Service feature adaptive traffic scheduling method and system in multi-cloud environment

Abstract

The invention discloses a service feature adaptive traffic scheduling method and system in a multi-cloud environment, and the method comprises the steps: randomly transmitting a user traffic request to any cloud service in the multi-cloud environment; the deep learning model obtains monitoring data of each cloud service in the multi-cloud environment, and judges whether a current user traffic request needs to be forwarded or not based on a set traffic balancing rule according to the user traffic request; if the judgment result is yes, the current user flow request is forwarded to the target cloud service, and if the judgment result is no, the current user flow request is received and processed by the current cloud service. The core of the method is that by means of a deep learning model, an entrance gateway is arranged in each cloud service to serve as an executor of flow scheduling, and the flow scheduling proportion among multiple cloud services is dynamically adjusted by taking a flow balancing rule abstracted from a specific service target as a guide, so that the set flow balancing rule target is achieved.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and specifically provides a method and system for adaptive traffic scheduling based on business characteristics in a multi-cloud environment. Background Art

[0002] A multi-cloud environment refers to an environment where an enterprise or organization uses and manages cloud computing resources from multiple cloud service providers at the same time. A multi-cloud environment may involve different types of cloud services such as public cloud, private cloud and hybrid cloud to meet different business needs and improve flexibility and reliability.

[0003] Traffic scheduling refers to the process of transmitting HTTP packets or requests from source nodes to target nodes according to specific strategies and algorithms. In a multi-cloud environment, traffic scheduling usually refers to dynamically adjusting the data flow based on factors such as load conditions, network latency, and cost considerations to achieve goals such as load balancing, performance optimization, or cost minimization.

[0004] The common traffic scheduling mechanisms in current multi-cloud environments are mostly based on manual experience and rule settings. The following describes in detail the existing technical solutions and the corresponding problems:

[0005] The traffic distribution ratio between multiple clouds is set based on manual experience and rules. For example, for two clouds A and B, the corresponding ratio is set according to the number of servers in each cloud. For example, if the servers are the same, the traffic ratio is set to 1:1.

[0006] The above traffic scheduling scheme is difficult to adapt to the changing cloud load conditions, and requires manual intervention when the corresponding situation changes. When the load of a certain cloud becomes high, manual intervention is required to adjust the traffic ratio. There is a lag effect, resulting in poor system performance and efficiency. It relies on human subjectivity and is difficult to ensure system stability.

[0007] On the other hand, the above traffic scheduling solution is also difficult to meet the changing business needs. For example, when the business goal is to make the order quantity between multiple clouds close to the same so as to make full use of the order server in each cloud to improve stability and efficiency, the traditional rule setting cannot achieve this goal.

[0008] In view of this, the present invention patent is proposed. Summary of the invention

[0009] The present invention proposes a business feature adaptive traffic scheduling method and system in a multi-cloud environment to solve the efficiency and stability problems caused by the traffic distribution ratio in a multi-cloud environment can only be based on artificial historical experience or simple rule settings.

[0010] Specifically, the following technical solutions are adopted:

[0011] In a first aspect, the present invention provides a method for adaptive traffic scheduling based on service characteristics in a multi-cloud environment, comprising:

[0012] Randomly send user traffic requests to any cloud service in a multi-cloud environment;

[0013] The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request;

[0014] If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

[0015] As an optional implementation of the present invention, in a method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment of the present invention, the deep learning model obtains monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request, including:

[0016] Abstract specific features based on business objectives and set traffic balancing rules corresponding to the specific features;

[0017] The abstracted specific features are converted into monitoring indicators. Each cloud service in the multi-cloud environment exposes the monitoring data corresponding to the monitoring indicators in the ingress gateway interface.

[0018] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in each cloud service in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

[0019] As an optional implementation manner of the present invention, a service feature adaptive traffic scheduling method in a multi-cloud environment of the present invention includes:

[0020] The deep learning model includes multiple sub-deep learning models, each corresponding to different traffic balancing rules;

[0021] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, selects a sub-deep learning model corresponding to the traffic balancing rules, and determines whether the current user traffic request needs to be forwarded.

[0022] As an optional implementation manner of the present invention, in a service feature adaptive traffic scheduling method under a multi-cloud environment of the present invention, the training process of the deep learning model includes:

[0023] Obtain historical traffic balancing data based on different traffic balancing rules in a multi-cloud environment;

[0024] The historical traffic balancing data under different traffic balancing rules are marked as positive and negative samples, and are respectively input into the sub-deep learning benchmark models corresponding to different traffic balancing rules for learning and training, so as to obtain the sub-deep learning models corresponding to different traffic balancing rules;

[0025] The historical traffic balancing data based on different traffic balancing rules in the multi-cloud environment is traffic balancing data obtained by manually processing traffic requests based on different traffic balancing rules.

