Application orchestration scheduling method and device based on computing power network, equipment and medium

Abstract

The embodiment of the application relates to the technical field of computers, and discloses an application arrangement and scheduling method and device based on a computing power network, equipment and a computer readable storage medium, the method comprising: a primary application scheduler sends a query request to a computing power network arrangement layer according to a computing power request, obtains computing power arrangement information and network routing information sent by the computing power network arrangement layer, and determines a target data center corresponding to the computing power request; an application controller of the target data center receives a first scheduling instruction sent by the primary application scheduler, pulls an application image from an image warehouse of the target data center, and the application image is synchronized from an image warehouse of an original data center to the image warehouse of the target data center; and the application controller of the target data center creates an application instance according to the application image. The embodiment of the application solves the problem of ubiquitous computing power empowerment business application in the computing power network in the prior art, and the application instance of the business application can be smoothly scheduled between different data centers.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, specifically to an application orchestration and scheduling method, apparatus, device, and computer-readable storage medium based on computing power networks. Background Technology

[0002] Computing power networks, centered on computing power and based on the network, enable users to enjoy computing power connectivity anytime, anywhere. By innovatively constructing ubiquitous and integrated computing power networks, they promote the convergence and development of computing power in physical, logical, and heterogeneous spaces, making computing power a social-level service like water and electricity—"one-point access, instant access"—achieving "ubiquitous network reach, ubiquitous computing power, and pervasive intelligence." Current research on computing power networks mainly focuses on exploring computing power, computing power awareness, computing power routing, and computing power management, relying on a unified computing power system and capability template to support a new computing power routing measurement mechanism of "optimal path + best computing power."

[0003] However, research on how to achieve smooth scheduling of applications based on computing power and network routing for computing power networks, so that applications can run on ubiquitous computing power, is still relatively lacking. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention provide an application orchestration and scheduling method, apparatus, device and computer-readable storage medium based on computing power network, which is used to solve the problem of how ubiquitous computing power in computing power network can empower business applications in the prior art, and the application instances of business applications can be smoothly scheduled between different data centers.

[0005] According to one aspect of the present invention, an application orchestration and scheduling method based on a computing power network is provided, the method comprising:

[0006] The first-level application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request;

[0007] The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, and synchronizes the application image from the image repository of the original data center to the image repository of the target data center.

[0008] The application controller in the target data center creates application instances based on the application image.

[0009] In an alternative approach, before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0010] The primary application scheduler determines the type of application requesting computing power.

[0011] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0012] The primary data manager receives the secondary scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0013] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0014] In one alternative approach, after the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0015] The original data center's application controller receives a third scheduling instruction from the primary application scheduler and destroys the original application instance corresponding to the computing power request.

[0016] In an alternative approach, after the application controller in the target data center creates an application instance based on the application image, the method further includes:

[0017] The application controller in the target data center controls the application implementation to consume the corresponding business data and obtain the processing result.

[0018] The application controller of the target data center sends the processing result to the primary application scheduler.

[0019] The primary application scheduler receives the processing results and aggregates them.

[0020] In an alternative approach, before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0021] Each data center's image repository synchronizes the data center's application images according to the scheduling situation and synchronizes the application images to the primary image repository;

[0022] The primary image repository backs up synchronized application images.

[0023] In an alternative approach, before the primary application scheduler sends a query request to the network orchestration layer based on the computing power request, the method further includes:

[0024] The computing network perception layer acquires real-time perception information from each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0025] After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the method further includes:

[0026] The computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and the network routing information to the first-level application scheduler.

[0027] According to another aspect of the present invention, an application orchestration and scheduling apparatus based on a computing power network is provided, comprising:

[0028] The first-level application scheduler is used to send query requests to the computing network orchestration layer based on computing power requests, obtain computing power orchestration information and network routing information sent by the computing network orchestration layer, and determine the target data center corresponding to the computing power request.

[0029] The application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, synchronizes the application image from the image repository of the original data center to the image repository of the target data center, and creates application instances based on the application image.

[0030] In one alternative embodiment, the device further includes:

[0031] The primary application scheduler is also used to receive the first scheduling instruction sent by the primary application scheduler at the application controller in the target data center, and to determine the application type of the computing power request before pulling the application image from the image repository of the target data center; in the case of a stateful application, it determines the data slice granularity based on the user service level agreement and computing network information.

