Main station layer computing resource collaborative scheduling method and system

Abstract

This application provides a method and system for collaborative scheduling of computing resources at the master station layer. The method includes: constructing a resource allocation constraint model and a routing constraint allocation model based on the pending business needs of multiple business master stations, the network topology relationships between business master stations, and the resource availability of each business master station; transforming the resource allocation constraint model and the routing constraint allocation model into optimization models according to preset business constraints, and decomposing them into resource allocation sub-models and routing optimization sub-models through a decoupling strategy; generating a collaborative scheduling scheme for computing resources by solving the resource allocation sub-model and the routing optimization sub-model respectively; and executing corresponding data transmission and resource scheduling operations according to the collaborative scheduling scheme. In this way, network topology constraints and dynamic computing power limitations are processed in a layered manner, accelerating convergence and ensuring real-time scheduling in high-dimensional complex scenarios; dynamic topology and elastic resource allocation are supported, making it suitable for scenarios such as edge computing and smart grid master station collaboration.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more particularly to a method and system for collaborative scheduling of master-site computing resources applicable to distributed computing services. Background Technology

[0002] With the widespread adoption of distributed computing services, service latency has become a key indicator of user experience. Existing technologies have the following limitations:

[0003] Resource scheduling fragmentation: Traditional methods typically separate routing planning from computing power allocation optimization (e.g., selecting paths before allocating resources), ignoring their coupling and leading to higher-than-expected latency. Insufficient latency modeling: Total service latency needs to encompass both data transmission latency (dependent on multi-hop link bandwidth) and computation processing latency (dependent on CPU resources), but existing solutions lack a unified optimization model. Constraint complexity: Topological relationships between master stations (e.g., hop count limits) and dynamic computing resource availability (e.g., CPU cycle constraints) require joint modeling, and existing methods struggle to efficiently solve such high-dimensional constraint problems. Local optimization defects: Independent optimization by a single master station easily leads to resource contention, failing to guarantee the overall service quality (e.g., some services time out due to insufficient paths or computing power).

[0004] Therefore, there is an urgent need for a distributed scheduling method that integrates routing and computing power collaboration and is driven by end-to-end latency. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for collaborative scheduling of computing resources at the master station layer, so as to improve the efficiency of computing resource allocation.

[0006] To achieve the above objectives, the master station layer general computing resource collaborative scheduling method provided in this application specifically includes: constructing a resource allocation constraint model and a routing constraint allocation model based on the pending business needs of multiple business master stations, the network topology relationship between the business master stations, and the resource availability of each business master station; transforming the resource allocation constraint model and the routing constraint allocation model into optimization models according to preset business constraints, and decomposing them into resource allocation sub-models and routing optimization sub-models through a decoupling strategy; generating a general computing resource collaborative scheduling scheme by solving the resource allocation sub-model and the routing optimization sub-model respectively; and executing corresponding data transmission and resource scheduling operations according to the general computing resource collaborative scheduling scheme.

[0007] In the above-mentioned master station layer computing resource collaborative scheduling method, optionally, constructing a resource allocation constraint model and a routing constraint allocation model based on the pending service requirements of multiple service master stations, the network topology relationship between the service master stations, and the resource availability of each service master station includes: constructing a service scheduling queue and a routing constraint model based on the pending service requirements of multiple service master stations and the network topology relationship between the service master stations; and constructing a resource allocation constraint model based on the pending service requirements of multiple service master stations and the resource availability of each service master station.

[0008] In the above-mentioned master station layer computing resource collaborative scheduling method, optionally, constructing a service scheduling queue and routing constraint model based on the pending service requirements of multiple service master stations and the network topology relationship between the service master stations includes: constructing a service queue to be served based on the pending service requirements of each service master station; constructing a network topology relationship graph based on the service queue to be served using the network relationship between each service master station; and establishing a routing constraint model through the network topology relationship graph.

[0009] In the above-mentioned main station layer computing resource collaborative scheduling method, optionally, the resource allocation constraint model and the routing constraint allocation model are transformed into an optimization model according to preset business constraints, including: calculating the predicted total service latency of each service according to preset business constraints, and transforming the resource allocation constraint model and the routing constraint allocation model into an optimization model with the goal of maximizing cumulative revenue by predicting the total service latency.

