entrance of the serverless computing function environment provided by the cloud service provider. It is responsible for maintaining the function routing table in this area and regularly notifying the function routing table entries (F, T, M) for which it is responsible to the function reflector. To ensure security, T is a time-varying parameter valid within the notification window.

[0052] In some embodiments, multiple target functions are called according to a function chain call request, including: Determine the request access credentials of each target function corresponding to the multi-level function router according to the function chain call request; Determine the function access credentials of each target function corresponding to the multi-level function router according to the function routing table; The request access credential of each target function is compared with the function access credential. Based on the comparison result, the request access credential of each target function is verified. If the verification passes, each target function is called.

[0053] It can be understood that by comparing the request access credentials of each target function with the function access credentials, each target function will be called only when the verification is passed, which can effectively prevent unauthorized function access and ensure that only legitimate requests can access the corresponding function, thereby enhancing the security, reliability and stability of the system.

[0054] The following describes a chain function scheduling method for serverless computing provided by an embodiment of the present invention. The chain function scheduling method for serverless computing described below and the chain function scheduling system for serverless computing described above can be referenced to each other.

[0055] Figure 2 This is a flow chart of a chain function scheduling method for serverless computing provided by an embodiment of the present invention. Figure 2 As shown, a chain function scheduling method for serverless computing is provided, and the chain function scheduling method for serverless computing is applied to Figure 1 The system in FIG. 2 is used as an example to illustrate the method, which includes the following steps: step 210, step 220, and step 230. The method flow steps are only a possible implementation of the present invention.

[0056] Step 210: Obtain the function routing table of the multi-level function router and construct a full function routing table; Step 220: Generate a function chain call request based on the business logic of the target task and the full function routing table; the business logic is determined based on the task request corresponding to the target task; Step 230: In a heterogeneous serverless computing environment, multiple target functions are called in sequence according to the function chain call request; the multiple target functions are used to execute business logic.

[0057] Among them, the function routing table of the multi-level function router includes but is not limited to: function names, function access credentials, function metadata and other information that function routers at different levels are responsible for.

[0058] Optionally, a target task request is received from an external source, where the target task request is generated through user interface input, other system calls, or timed task triggering.

[0059] Optionally, the target task request includes relevant information of the target task, such as the task type, business logic, task requirements, etc.

[0060] Business logic refers to the series of operational steps and judgment conditions required to achieve task objectives. For example, a cross-cloud video processing application requires a series of steps, including video upload and format checking, video transcoding, video analysis and tag generation, and video storage and distribution.

[0061] Optionally, the business logic is decomposed into multiple specific operation steps, each step corresponding to the call of one or more target functions.

[0062] The function chain call request includes a function call list arranged in a specific order, and the function call list includes information such as the function name, input parameters, and expected output results of each function.

[0063] It's important to note that when generating a function chain call request, you need to consider the data transfer and dependencies between functions. This ensures that each function receives the required input data and that the function's execution result is correctly passed to the next function. For example, if the output of function A is an input parameter for function B, this relationship must be clearly stated in the request to ensure that function B can execute correctly after receiving the result of function A.

[0064] In an embodiment of the present invention, a full function routing table is constructed by aggregating the function routing tables of multi-level function routers. On this basis, a function chain call request is generated in combination with the business logic of the target task. In a heterogeneous serverless computing environment, multiple target functions are called in sequence according to the function chain call request, which effectively simplifies the function call process, improves the efficiency and reliability of function calls, has strong scalability, and is suitable for complex serverless computing scenarios.

[0065] Figure 3 The second flow chart of the chain function scheduling method for serverless computing provided by the embodiment of the present invention is as follows: Figure 3 As shown, a chain function scheduling method for serverless computing is provided, including the following steps: The function reflector obtains the function routing table of the multi-level function router and builds the full function routing table; The function orchestrator obtains the target task request, determines the business logic of the target task, generates a function chain call request based on the business logic of the target task and the full function routing table, and sends the function chain call request to the headend function router; The head-end function router receives the function chain call request sent by the function orchestrator, calls the first target function according to the function chain call request, and forwards the function chain call request to the intermediate function router; The intermediate function router receives the function chain call request sent by the head-end function router, calls the second target function according to the function chain call request, and forwards the function chain call request to the end function router; The end function router receives the function chain call request sent by the intermediate function router, calls the third target function according to the function chain call request, obtains the function chain call result, and sends the function chain call result to the function orchestrator.