[0032] The primary data manager receives the secondary scheduling instructions from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0033] The original data center's data controller is used to receive data slice granularity and target data center information, slice the application data corresponding to the computing power request, and migrate the sliced ​​data to the target data center.

[0034] In one alternative embodiment, the device further includes:

[0035] The application controller in the original data center is used to receive the first scheduling instruction sent by the first-level application scheduler in the application controller of the target data center, pull the application image from the image repository of the target data center, and then receive the third scheduling instruction sent by the first-level application scheduler to destroy the original application instance corresponding to the computing power request.

[0036] In one alternative approach, the application controller of the target data center is further configured to control the application instance to consume the corresponding business data and obtain the processing result after the application controller of the target data center creates the application instance based on the application image; and send the processing result to the primary application scheduler.

[0037] The primary application scheduler is also used to receive processing results and aggregate them.

[0038] In one alternative embodiment, the device further includes:

[0039] The data center's image repository is used to synchronize application images in the data center according to scheduling conditions and to synchronize the application images to the primary image repository.

[0040] The primary image repository is used to back up synchronized application images.

[0041] In one alternative embodiment, the device further includes:

[0042] The computing network perception layer is used to acquire real-time perception information from each data center and register the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0043] The computing network orchestration layer is used to perform computing power orchestration and network routing planning based on computing power information and network information, obtain computing power orchestration information and network routing information, and send the computing power orchestration information and network routing information to the first-level application scheduler.

[0044] According to another aspect of the present invention, an application orchestration and scheduling device is provided, comprising:

[0045] The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus.

[0046] The memory is used to store at least one executable instruction, which causes the processor to perform the following operations:

[0047] The first-level application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request;

[0048] The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, and synchronizes the application image from the image repository of the original data center to the image repository of the target data center.

[0049] The application controller in the target data center creates application instances based on the application image.

[0050] In an alternative manner, the executable instructions cause the processor to perform the following operations:

[0051] Before the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository in the target data center, the first-level application scheduler determines the application type of the computing power request.

[0052] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0053] The primary data manager receives the secondary scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0054] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0055] In an alternative manner, the executable instructions cause the processor to perform the following operations:

[0056] After the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository of the target data center, the application controller in the original data center receives the third scheduling instruction sent by the first-level application scheduler and destroys the original application instance corresponding to the computing power request.

[0057] In an alternative manner, the executable instructions cause the processor to perform the following operations:

[0058] After the application controller in the target data center creates an application instance based on the application image, the application controller in the target data center controls the application instance to consume the corresponding business data and obtain the processing result.

[0059] The application controller of the target data center sends the processing results to the primary application scheduler.

[0060] The primary application scheduler receives the processing results and aggregates them.

[0061] In an alternative manner, the executable instructions cause the processor to perform the following operations:

[0062] Before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the image repositories in each data center synchronize the application images of the data center according to the scheduling situation and synchronize the application images to the primary image repository.

[0063] The primary image repository backs up synchronized application images.

[0064] In an alternative manner, the executable instructions cause the processor to perform the following operations:

[0065] Before the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network perception layer obtains the real-time perception information of each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0066] After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and network routing information to the primary application scheduler.

[0067] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction that causes an application orchestration and scheduling device to perform the following operations:

[0068] The first-level application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request;

[0069] The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, and synchronizes the application image from the image repository of the original data center to the image repository of the target data center.

[0070] The application controller in the target data center creates application instances based on the application image.

[0071] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0072] Before the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository in the target data center, the first-level application scheduler determines the application type of the computing power request.

[0073] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0074] The primary data manager receives the secondary scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0075] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0076] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0077] After the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository of the target data center, the application controller in the original data center receives the third scheduling instruction sent by the first-level application scheduler and destroys the original application instance corresponding to the computing power request.

[0078] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0079] After the application controller in the target data center creates an application instance based on the application image, the application controller in the target data center controls the application instance to consume the corresponding business data and obtain the processing result.

[0080] The application controller of the target data center sends the processing results to the primary application scheduler.

[0081] The primary application scheduler receives the processing results and aggregates them.

[0082] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0083] Before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the image repositories in each data center synchronize the application images of the data center according to the scheduling situation and synchronize the application images to the primary image repository.