[0010] In the above-mentioned master station layer computing resource collaborative scheduling method, optionally, calculating the predicted total service latency of each service according to preset service constraints includes: calculating the transmission latency of each service to be processed based on the data volume corresponding to the service to be processed in each service master station, calculating the processing cycle data of the service to be processed, and the historical average bandwidth between each service master station; and obtaining the predicted total service latency of each service based on the transmission latency.

[0011] In the above-mentioned main station layer computing resource collaborative scheduling method, the decoupling strategy may optionally include: dividing the optimization model into a resource allocation sub-model that fixes the routing path and solves for the optimal resource allocation, and a routing optimization sub-model that fixes the resource allocation scheme and iteratively optimizes the routing path.

[0012] In the above-mentioned main station layer general computing resource collaborative scheduling method, optionally, generating a general computing resource collaborative scheduling scheme by solving the resource allocation sub-model and the routing optimization sub-model respectively includes: generating an initial resource allocation scheme or an initial routing path based on the resource allocation sub-model or the routing optimization sub-model; obtaining an updated resource scheduling scheme and a path scheduling scheme by alternately iterating through the initial resource allocation scheme or the initial routing path using the resource allocation sub-model and the routing optimization sub-model; and generating a general computing resource collaborative scheduling scheme when the resource scheduling scheme and the path scheduling scheme meet a preset convergence condition.

[0013] This application also provides a master station layer computing resource collaborative scheduling system, the system including a constraint module, a decoupling module, and an analysis module; the constraint module is used to construct a resource allocation constraint model and a routing constraint allocation model based on the pending business requirements of multiple business master stations, the network topology relationship between the business master stations, and the resource availability of each business master station; the decoupling module is used to transform the resource allocation constraint model and the routing constraint allocation model into an optimization model according to preset business constraints, and decompose them into a resource allocation sub-model and a routing optimization sub-model through a decoupling strategy; the analysis module is used to generate a computing resource collaborative scheduling scheme by solving the resource allocation sub-model and the routing optimization sub-model respectively, and execute corresponding data transmission and resource scheduling operations according to the computing resource collaborative scheduling scheme.

[0014] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0015] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0016] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0017] The beneficial technical effects of this application are as follows: synchronously integrating data transmission and computation processing latency, directly driving scheduling decisions through a revenue quantification model to ensure that services meet latency requirements; adopting a hierarchical decoupling mechanism for routing and computing power, with dynamic allocation of computing power at the upper layer and iterative path updates at the lower layer, breaking resource fragmentation and improving utilization; hierarchically processing network topology constraints and dynamic computing power limitations, accelerating convergence and ensuring real-time scheduling in high-dimensional and complex scenarios; supporting dynamic topology and elastic resource allocation, suitable for scenarios such as edge computing and smart grid master station collaboration. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. In the drawings:

[0019] Figure 1 A flowchart illustrating a method for collaborative scheduling of main station layer computing resources provided in an embodiment of this application;

[0020] Figure 2 This is a schematic diagram provided for an embodiment of this application;

[0021] Figure 3 This is a schematic diagram provided for an embodiment of this application;

[0022] Figure 4 This is a schematic diagram provided for an embodiment of this application;

[0023] Figure 5 This is a schematic diagram provided for an embodiment of this application;

[0024] Figure 6 A schematic diagram of the main station layer computing resource collaborative scheduling system provided in an embodiment of this application;

[0025] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] The following will describe in detail the implementation methods of this application with reference to the accompanying drawings and embodiments, so as to fully understand how this application uses technical means to solve technical problems and achieve technical effects, and to implement it accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in each embodiment of this application can be combined with each other, and the resulting technical solutions are all within the protection scope of this application.

[0027] Furthermore, the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0028] Please refer to Figure 1 As shown, the main station layer computing resource collaborative scheduling method provided in this application specifically includes:

[0029] S101 constructs a resource allocation constraint model and a routing constraint allocation model based on the pending business requirements of multiple business master stations, the network topology relationship between the business master stations, and the resource availability of each business master station;

[0030] S102 transforms the resource allocation constraint model and the routing constraint allocation model into an optimization model according to the preset business constraints, and decomposes them into a resource allocation sub-model and a routing optimization sub-model through a decoupling strategy.