[0026] As an optional implementation manner of the present invention, in a service feature adaptive traffic scheduling method under a multi-cloud environment of the present invention, forwarding the current user traffic request to the target cloud service includes:

[0027] Add a forwarding identifier to the current user traffic request and forward it to the target cloud service;

[0028] The target cloud service receives the forwarded user traffic request and parses it to find out that it contains a forwarding identifier. It then directly receives and processes the user traffic request without using a deep learning model to determine whether to forward the request.

[0029] As an optional implementation manner of the present invention, a service feature adaptive traffic scheduling method in a multi-cloud environment of the present invention includes:

[0030] Recording and storing the traffic scheduling data of the user traffic request in a monitoring indicator database;

[0031] The traffic scheduling data in the monitoring indicator database is regularly fed back to the deep learning model to perform optimization training of the deep learning model.

[0032] In a second aspect, the present invention provides a service feature adaptive traffic scheduling system in a multi-cloud environment, comprising:

[0033] Multiple cloud services deploy business services separately and randomly send user traffic requests to any cloud service in the multi-cloud environment;

[0034] The deep learning model module deploys a deep learning model that communicates with each cloud service respectively. The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the user traffic request and the set traffic balancing rules. If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

[0035] As an optional implementation mode of the present invention, in a service feature adaptive traffic scheduling system under a multi-cloud environment of the present invention, each cloud service independently deploys its own ingress gateway;

[0036] The cloud service's ingress gateway receives the user's traffic request and synchronously forwards it to the deep learning model;

[0037] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

[0038] As an optional embodiment of the present invention, a service characteristic adaptive traffic scheduling system in a multi-cloud environment of the present invention includes a monitoring indicator database module, and deploys a monitoring indicator database that communicates with each cloud service respectively:

[0039] The monitoring indicator database obtains and records the monitoring data exposed by the business services in each cloud service, and the deep learning model obtains the monitoring data of the business services in each cloud service in the multi-cloud environment based on the monitoring indicators to determine whether forwarding is required for the current user traffic request;

[0040] and, the monitoring indicator database acquires and records the traffic scheduling data of the user traffic request;

[0041] The monitoring indicator database regularly feeds back the traffic scheduling data to the deep learning model to perform optimization training of the deep learning model.

[0042] In a third aspect, the present invention provides a computer-readable recording medium storing a computer-executable program, which, when executed, implements a method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The core of the present invention is to set up an entry gateway in each cloud service with the help of a deep learning model. The entry gateway acts as the executor of traffic scheduling and dynamically adjusts the traffic scheduling ratio between multi-cloud services based on the traffic balancing rules abstracted from specific business objectives to achieve the set traffic balancing rule objectives.

[0045] Therefore, the service feature adaptive traffic scheduling method and system in a multi-cloud environment of the present invention has the following advantages:

[0046] 1. Since the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment of the present invention adopts a deep learning model, it can more accurately predict and identify the relationship between different business characteristics and cloud load conditions through a large amount of data training and learning, thereby improving the intelligence level of the system for dynamic diversion in a multi-cloud environment.

[0047] 2. Through real-time monitoring data and analysis, the system can quickly compare with the traffic balancing rule targets and automatically adjust the diversion strategy, making the system more responsive and adaptable in real time, improving the flexibility and stability of the system.

[0048] 3. By reducing human intervention to improve the accuracy of decision-making, the system can reduce operation and maintenance costs, improve management efficiency, realize intelligent traffic distribution and management, and promote the optimization and improvement of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flow chart of a method for adaptive traffic scheduling based on service characteristics in a multi-cloud environment according to a first embodiment of the present invention;

[0050] Figure 2 A system framework diagram of a traffic scheduling system for adaptively adjusting traffic based on service characteristics in a multi-cloud environment according to a first embodiment of the present invention;

[0051] Figure 3 A schematic structural diagram of an electronic device according to a second embodiment of the present invention;

[0052] Figure 4 A schematic diagram of a computer-readable recording medium according to a second embodiment of the present invention. DETAILED DESCRIPTION

[0053] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be described clearly and completely in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0054] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the invention claimed for protection, but merely represents some embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions in the embodiments may be combined with each other.