[0084] The primary image repository backs up synchronized application images.

[0085] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0086] Before the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network perception layer obtains the real-time perception information of each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0087] After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and network routing information to the primary application scheduler.

[0088] This invention, through a primary application scheduler, sends a query request to the computing network orchestration layer based on computing power requests. This query retrieves computing power orchestration information and network routing information from the computing network orchestration layer, thereby identifying a target data center with optimal computing resources suitable for executing computing tasks and providing external services. The application controller of the target data center receives a first scheduling instruction from the primary application scheduler and pulls an application image from the target data center's image repository. This application image is synchronized from the original data center's image repository to the target data center's image repository. The application controller then creates application instances based on these application images. By constructing a hierarchical control mechanism for application scheduling, this invention enables smooth scheduling and deployment of application instances across different data centers, launching application instances and providing external services, thus addressing the challenge of how ubiquitous computing power in a computing network can empower business applications.

[0089] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0090] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0091] Figure 1 This diagram illustrates a computing power processing scenario provided by an embodiment of the present invention.

[0092] Figure 2 A flowchart illustrating the application orchestration and scheduling method based on computing power networks provided in an embodiment of the present invention is shown.

[0093] Figure 3 The diagram illustrates a flowchart of the application invocation processing method in an application orchestration and scheduling method provided by an embodiment of the present invention.

[0094] Figure 4 The diagram shows a flowchart of a processing method for the network-aware orchestration part in an application orchestration and scheduling method provided by an embodiment of the present invention.

[0095] Figure 5 This diagram illustrates the structure of an application orchestration and scheduling device based on a computing power network, as provided in an embodiment of the present invention.

[0096] Figure 6 A schematic diagram of the application orchestration and scheduling device provided in an embodiment of the present invention is shown. Detailed Implementation

[0097] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0098] First, the technical terms involved in the embodiments of the present invention will be explained.

[0099] Computing power, or computing ability, specifically refers to the ability to collect, transmit, compute, and store data. The magnitude of computing power indicates the strength of the ability to process digital information.

[0100] A computing power network can refer to a network that connects computing power. Through a computing power network, the demand side that accesses the network can share all the computing power supply in the network.

[0101] An image can be a lightweight, executable, standalone software package used to package a software runtime environment and the software developed based on that runtime environment. It contains everything needed to run a piece of software, including code, runtime, libraries, environment variables, and configuration files.

[0102] Computing power networks are designed to solve the routing problem of computing power networks based on ubiquitous computing, but the following issues still exist for running business applications to actually use computing power:

[0103] 1. Smooth scheduling of business applications: While computing power network is aware of and orchestrates ubiquitous computing resources and optimizes network routing, business applications cannot smoothly migrate between different data centers based on the above information, and computing power resources cannot truly launch application instances and provide services to the outside world.

[0104] 2. Data remote access latency issues for stateful services: For stateful services, even a simple migration of application instances can lead to latency issues when accessing data across data centers, resulting in a decrease in the perceived performance of business applications.

[0105] In view of this, embodiments of the present invention provide an application orchestration and scheduling method, apparatus, device, and computer-readable storage medium based on a computing power network. This method addresses the issue of how ubiquitous computing power in a computing power network can empower business applications through a hierarchical control mechanism, enabling smooth scheduling of application instances across different data centers. Furthermore, it solves the latency problem caused by stateful services accessing data centers across data centers in a computing power network by using data slicing and automatic migration for distributed data consumption, thereby improving data consumption efficiency.

[0106] Figure 1 This diagram illustrates a computing power processing scenario provided by an embodiment of the present invention. Figure 1 As shown, the user terminal 11 sends a computing power request to the application orchestration and scheduling device 12 based on the computing power network. The application orchestration and scheduling device 12 can execute the application orchestration and scheduling method based on the computing power network as described in any embodiment of the present invention in response to the computing power request.

[0107] Here, the application orchestration and scheduling device 12 may include a network-aware orchestration section and an application scheduling section. The network-aware orchestration section includes a network awareness layer and a network orchestration layer. The network orchestration layer, used for computing power orchestration and network routing planning, may include a computing power orchestration module and a network routing module. The network awareness layer is used to acquire real-time computing power information such as remaining computing power information for each data center, and network information such as data center network topology and remaining bandwidth.