[0031] S103 generates a collaborative scheduling scheme for general computing resources by solving the resource allocation sub-model and the routing optimization sub-model respectively, and performs corresponding data transmission and resource scheduling operations according to the collaborative scheduling scheme for general computing resources.

[0032] Therefore, the master station layer computing resource collaborative scheduling method provided in this application adopts a routing and computing power layer decoupling mechanism, with the upper layer dynamically allocating computing power and the lower layer iteratively updating paths, breaking resource fragmentation and improving utilization; it handles network topology constraints and dynamic computing power limitations in a layered manner, accelerating convergence and ensuring real-time scheduling in high-dimensional and complex scenarios; it supports dynamic topology and elastic resource allocation, and is suitable for edge computing, smart grid master station collaboration and other scenarios; the implementation methods of each step will be described one by one in subsequent embodiments, and will not be detailed here.

[0033] Please refer to Figure 2 As shown, in some embodiments of this application, constructing a resource allocation constraint model and a routing constraint allocation model based on the pending service requirements of multiple service master stations, the network topology relationship between the service master stations, and the resource availability of each service master station includes:

[0034] S201 constructs a service scheduling queue and routing constraint model based on the pending service requirements of multiple service master stations and the network topology relationship between the service master stations;

[0035] S202 constructs a resource allocation constraint model based on the pending business needs of multiple business master stations and the resource availability of each business master station.

[0036] In this embodiment, the connection relationships and resource scheduling relationships between various business master stations are analyzed primarily based on the business requirements to be processed. Taking resource scheduling as an example, when it is determined that the business of master station A requires computing power resource 'a', and master stations A and its associated master stations B and C each have remaining computing power resources of b, c, and d respectively, the computing power difference can be determined by subtracting a from b. Then, the amount of computing power resources in c and d that can be used to make up for the computing power difference is calculated. This logic is then extrapolated to the entire network to construct the corresponding resource allocation constraint model. In this process, determining which master stations are associated with master station A is done through a routing constraint model. The routing constraint model can limit the number of hops to analyze the connection relationships between master stations. The specific construction method will be explained in detail in subsequent embodiments.

[0037] For further details, please refer to... Figure 3As shown, the business scheduling queue and routing constraint model constructed based on the pending business needs of multiple business master stations and the network topology relationships between the business master stations includes:

[0038] S301 constructs a queue of pending services based on the pending service needs of each business master station;

[0039] S302 constructs a network topology diagram based on the network relationships between the service master stations according to the queue of services to be served, and establishes a routing constraint model through the network topology diagram.

[0040] Specifically, based on the queue of services to be served, a network topology diagram can be constructed using the network relationships between each service master station. This can be done using data such as the network connectivity, communication distance, and location relationships of each service master station. For example, if there is no direct connection path between master station A and master station D, but master station A is connected to master station B, master station B is connected to master station C, and master station C is connected to master station D, then master station A and master station D have connection path relationships from A to B, B to C, and C to D. And so on. By summarizing all the connectable paths between all master stations, the corresponding network topology diagram can be obtained.

[0041] In practice, a queue of pending services for each main station can be established at the beginning of a work cycle. This queue of pending services can be represented by the following model:

[0042]

[0043] Where z is the business master station number, Z is the set of all master station numbers, |I z | The number of services pending on the main site z. The |I of the main site z z | Business.

[0044] The routing constraint model is expressed as follows:

[0045]

[0046] Among them, l i,j For decision variables, if main station i and main station j are directly connected, then l i,j =1, otherwise 0, n i,j N is the number of hops between master station i and master station j, where N is a constant that limits the number of hops.

[0047] Similarly, when constructing the resource allocation constraint model, the main analysis target is the available resources, and the computing power resources required by the business set corresponding to each main station are the constraints. The resource allocation constraint model includes the following:

[0048]

[0049] Among them, c i,j Let c be the decision variable. If master station j allocates computing resources to master station i, then c i,j =1, otherwise 0, c j,r Available computing resources for main station j, c z,j,k This refers to the computing resources allocated by main station j to the k-th service in main station z.