[0056] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0057] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the invention product is usually placed when in use, or the orientation or positional relationship commonly understood by those skilled in the art. Such terms are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0058] Embodiment 1

[0059] See also Figure 1 As shown, a method for adaptive traffic scheduling based on service characteristics in a multi-cloud environment in this embodiment includes:

[0060] Randomly send user traffic requests to any cloud service in a multi-cloud environment;

[0061] The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request;

[0062] If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

[0063] The core of the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment of this embodiment is to set an entry gateway in each cloud service with the help of a deep learning model. The entry gateway acts as the executor of traffic scheduling and dynamically adjusts the traffic scheduling ratio between multi-cloud services based on the traffic balancing rules abstracted from specific business objectives to achieve the set traffic balancing rule objectives.

[0064] Therefore, the service feature adaptive traffic scheduling method in a multi-cloud environment of this embodiment has the following advantages:

[0065] 1. Since the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment of this embodiment adopts a deep learning model, it can more accurately predict and identify the relationship between different business characteristics and cloud load conditions through a large amount of data training and learning, thereby improving the intelligence level of the system for dynamic diversion in a multi-cloud environment.

[0066] 2. Through real-time monitoring data and analysis, the system can quickly compare with the traffic balancing rule targets and automatically adjust the diversion strategy, making the system more responsive and adaptable in real time, improving the flexibility and stability of the system.

[0067] 3. By reducing human intervention to improve the accuracy of decision-making, the system can reduce operation and maintenance costs, improve management efficiency, realize intelligent traffic distribution and management, and promote the optimization and improvement of the entire system.

[0068] Furthermore, in a method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment of this embodiment, the deep learning model obtains monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request, including:

[0069] Abstract specific features based on business objectives and set traffic balancing rules corresponding to the specific features;

[0070] The abstracted specific features are converted into monitoring indicators. Each cloud service in the multi-cloud environment exposes the monitoring data corresponding to the monitoring indicators in the ingress gateway interface.

[0071] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in each cloud service in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

[0072] This embodiment abstracts specific features according to business objectives, and sets corresponding traffic balancing rules for specific features. When processing specific user traffic requests, traffic scheduling can be performed according to the specific feature data in the user traffic request. Specifically, this step requires manual intervention. For example, order volume balancing can be mapped to request volume balancing of the order creation interface, and overall traffic balancing can be mapped to the sum of traffic balancing of all interfaces, etc. Different businesses can set target rules separately.

[0073] In this embodiment, the abstracted specific features are converted into monitoring indicators. Each cloud service in the multi-cloud environment exposes the monitoring data corresponding to the monitoring indicators in the inlet gateway interface. In this way, the deep learning model can obtain the monitoring data in real time, and obtain the resource utilization status in the current multi-cloud environment based on the monitoring data, providing a basis for processing new user traffic request scheduling.

[0074] Specifically, the monitoring data includes the amount of user traffic requests scheduled for each cloud service, resource utilization of each cloud service, processing of user traffic requests by each cloud service, etc.

[0075] As an optional implementation manner of this embodiment, a service feature adaptive traffic scheduling method in a multi-cloud environment of this embodiment includes:

[0076] The deep learning model includes multiple sub-deep learning models, each corresponding to different traffic balancing rules;

[0077] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, selects a sub-deep learning model corresponding to the traffic balancing rules, and determines whether the current user traffic request needs to be forwarded.

[0078] Since different business objectives abstract different specific features, the traffic balancing rules corresponding to the specific features are also different. Therefore, in order to improve the compatibility of the deep learning model of this embodiment and meet the traffic balancing problem in a multi-cloud environment with different business objectives, the deep learning model of this embodiment includes multiple sub-deep learning models, each corresponding to different traffic balancing rules. In this way, as long as the business objectives are determined and the corresponding traffic balancing rules are set, the deep learning model selects the sub-deep learning model corresponding to the traffic balancing rule to handle the balanced distribution of user traffic requests.

[0079] Specifically, in a method for adaptive traffic scheduling based on service characteristics in a multi-cloud environment of this embodiment, the training process of the deep learning model includes:

[0080] Obtain historical traffic balancing data based on different traffic balancing rules in a multi-cloud environment;

[0081] The historical traffic balancing data under different traffic balancing rules are marked as positive and negative samples, and are respectively input into the sub-deep learning benchmark models corresponding to different traffic balancing rules for learning and training, so as to obtain the sub-deep learning models corresponding to different traffic balancing rules;

[0082] The historical traffic balancing data based on different traffic balancing rules in the multi-cloud environment is traffic balancing data obtained by manually processing traffic requests based on different traffic balancing rules.