[0108] The application scheduling component can include a data plane and a control plane. The control plane can adopt a distributed architecture and may include an application scheduling center, data center application controllers deployed in various data centers, and data center data controllers. The application scheduling center may include a primary application scheduler and a primary data manager.

[0109] The Level 1 Application Scheduler is used to query computing network information based on computing power requests and the computing network orchestration layer, and to decide whether to perform application migration and data migration based on the Service-Level Agreement (SLA) and computing network information; it also aggregates the computing results of the same service in different data centers.

[0110] The primary data manager receives scheduling instructions from the primary application scheduler and schedules the data center data controllers in the data center.

[0111] The data center application controller receives scheduling instructions from the primary application controller, and starts or destroys application instances by pulling application images from the image repository within the data center; it also reports the business processing results of the data center to the primary application scheduler.

[0112] Here, the data center application controller can control, but is not limited to, container and virtual machine processes.

[0113] The data center data controller receives scheduling instructions from the primary data manager, slices the data of the business system in the data center, and migrates it to the data storage of the target data center.

[0114] The data plane includes application instances, data storage, image repositories, and primary image repositories deployed in various data centers.

[0115] Application instances can be application processes running in a data center, including but not limited to container or virtual machine processes.

[0116] Data storage is a device used to store data consumed by business systems, including but not limited to databases, block storage, and network attached storage (NAS).

[0117] The image repository is deployed within the data center and can be used to store application images of applications within the data center.

[0118] The primary image repository is used to store application images from across the network, serving as an image backup and centralized management tool.

[0119] In the above embodiments, the hierarchical control mechanism enables smooth scheduling and deployment of application instances across different data centers. Furthermore, by sharding and migrating data from stateful applications, latency for stateful application services accessing data centers across data centers is reduced. In addition, the computing power network and application hierarchical scheduling structure supplement the application deployment capabilities of the computing power network, achieving true WYSIWYG (What You See Is What You Get) and immediate usability of computing power through smooth hierarchical application deployment.

[0120] Figure 2 This diagram illustrates a flowchart of an application orchestration and scheduling method based on a computing power network provided in an embodiment of the present invention. The method comprises, as shown in the diagram... Figure 1 The application orchestration and scheduling device 12 based on the computing power network shown is executed. For example... Figure 2 As shown, the method includes the following steps:

[0121] Step 21: The primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request.

[0122] Here, a computing power request can be a computing power request sent by the user client for the target business application, which may include the computing power requirements of the target business. A query request can be to query the current optimal computing power orchestration information and network routing information. A target data center can be one or more data centers in the computing power network with superior computing and storage resources.

[0123] Step 22: The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository of the target data center.

[0124] Application images can be synchronized from the image repository of the original data center to the image repository of the target data center.

[0125] Here, the original data center and the target data center are different data centers. The first scheduling instruction can be used to instruct the application controller of the target data center to pull application images from the image repository of the target data center. The application image can be an image of the business application corresponding to the computing power request.

[0126] In this embodiment of the invention, the original data center's image repository synchronizes the application image to the primary image repository, which in turn synchronizes the application image to the target data center's image repository. This allows for the smooth scheduling and deployment of application images across different data centers.

[0127] Step 23: The application controller in the target data center creates an application instance based on the application image.

[0128] In the above embodiments, the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request to obtain the computing power orchestration information and network routing information sent by the computing network orchestration layer. This allows for the identification of a target data center with optimal current computing power resources, suitable for executing computing tasks and providing external services. The application controller of the target data center receives a first scheduling instruction from the primary application scheduler, pulls an application image from the target data center's image repository, synchronizes the application image from the original data center's image repository to the target data center's image repository, and creates application instances based on the application images. By constructing a hierarchical control mechanism for application scheduling, it is possible to achieve smooth scheduling and deployment of application instances of business applications across different data centers, launch application instances, and provide external services, thus solving the problem of how ubiquitous computing power in the computing network can empower business applications.

[0129] In an alternative approach, before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0130] The primary application scheduler determines the type of application requesting computing power.

[0131] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0132] The primary data manager receives the second scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0133] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0134] Here, application types can include stateful applications and stateless applications. A stateful application is one where the application's computing power request must be submitted to a data center that stores its relevant state information. A stateless application is one where the application's computing power request can include all the information needed to respond to the request.