[0050] In one embodiment of this application, transforming the resource allocation constraint model into an optimization model based on preset business constraints includes: calculating the predicted total service latency for each service based on preset business constraints, and transforming the resource allocation constraint model into an optimization model with the goal of maximizing cumulative revenue through the predicted total service latency.

[0051] For reference Figure 4 As shown, in the above embodiments, calculating the predicted total service latency for each service based on preset service constraints includes:

[0052] S401 calculates the transmission delay of each pending service based on the data volume corresponding to the pending service in each service master station, the processing cycle data of the pending service, and the historical average bandwidth between each service master station.

[0053] S402 obtains the predicted total service latency for each service based on the transmission latency.

[0054] Specifically, before constructing the optimization model, it is necessary to determine the total service latency and the target benefit. The total service latency includes both service transmission latency and service computation latency. In some cases, such as when a service takes 2 hours to process at master station A but only 1 hour at master station B, transmitting the service from master station A to master station B in less than 1 hour is significantly more efficient. Therefore, in this embodiment, it is necessary to calculate the service transmission latency and service computation latency separately. The service transmission latency is:

[0055]

[0056] Among them, D z,k The amount of data corresponding to business k in the main site z, L i,j Given a path from master station i to master station j, this set contains the master station numbers of all master stations in the multi-hop link except for the target master station j. v,next For L i,j The historical average bandwidth of the relay master station v and its next hop involved in the process, and the service calculation latency are as follows:

[0057]

[0058] Where, δ z,kTo calculate the number of CPU cycles required to process k bits of data, the predicted total service latency is:

[0059] d i,j,k =d i,j,t,k +d i,j,c,k

[0060] The corresponding optimization objective benefit is:

[0061]

[0062] Where, τ i,j,k For latency requirements.

[0063] Therefore, an optimization model can be constructed based on the predicted total service latency and the target benefit. This optimization model can be referenced as follows:

[0064]

[0065] Among them, I z,ι This refers to the set of services that the main site z cannot serve due to limited computing resources.

[0066] After constructing the above optimization model, two sub-models can be decoupled. The specific decoupling strategy includes dividing the optimization model into a resource allocation sub-model with fixed routing paths and solving for optimal resource allocation, and a routing optimization sub-model with fixed resource allocation schemes and iteratively optimizing routing paths.

[0067] After obtaining the resource allocation sub-model and the routing optimization sub-model, the corresponding collaborative scheduling analysis can begin. Overall, the main implementation process of the above embodiment includes: clearly specifying the queue of services to be served by each master station at the beginning of the specified work cycle; drawing the network topology diagram of each master station and establishing a routing constraint model; establishing a resource allocation constraint model based on the availability of computing resources of each master station; calculating the expected total service latency of each service under given computing power and path planning, and defining the corresponding optimization target revenue accordingly; establishing an optimization model with the goal of maximizing cumulative revenue; decoupling the original optimization problem of the optimization model into two sub-problems and solving them separately; and executing the corresponding data transmission and computing power allocation scheme based on the converged solution of the optimization problem.

[0068] Finding the convergent solution mainly involves transforming the complex optimization model into two relatively simple sub-models: one with a fixed routing path and the other with fixed resources, to analyze the optimized resource allocation scheme and path optimization scheme. For details, please refer to... Figure 5 As shown, in one embodiment of this application, generating a collaborative scheduling scheme for general computing resources by solving the resource allocation sub-model and the routing optimization sub-model respectively includes:

[0069] S501 generates an initial resource allocation scheme or an initial routing path based on the resource allocation sub-model or the routing optimization sub-model;

[0070] S502 uses the initial resource allocation scheme or the initial routing path to perform alternating iterative solutions using the resource allocation sub-model and the routing optimization sub-model to obtain the updated resource scheduling scheme and path scheduling scheme;

[0071] S503 When the resource scheduling scheme and the path scheduling scheme meet the preset convergence conditions, a collaborative scheduling scheme for general computing resources is generated.