[0083] Deep learning is a branch of machine learning. It is based on artificial neural network models and uses multi-level neural network structures to learn and implement complex pattern recognition and data analysis tasks. In a multi-cloud environment, deep learning technology can be applied to data analysis, load forecasting, behavioral pattern recognition, etc., providing intelligent decision support for traffic scheduling systems.

[0084] As an optional implementation of this embodiment, in a service feature adaptive traffic scheduling method under a multi-cloud environment of this embodiment, forwarding the current user traffic request to the target cloud service includes:

[0085] Add a forwarding identifier to the current user traffic request and forward it to the target cloud service;

[0086] The target cloud service receives the forwarded user traffic request and parses it to find out that it contains a forwarding identifier. It no longer uses the deep learning model to determine whether to forward the request, but directly receives and processes the user traffic request, thereby avoiding an infinite loop and preventing it from being executed.

[0087] As an optional implementation manner of this embodiment, a service feature adaptive traffic scheduling method in a multi-cloud environment of this embodiment includes:

[0088] Recording and storing the traffic scheduling data of the user traffic request in a monitoring indicator database;

[0089] The traffic scheduling data in the monitoring indicator database is regularly fed back to the deep learning model to perform optimization training of the deep learning model.

[0090] This embodiment optimizes the deep learning model by using traffic scheduling data to ensure the accuracy and reliability of the deep learning model in determining whether to forward traffic requests.

[0091] As an optional implementation mode of this embodiment, a method for adaptive traffic scheduling based on service characteristics in a multi-cloud environment of this embodiment, the specific implementation process includes:

[0092] 1. Abstract specific features based on business goals. This step requires manual intervention. For example, order volume balance can be mapped to request volume balance of the order creation interface, and overall traffic balance can be mapped to the sum of traffic balance of all interfaces, etc. Different businesses can set target rules separately.

[0093] 2. Convert the specific features abstracted in step 1 into monitoring indicators, that is, expose the corresponding data on the http interface and collect data regularly through a central monitoring collector. The data itself carries the different identification of each cloud.

[0094] 3. All clouds only need to deploy one deep learning inference program. Each cloud independently deploys a gateway program to receive http requests. The model reads monitoring data. When an http request is received, it is compared with the rules set in the first step, and decides whether it needs to be forwarded to other clouds based on the results. If necessary, a forwarding flag is attached to the http request.

[0095] 4. If the http request received by the target cloud has a forwarding flag, it will no longer be forwarded to avoid an infinite loop.

[0096] 5. Regularly deliver traffic scheduling data as feedback data to the deep learning model to optimize the model effect.

[0097] See also Figure 2 As shown, this embodiment also provides a service feature adaptive traffic scheduling system in a multi-cloud environment, including:

[0098] Multiple cloud services deploy business services separately and randomly send user traffic requests to any cloud service in the multi-cloud environment;

[0099] The deep learning model module deploys a deep learning model that communicates with each cloud service respectively. The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the user traffic request and the set traffic balancing rules. If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

[0100] The core of the business feature adaptive traffic scheduling system in a multi-cloud environment of this embodiment is to set an entry gateway in each cloud service with the help of the deep learning model of the deep learning model module. This entry gateway serves as the executor of traffic scheduling and dynamically adjusts the traffic scheduling ratio between multi-cloud services based on the traffic balancing rules abstracted from specific business goals to achieve the set traffic balancing rule goals.

[0101] Therefore, the service feature adaptive traffic scheduling system in a multi-cloud environment of this embodiment has the following advantages:

[0102] 1. Since the service feature adaptive traffic scheduling system in a multi-cloud environment of this embodiment adopts a deep learning model, it can more accurately predict and identify the relationship between different service features and cloud load conditions through a large amount of data training and learning, thereby improving the system's intelligence level for dynamic diversion in a multi-cloud environment.

[0103] 2. Through real-time monitoring data and analysis, the system can quickly compare with the traffic balancing rule targets and automatically adjust the diversion strategy, making the system more responsive and adaptable in real time, improving the flexibility and stability of the system.

[0104] 3. By reducing human intervention to improve the accuracy of decision-making, the system can reduce operation and maintenance costs, improve management efficiency, realize intelligent traffic distribution and management, and promote the optimization and improvement of the entire system.