[0135] The second scheduling instruction can be used to instruct the primary data manager to send data slice granularity and target data center information to the original data center. Application data can be stateful application data stored in the data center. Sliced ​​data can be data obtained by slicing application data.

[0136] In this embodiment of the invention, the data controller of the original data center receives the data slicing granularity and the target data center information, executes the data slicing instruction, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0137] In the above embodiments, the data of stateful applications is sharded and migrated according to the queried computing network information and the application type of the business system. Combined with application instance scheduling, the latency problem caused by stateful services accessing data centers across data centers in the computing power network can be solved, so as to enable distributed consumption of data and improve data consumption efficiency.

[0138] In an alternative approach, after the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0139] The original data center's application controller receives a third scheduling instruction from the primary application scheduler and destroys the original application instance corresponding to the computing power request.

[0140] Here, the third scheduling instruction can be used to instruct the destruction of the original application instance in the original data center. The original application instance corresponding to the computing power request can be the application process in the original data center that corresponds to the computing power request.

[0141] In the above embodiments, by destroying the original application implementation of the original data center, the data center's computing resources can be released from occupation, so that it can participate in application orchestration and scheduling in the future.

[0142] In an alternative approach, after the application controller in the target data center creates an application instance based on the application image, the method further includes:

[0143] The application controller in the target data center controls the application implementation to consume the corresponding business data and obtain the processing result.

[0144] The application controller of the target data center sends the processing results to the primary application scheduler.

[0145] The primary application scheduler receives the processing results and aggregates them.

[0146] Here, the processing result can be the result of computing power processing. In this embodiment of the invention, the target data center can be one or more data centers. The application controller of the target data center controls the respective application implementation to consume the corresponding business data, obtain their respective processing results, and send their respective processing results to the primary application scheduler. The primary application scheduler aggregates one or more processing results and performs aggregation processing on the processing results.

[0147] It should be noted that the multiple processing results are the processing results of the same business application in different data centers.

[0148] In the above embodiments, processing efficiency can be improved by distributing business data and aggregating processing results.

[0149] To fully understand application orchestration and scheduling methods based on computing power networks Figure 3 This diagram illustrates a flowchart of the application invocation processing method in an application orchestration and scheduling method provided by an embodiment of the present invention. Figure 3 As shown, the application's call to some processing methods includes the following steps:

[0150] Step 31: The primary application scheduler of the application scheduling center receives the computing power request for the target business application sent by the user terminal.

[0151] Step 32: The first-level application scheduler initiates a query operation (i.e., sends a query request) to the computing network orchestration layer based on the computing power request to obtain computing network information;

[0152] The computing network information may include the current optimal computing power arrangement information and network routing information.

[0153] Step 33: The primary application scheduler checks whether the computing power request is a stateful application. If the computing power request is a stateful application, proceed to step 37. If the computing power request is a stateless application, proceed to step 34.

[0154] Step 34: The primary application scheduler schedules the application controllers deployed within the target data center.

[0155] Step 35: The application controller of the target data center pulls the application image from its own image repository.

[0156] Step 36: The application controller of the target data center deploys the application implementation in the target data center.

[0157] Here, the application controller of the target data center creates an application instance based on the application image. The application instance is launched in this data center, and the application controller of the original data center destroys the original application instance.

[0158] Step 37: The primary application scheduler determines the data granularity based on the user SLA and computing network information.

[0159] Step 38: The first-level application scheduler schedules the first-level data manager to distribute data slice granularity information and target data center information.

[0160] Step 39: The target data center application controller pulls the application image.

[0161] Step 310: The original data center data controller executes the data slicing instruction and migrates the sliced ​​data to the target data center.

[0162] Step 311: The application controllers of each data center deploy and destroy application implementation examples according to instructions.

[0163] Here, the application controller in the target data center launches the application instance in the local data center, while the application controller in the original data center destroys the original application instance.

[0164] Step 312: Application instances in each data center consume their respective business data and aggregate the processing results to the primary application scheduler.

[0165] Step 313: The primary application scheduler aggregates the processing results from each data center.

[0166] In an alternative approach, before the primary application scheduler sends a query request to the network orchestration layer based on the computing power request, the method further includes:

[0167] The computing network perception layer acquires real-time perception information from each data center and registers this real-time perception information to the computing network orchestration layer.