[0072] To facilitate a clearer understanding of the above embodiments, the following uses four main stations A, B, C, and D as an example. Main station A has an initial computing power of 40, B has an initial computing power of 30, C has an initial computing power of 20, and D has an initial computing power of 10. The initial routing path is A to B to D, and the backup routing path is A to C to D. In step S502, the path is initially fixed at A to B to D. During the resource allocation sub-model optimization process, it is determined that main station B has excess computing power and needs to be reallocated. Therefore, 15% of the computing power is transferred from B to main stations A, B, and C. At this time, A... The main station has a capacity of 45, B has 15, C has 25, and D has 15. Analyzing this computing power result in the routing optimization sub-model reveals that the initial path from A to B to D increases latency by 50ms, necessitating an optimized path from A to C to D. Next, this optimized path from A to C to D is re-introduced into the resource allocation sub-model, revealing an excess of computing power resources at C. Therefore, the aforementioned method is used to reallocate the computing power resources of C. This process continues until the optimal routing path and computing power resource allocation scheme are determined as the overall computing resource collaborative scheduling scheme.

[0073] This approach addresses resource load imbalance and response node capabilities through fixed-path computing power optimization. It eliminates transmission bottlenecks and enhances reliability by utilizing path optimization under fixed computing power. The optimization model approximates its optimal solution through an iterative, alternating solution process. A hierarchical decoupling mechanism between routing and computing power is employed, with the upper layer dynamically allocating computing power and the lower layer iteratively updating paths. This breaks down resource fragmentation and improves utilization. Layered processing of network topology constraints and dynamic computing power limitations accelerates convergence and ensures real-time scheduling in high-dimensional and complex scenarios. It supports dynamic topology and elastic resource allocation, making it suitable for edge computing, smart grid master station collaboration, and other scenarios.

[0074] Please refer to Figure 6As shown, this application also provides a master station layer computing resource collaborative scheduling system, the system including a constraint module, a decoupling module, and an analysis module; the constraint module is used to construct a resource allocation constraint model based on the pending business requirements of multiple business master stations, the network topology relationship between the business master stations, and the resource availability of each business master station; the decoupling module is used to transform the resource allocation constraint model into an optimization model according to preset business constraints, and decompose it into a resource allocation sub-model and a routing optimization sub-model through a decoupling strategy; the analysis module is used to generate a computing resource collaborative scheduling scheme by solving the resource allocation sub-model and the routing optimization sub-model respectively, and execute corresponding data transmission and resource scheduling operations according to the computing resource collaborative scheduling scheme.

[0075] Since the principle behind this system's problem-solving approach is similar to the collaborative scheduling method for general computing resources at the main station level, the implementation of this system can be found in the implementation of the collaborative scheduling method for general computing resources at the main station level. Further details will not be repeated.

[0076] The beneficial technical effects of this application are as follows: synchronously integrating data transmission and computation processing latency, directly driving scheduling decisions through a revenue quantification model to ensure that services meet latency requirements; adopting a hierarchical decoupling mechanism for routing and computing power, with dynamic allocation of computing power at the upper layer and iterative path updates at the lower layer, breaking resource fragmentation and improving utilization; hierarchically processing network topology constraints and dynamic computing power limitations, accelerating convergence and ensuring real-time scheduling in high-dimensional and complex scenarios; supporting dynamic topology and elastic resource allocation, suitable for scenarios such as edge computing and smart grid master station collaboration.

[0077] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0078] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0079] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0080] like Figure 7 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 7 All components shown; in addition, the electronic device 600 may also include Figure 7 For components not shown, please refer to existing technologies.

[0081] like Figure 7 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0082] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.

[0083] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0084] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.

[0085] The memory 140 may also include a data storage unit (data 143) for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit (driver 144) of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0086] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0087] Based on different communication technologies, multiple communication modules 110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling typical telecommunications functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 130 is coupled to a central processing unit 100, enabling on-device recording via the microphone 132 and on-device playback of stored audio via the speaker 131.