[0066] Optionally, the function orchestrator receives an external target task request, generates a function chain call request based on the business logic, carries the function call chain (F1, F2, ...Fn) and function call credentials (T1, T2, ...Tn), searches the full function routing table, obtains the address of the head-end function router R1, and sends the function chain call request to the head-end function router.

[0067] Optionally, the function orchestrator receives the function chain call result sent by the terminal function router.

[0068] Among them, the head-end function router supports function chain nesting, that is, it can call multiple target functions in this area.

[0069] Optionally, after the head-end function router completes the function call task in the region, it forwards the request to a designated function router according to the type and routing policy of the function chain call request.

[0070] In some embodiments, before forwarding the function chain call request to the intermediate function router, the method further includes: Send a function routing request to the function reflector and receive the address information of the intermediate function router sent by the function reflector; Before forwarding the function chain call request to the terminal function router, it also includes: Send a function routing request to the function reflector and receive the address information of the end function router sent by the function reflector.

[0071] Optionally, the function reflector is further configured to receive a function routing request sent by the head-end function router, determine the address information of the selected intermediate function router, and send the address information of the intermediate function router to the head-end function router.

[0072] Optionally, the function reflector is further configured to receive a function routing request sent by the intermediate function router, determine the address information of the selected end function router, and send the address information of the end function router to the intermediate function router.

[0073] In some embodiments, the function routing table includes a function name, function access credentials, and function metadata; the function chain call request includes a function call chain and corresponding request access credentials.

[0074] Optionally, the full function routing table includes but is not limited to the function router address, function name, function access credentials, function metadata, etc.

[0075] Optionally, specific access credential rules can be pre-set for different functions based on specific business needs and security policies to achieve refined management of function calls.

[0076] In some embodiments, multiple target functions are called sequentially according to a function chain call request, including: Determine the request access credentials of each target function corresponding to the multi-level function router according to the function chain call request; Determine, according to the function routing table of the multi-level function router, the function access credentials of each target function corresponding to the multi-level function router; The request access credential of each target function is compared with the function access credential. Based on the comparison result, the request access credential of each target function is verified. If the verification passes, each target function is called.

[0077] It can be understood that by comparing and verifying the request access credentials of each target function with the function access credentials, and calling each target function when the verification is passed, refined management and control of function calls are achieved, the security and flexibility of function calls are improved, the probability of errors during function execution can be reduced, the stable operation of the system can be maintained, and the user experience and system availability can be improved.

[0078] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410 may be a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits for controlling the execution of the program. In different embodiments, the communications interface 420 includes a network interface, a high-level bus interface, etc. The memory 430 may be a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory, a solid-state drive, or other type of static or dynamic storage device. The processor 410, the communications interface 420, and the memory 430 communicate with each other via the communications bus 440. The processor 410 can call the logic instructions in the memory 430 to execute a chain function scheduling method for serverless computing, which includes: generating a function chain call request based on the business logic of the target task and the full function routing table; the business logic is determined based on the task request of the target task; in a heterogeneous serverless environment, multiple target functions are called in sequence according to the function chain call request; multiple target functions are used to execute the business logic.

[0079] Furthermore, the logic instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for causing a computer device (such as a personal computer, server, or network device) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0080] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the chain function scheduling method for serverless computing provided by the above methods, the method including: generating a function chain call request based on the business logic of the target task and the full function routing table; the business logic is determined based on the task request of the target task; in a heterogeneous serverless environment, multiple target functions are called in sequence according to the function chain call request; multiple target functions are used to execute the business logic.

[0081] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the chain function scheduling method for serverless computing provided by the above methods. The method includes: generating a function chain call request based on the business logic of the target task and the full function routing table; the business logic is determined based on the task request of the target task; in a heterogeneous serverless environment, multiple target functions are called in sequence according to the function chain call request; multiple target functions are used to execute the business logic.