[0168] Here, real-time sensing information may include, but is not limited to, computing power information and network information. Computing power information may include, but is not limited to, the remaining computing power information of the data center. Network information may include, but is not limited to, the network topology information, remaining bandwidth information, and network device information of the data center.

[0169] In an alternative approach, after the primary application scheduler sends a query request to the network orchestration layer based on the computing power request, the method further includes:

[0170] The computing network orchestration layer performs computing power orchestration and network routing planning based on computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and network routing information to the first-level application scheduler.

[0171] Here, the computing power orchestration information can include the computing power orchestration results. The network routing information can include the routing path information to the target data center.

[0172] To fully understand application orchestration and scheduling methods based on computing power networks Figure 4 This diagram illustrates a flowchart of a network-aware orchestration processing method in an application orchestration and scheduling method according to an embodiment of the present invention. Figure 4 As shown, the processing method for the network perception orchestration part may include the following steps.

[0173] Step 41, computing power and network information perception.

[0174] Here, the computing network perception layer deployed in the data center perceives the remaining computing power, network equipment information and network bandwidth information of the data center in real time (i.e., the computing network perception layer perceives real-time information).

[0175] Step 42: Report computing power and network information.

[0176] Here, the computing network perception layer registers real-time perception information to the computing network orchestration layer.

[0177] Step 43, computational orchestration and network routing planning.

[0178] Here, after receiving the query request, the computing power orchestration layer performs computing power orchestration and network routing planning based on real-time sensing information, obtains computing power orchestration information and network routing information, and synchronizes the computing power orchestration results and network routing information to the application scheduling center.

[0179] In an alternative approach, before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the method further includes:

[0180] Each data center's image repository synchronizes the data center's application images according to the scheduling situation and synchronizes the application images to the primary image repository;

[0181] The primary image repository backs up synchronized application images.

[0182] Here, the scheduling status can refer to the scheduling of application images. The primary image repository can be used to manage image information for image repositories in various data centers.

[0183] In this embodiment, each data center's image repository synchronizes the scheduled application images to the target data center's image repository based on the application image scheduling status of the primary application scheduler. Furthermore, each data center's image repository uploads the synchronized application images to the primary image repository, enabling the primary image repository to back up the synchronized application images.

[0184] Figure 5 A schematic diagram of an application orchestration and scheduling device based on a computing power network, provided by an embodiment of the present invention, is shown. Figure 5 As shown, the device 50 includes: a primary application scheduler 51 and an application controller 52 for the target data center.

[0185] The first-level application scheduler 51 is used to send a query request to the computing network orchestration layer according to the computing power request, obtain the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determine the target data center corresponding to the computing power request.

[0186] The application controller 52 of the target data center is used to receive the first scheduling instruction sent by the first-level application scheduler, pull the application image from the image repository of the target data center, synchronize the application image from the image repository of the original data center to the image repository of the target data center, and create application instances based on the application image.

[0187] In the above embodiments, the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request to obtain the computing power orchestration information and network routing information sent by the computing network orchestration layer. This allows for the identification of a target data center with optimal current computing power resources, suitable for executing computing tasks and providing external services. The application controller of the target data center receives a first scheduling instruction from the primary application scheduler, pulls an application image from the target data center's image repository, synchronizes the application image from the original data center's image repository to the target data center's image repository, and creates application instances based on the application images. By constructing a hierarchical control mechanism for application scheduling, it is possible to achieve smooth scheduling and deployment of application instances of business applications across different data centers, launch application instances, and provide external services, thus solving the problem of how ubiquitous computing power in the computing network can empower business applications.

[0188] In an alternative embodiment, the device 50 also includes a primary data manager and a data controller for the original data center.

[0189] The primary application scheduler 51 is also used to receive the first scheduling instruction sent by the primary application scheduler at the application controller of the target data center, and determine the application type of the computing power request before pulling the application image from the image repository of the target data center; if the application type is a stateful application, it decides the data slice granularity according to the user service level agreement and computing network information.

[0190] The primary data manager receives the secondary scheduling instructions from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0191] The original data center's data controller is used to receive data slice granularity and target data center information, slice the application data corresponding to the computing power request, and migrate the sliced ​​data to the target data center.

[0192] In an alternative embodiment, the device 50 also includes the application controller of the original data center.