[0105] Furthermore, in a service feature adaptive traffic scheduling system in a multi-cloud environment of this embodiment, each cloud service independently deploys its own ingress gateway;

[0106] The cloud service's ingress gateway receives the user's traffic request and synchronously forwards it to the deep learning model;

[0107] The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

[0108] In this embodiment, a business feature adaptive traffic scheduling system in a multi-cloud environment sets an entry gateway in each cloud service. This entry gateway serves as the executor of traffic scheduling and dynamically adjusts the traffic scheduling ratio between multi-cloud services based on the traffic balancing rules abstracted from specific business objectives to achieve the set traffic balancing rule objectives.

[0109] Furthermore, a traffic scheduling system for adaptively adjusting traffic characteristics in a multi-cloud environment according to the present embodiment includes a monitoring indicator database module, and a monitoring indicator database is deployed to communicate with each cloud service respectively:

[0110] The monitoring indicator database obtains and records the monitoring data exposed by the business services in each cloud service, and the deep learning model obtains the monitoring data of the business services in each cloud service in the multi-cloud environment based on the monitoring indicators to determine whether forwarding is required for the current user traffic request;

[0111] and, the monitoring indicator database acquires and records the traffic scheduling data of the user traffic request;

[0112] The monitoring indicator database regularly feeds back the traffic scheduling data to the deep learning model to perform optimization training of the deep learning model.

[0113] Embodiment 2

[0114] The following describes an electronic device embodiment of the present invention, which can be regarded as a specific physical implementation of the method and device embodiments of the present invention. The details described in the electronic device embodiment of the present invention should be regarded as a supplement to the above method or device embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above method or device embodiments.

[0115] Figure 3It is a structural diagram of an electronic device of an embodiment of the present invention, the electronic device includes a processor and a memory, the memory is used to store a computer executable program, when the computer program is executed by the processor, the processor executes a service feature adaptive traffic scheduling method in a multi-cloud environment of embodiment one.

[0116] like Figure 3 As shown, the electronic device is presented in the form of a general computing device. The processor may be one or more and work in coordination. The present invention does not exclude distributed processing, that is, the processor may be dispersed in different physical devices. The electronic device of the present invention is not limited to a single entity, but may also be the sum of multiple physical devices.

[0117] The memory stores a computer executable program, which is usually a machine-readable code. The computer-readable program can be executed by the processor to enable the electronic device to perform the method of the present invention, or at least part of the steps in the method.

[0118] The memory includes a volatile memory, such as a random access memory unit (RAM) and / or a cache memory unit, and may also be a non-volatile memory, such as a read-only memory unit (ROM).

[0119] Optionally, in this embodiment, the electronic device further includes an I / O interface, which is used for the electronic device to exchange data with an external device. The I / O interface can represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of a variety of bus structures.

[0120] It should be understood that Figure 3 The electronic device shown is only an example of the present invention, and the electronic device of the present invention may also include elements or components not shown in the above examples. For example, some electronic devices also include display units such as display screens, and some electronic devices also include human-computer interaction elements such as buttons, keyboards, etc. As long as the electronic device can execute the computer-readable program in the memory to implement the method of the present invention or at least part of the steps of the method, it can be considered as an electronic device covered by the present invention.

[0121] Figure 4 Schematic diagram of a computer readable recording medium according to an embodiment of the present invention. Figure 4As shown, a computer-readable recording medium stores a computer executable program, and when the computer executable program is executed, a method for adaptive traffic scheduling of service characteristics in a multi-cloud environment according to the first embodiment of the present invention is implemented. The computer-readable recording medium may include a data signal propagated in a baseband or as part of a carrier, which carries a readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable recording medium may also be any readable medium other than a readable recording medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the readable recording medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0122] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0123] Through the above description of the implementation mode, it is easy for those skilled in the art to understand that the present invention can be implemented by hardware capable of executing a specific computer program, such as the system of the present invention, and the electronic processing unit, server, client, mobile phone, control unit, processor, etc. contained in the system. The present invention can also be implemented by computer software that executes the method of the present invention, such as control software executed by a microprocessor, an electronic control unit, a client, a server, etc. However, it should be noted that the computer software that executes the method of the present invention is not limited to being executed by one or a specific hardware entity, and it can also be implemented in a distributed manner by unspecified specific hardware. For computer software, the software product can be stored in a computer-readable recording medium (which can be a CD-ROM, a USB flash drive, a mobile disk, etc.), and can also be distributed and stored on the network, as long as it enables the electronic device to execute the method according to the present invention.