[0082] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0083] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A chain function scheduling system for serverless computing, characterized in that: include: Function reflector, used to obtain the function routing table of multi-level function routers and build the full function routing table; A function orchestrator, configured to generate a function chain call request based on the business logic of the target task and the full function routing table; The business logic is determined according to the task request corresponding to the target task; A multi-level function router is used to call multiple target functions in sequence according to the function chain call request in a heterogeneous serverless computing environment; the multiple target functions are used to execute the business logic.

2. The system according to claim 1, wherein: The multi-level function router comprises: A head-end function router, configured to call the first target function according to the function chain call request, and forward the function chain call request to an intermediate function router; an intermediate function router, configured to receive the function chain call request sent by the head-end function router, call the second target function according to the function chain call request, and forward the function chain call request to the end function router; The end function router is used to receive the function chain call request sent by the intermediate function router, call the third target function according to the function chain call request, obtain the function chain call result, and send the function chain call result to the function orchestrator.

3. The system according to claim 2, characterized in that The head-end function router is also used to send a function routing request to the intermediate function reflector and receive the address information of the intermediate function router sent by the function reflector; the intermediate function router is also used to send a function routing request to the function reflector and receive the address information of the end function router sent by the function reflector.

4. The system according to claim 1, wherein: The function routing table includes a function name, a function access credential, and function metadata; the function chain call request includes a function call chain and a corresponding request access credential.

5. The system according to claim 4, characterized in that The calling of multiple target functions in sequence according to the function chain calling request includes: Determining, according to the function chain call request, request access credentials for each target function corresponding to the multi-level function router; Determining, according to the function routing table, function access credentials of each target function corresponding to the multi-level function router; The request access credential of each target function is compared with the function access credential, and the request access credential of each target function is verified according to the comparison result. If the verification passes, each target function is called.

6. A chain function scheduling method for serverless computing, characterized in that: Applied to the chained function scheduling system for serverless computing according to any one of claims 1 to 5, the method comprising: Obtain the function routing table of the multi-level function router and build the full function routing table; Generate a function chain call request according to the business logic of the target task and the full function routing table; the business logic is determined according to the task request corresponding to the target task; In a heterogeneous serverless computing environment, multiple target functions are called in sequence according to the function chain call request; the multiple target functions are used to execute the business logic.

7. The chained function scheduling method for serverless computing according to claim 6, characterized in that: The calling of multiple target functions in sequence according to the function chain calling request includes: The head-end function router calls the first target function according to the function chain call request, and forwards the function chain call request to the intermediate function router; The intermediate function router receives the function chain call request sent by the head-end function router, calls the second target function according to the function chain call request, and forwards the function chain call request to the end function router; The end function router receives the function chain call request sent by the intermediate function router, calls the third target function according to the function chain call request, obtains a function chain call result, and sends the function chain call result to the function orchestrator.

8. The chain function scheduling method for serverless computing according to claim 7, characterized in that: Before forwarding the function chain call request to the intermediate function router, the method further includes: Sending a function routing request to the function reflector, and receiving address information of the intermediate function router sent by the function reflector; Before forwarding the function chain call request to the terminal function router, the method further includes: Send a function routing request to the function reflector, and receive the address information of the terminal function router sent by the function reflector.

9. The chain function scheduling method for serverless computing according to claim 6, characterized in that: The function routing table includes a function name, a function access credential, and function metadata; the function chain call request includes a function call chain and a corresponding request access credential.

10. The chained function scheduling method for serverless computing according to claim 9, characterized in that: The calling of multiple target functions according to the function chain calling request includes: Determining, according to the function chain call request, request access credentials for each target function corresponding to the multi-level function router; Determining, according to the function routing table of the multi-level function router, the function access credentials of each target function corresponding to the multi-level function router; The request access credential of each target function is compared with the function access credential, and the request access credential of each target function is verified according to the comparison result. If the verification passes, each target function is called.

Images