[0088] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for collaborative scheduling of computing resources at the master station layer, characterized in that, The method includes: Based on the pending business requirements of multiple business master stations, the network topology between the business master stations, and the resource availability of each business master station, a resource allocation constraint model and a routing constraint allocation model are constructed. Based on preset business constraints, the resource allocation constraint model and the routing constraint allocation model are transformed into optimization models, and then decomposed into resource allocation sub-models and routing optimization sub-models through a decoupling strategy. A collaborative scheduling scheme for computing resources is generated by solving the resource allocation sub-model and the routing optimization sub-model respectively, and corresponding data transmission and resource scheduling operations are performed according to the collaborative scheduling scheme for computing resources. Transforming the resource allocation constraint model and the routing constraint allocation model into an optimization model based on preset business constraints includes: The predicted total service latency of each service is calculated based on the preset business constraints. The resource allocation constraint model and the routing constraint allocation model are transformed into an optimization model with the goal of maximizing cumulative revenue by predicting the total service latency. The decoupling strategy includes: The optimization model is decomposed into a resource allocation sub-model that solves for the optimal resource allocation by fixing the routing path, and a routing optimization sub-model that iteratively optimizes the routing path by fixing the resource allocation scheme.

2. The method for collaborative scheduling of computing resources at the master station layer according to claim 1, characterized in that, Based on the pending business needs of multiple business master stations, the network topology relationships between the business master stations, and the resource availability of each business master station, a resource allocation constraint model and a routing constraint allocation model are constructed, including: A service scheduling queue and routing constraint model are constructed based on the pending service requirements of multiple service master stations and the network topology relationship between the service master stations. A resource allocation constraint model is constructed based on the pending business needs of multiple business master stations and the resource availability of each business master station.

3. The method for collaborative scheduling of computing resources at the main station layer according to claim 2, characterized in that, The service scheduling queue and routing constraint model are constructed based on the pending service requirements of multiple service master stations and the network topology relationships between the service master stations, including: Construct a queue of pending business requests based on the pending business requests of each main business station; Based on the queue of services to be served, a network topology graph is constructed using the network relationships between the service master stations, and a routing constraint model is established using the network topology graph.

4. The method for collaborative scheduling of computing resources at the master station layer according to claim 1, characterized in that, The predicted total service latency for each service is calculated based on preset business constraints, including: The transmission latency of each pending service is calculated based on the data volume of the pending services in each business master station, the processing cycle data of the pending services, and the historical average bandwidth between each business master station. The predicted total service latency for each service is obtained based on the transmission latency.

5. The method for collaborative scheduling of computing resources at the master station layer according to claim 1, characterized in that, The generation of a collaborative scheduling scheme for general computing resources by solving the resource allocation sub-model and the routing optimization sub-model respectively includes: Generate an initial resource allocation scheme or an initial routing path based on the resource allocation sub-model or the routing optimization sub-model; The updated resource scheduling scheme and path scheduling scheme are obtained by iteratively solving the resource allocation sub-model and the route optimization sub-model using the initial resource allocation scheme or the initial route path; When the resource scheduling scheme and the path scheduling scheme meet the preset convergence conditions, a collaborative scheduling scheme for general computing resources is generated.

6. A master-station layer computing resource collaborative scheduling system, characterized in that, The system includes a constraint module, a decoupling module, and an analysis module; The constraint module is used to construct a resource allocation constraint model and a routing constraint allocation model based on the pending business requirements of multiple business master stations, the network topology relationship between the business master stations, and the resource availability of each business master station. The decoupling module is used to transform the resource allocation constraint model and the routing constraint allocation model into an optimization model according to preset business constraints, and decompose them into a resource allocation sub-model and a routing optimization sub-model through a decoupling strategy. The process of transforming the resource allocation constraint model and the routing constraint allocation model into an optimization model based on preset business constraints includes: calculating the predicted total service latency for each service based on preset business constraints, and transforming the resource allocation constraint model and the routing constraint allocation model into an optimization model with the goal of maximizing cumulative revenue through the predicted total service latency; the decoupling strategy includes: decomposing the optimization model into a resource allocation sub-model that solves for the optimal resource allocation by fixing the routing path, and a routing optimization sub-model that iteratively optimizes the routing path by fixing the resource allocation scheme; The analysis module is used to generate a collaborative scheduling scheme for general computing resources by solving the resource allocation sub-model and the routing optimization sub-model respectively, and to execute corresponding data transmission and resource scheduling operations according to the collaborative scheduling scheme for general computing resources.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1 to 5.

Images