[0124] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the above specific implementation methods. Therefore, any modification or equivalent replacement of the present invention; and all technical solutions and improvements thereof that do not depart from the spirit and scope of the invention are included in the scope of the claims of the present invention.

Claims

1. A traffic scheduling method for business characteristics adaptive to traffic in a multi-cloud environment, characterized in that: include: Randomly send user traffic requests to any cloud service in a multi-cloud environment; The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request; If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

2. According to the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment according to claim 1, it is characterized in that: The deep learning model obtains monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the set traffic balancing rules according to the user traffic request, including: Abstract specific features based on business objectives and set traffic balancing rules corresponding to the specific features; The abstracted specific features are converted into monitoring indicators. Each cloud service in the multi-cloud environment exposes the monitoring data corresponding to the monitoring indicators in the ingress gateway interface. The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in each cloud service in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

3. According to the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment according to claim 2, it is characterized in that: include: The deep learning model includes multiple sub-deep learning models, each corresponding to different traffic balancing rules; The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, selects a sub-deep learning model corresponding to the traffic balancing rules, and determines whether the current user traffic request needs to be forwarded.

4. According to the method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment according to claim 2, it is characterized in that: The training process of the deep learning model includes: Obtain historical traffic balancing data based on different traffic balancing rules in a multi-cloud environment; The historical traffic balancing data under different traffic balancing rules are marked as positive and negative samples, and are respectively input into the sub-deep learning benchmark models corresponding to different traffic balancing rules for learning and training, so as to obtain the sub-deep learning models corresponding to different traffic balancing rules; The historical traffic balancing data based on different traffic balancing rules in the multi-cloud environment is traffic balancing data obtained by manually processing traffic requests based on different traffic balancing rules.

5. The method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment according to claim 1 is characterized in that: The forwarding of the current user traffic request to the target cloud service includes: Add a forwarding identifier to the current user traffic request and forward it to the target cloud service; The target cloud service receives the forwarded user traffic request and parses it to find out that it contains a forwarding identifier. It then directly receives and processes the user traffic request without using a deep learning model to determine whether to forward the request.

6. The method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment according to claim 1, characterized in that: include: Recording and storing the traffic scheduling data of the user traffic request in a monitoring indicator database; The traffic scheduling data in the monitoring indicator database is regularly fed back to the deep learning model to perform optimization training of the deep learning model.

7. A traffic scheduling system that is adaptive to business characteristics in a multi-cloud environment, characterized in that: include: Multiple cloud services deploy business services separately and randomly send user traffic requests to any cloud service in the multi-cloud environment; The deep learning model module deploys a deep learning model that communicates with each cloud service respectively. The deep learning model obtains the monitoring data of each cloud service in the multi-cloud environment, and determines whether the current user traffic request needs to be forwarded based on the user traffic request and the set traffic balancing rules. If the judgment result is yes, the current user traffic request is forwarded to the target cloud service. If the judgment result is no, the current user traffic request is received and processed by the current cloud service.

8. The service feature adaptive traffic scheduling system in a multi-cloud environment according to claim 7, characterized in that: Each cloud service independently deploys its own entry gateway; The cloud service's ingress gateway receives the user's traffic request and synchronously forwards it to the deep learning model; The deep learning model obtains monitoring indicators according to user traffic requests, obtains traffic balancing rules based on specific features corresponding to the monitoring indicators, and obtains monitoring data of business services in a multi-cloud environment based on the monitoring indicators. The monitoring data has identification information of the corresponding cloud service, and determines whether forwarding is required for the current user traffic request.

9. The service feature adaptive traffic scheduling system in a multi-cloud environment according to claim 7, characterized in that: Including the monitoring indicator database module, deploying the monitoring indicator database that communicates with each cloud service separately: The monitoring indicator database obtains and records the monitoring data exposed by the business services in each cloud service, and the deep learning model obtains the monitoring data of the business services in each cloud service in the multi-cloud environment based on the monitoring indicators to determine whether forwarding is required for the current user traffic request; and, the monitoring indicator database acquires and records the traffic scheduling data of the user traffic request; The monitoring indicator database regularly feeds back the traffic scheduling data to the deep learning model to perform optimization training of the deep learning model.

10. A computer-readable recording medium storing a computer-executable program, characterized in that: When the computer executable program is executed, a method for adaptive traffic scheduling based on business characteristics in a multi-cloud environment as described in any one of claims 1 to 6 is implemented.

Images