[0193] The application controller in the original data center is used to receive the first scheduling instruction sent by the first-level application scheduler in the application controller of the target data center, pull the application image from the image repository of the target data center, and then receive the third scheduling instruction sent by the first-level application scheduler to destroy the original application instance corresponding to the computing power request.

[0194] In one alternative approach, the application controller of the target data center is further configured to control the application instance to consume the corresponding business data and obtain the processing result after the application controller of the target data center creates the application instance based on the application image; and send the processing result to the primary application scheduler.

[0195] The primary application scheduler is also used to receive processing results and aggregate them.

[0196] In an alternative embodiment, the device 50 also includes a data center mirror repository and a primary mirror repository.

[0197] The data center's image repository is used to synchronize application images in the data center according to scheduling conditions and to synchronize the application images to the primary image repository.

[0198] The primary image repository is used to back up synchronized application images.

[0199] In an alternative embodiment, the device 50 further includes a network sensing layer and a network orchestration layer.

[0200] The computing network perception layer is used to acquire real-time perception information from each data center and register the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0201] The computing network orchestration layer is used to perform computing power orchestration and network routing planning based on computing power information and network information, obtain computing power orchestration information and network routing information, and send the computing power orchestration information and network routing information to the first-level application scheduler.

[0202] Figure 6 The diagram shows a schematic of the application orchestration and scheduling device provided in an embodiment of the present invention. The specific implementation of the application orchestration and scheduling device is not limited by the specific embodiments of the present invention.

[0203] like Figure 6 As shown, the application orchestration and scheduling device may include: a processor 602, a communications interface 604, a memory 606, and a communications bus 608.

[0204] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608. Communication interface 604 is used to communicate with other network elements such as clients or other servers. The processor 602 executes program 610, specifically performing the relevant steps described in the embodiment of the application orchestration and scheduling method for computing power networks.

[0205] Specifically, program 610 may include program code, which includes computer-executable instructions.

[0206] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The application orchestration and scheduling device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0207] Memory 606 is used to store program 610. Memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0208] Specifically, program 610 can be called by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0209] The first-level application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request;

[0210] The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, and synchronizes the application image from the image repository of the original data center to the image repository of the target data center.

[0211] The application controller in the target data center creates application instances based on the application image.

[0212] In the above embodiments, the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request to obtain the computing power orchestration information and network routing information sent by the computing network orchestration layer. This allows for the identification of a target data center with optimal current computing power resources, suitable for executing computing tasks and providing external services. The application controller of the target data center receives a first scheduling instruction from the primary application scheduler, pulls an application image from the target data center's image repository, synchronizes the application image from the original data center's image repository to the target data center's image repository, and creates application instances based on the application images. By constructing a hierarchical control mechanism for application scheduling, it is possible to achieve smooth scheduling and deployment of application instances of business applications across different data centers, launch application instances, and provide external services, thus solving the problem of how ubiquitous computing power in the computing network can empower business applications.

[0213] In an alternative approach, program 610 may be invoked by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0214] Before the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository in the target data center, the first-level application scheduler determines the application type of the computing power request.

[0215] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0216] The primary data manager receives the secondary scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0217] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0218] In an alternative approach, program 610 may be invoked by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0219] After the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository of the target data center, the application controller in the original data center receives the third scheduling instruction sent by the first-level application scheduler and destroys the original application instance corresponding to the computing power request.

[0220] In an alternative approach, program 610 may be invoked by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0221] After the application controller in the target data center creates an application instance based on the application image, the application controller in the target data center controls the application instance to consume the corresponding business data and obtain the processing result.

[0222] The application controller of the target data center sends the processing results to the primary application scheduler.

[0223] The primary application scheduler receives the processing results and aggregates them.

[0224] In an alternative approach, program 610 may be invoked by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0225] Before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the image repositories in each data center synchronize the application images of the data center according to the scheduling situation and synchronize the application images to the primary image repository.

[0226] The primary image repository backs up synchronized application images.

[0227] In an alternative approach, program 610 may be invoked by processor 602 to cause the application orchestration and scheduling device to perform the following operations:

[0228] Before the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network perception layer obtains the real-time perception information of each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0229] After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and network routing information to the primary application scheduler.

[0230] This invention provides a computer-readable storage medium storing at least one executable instruction. When the executable instruction is executed on an application orchestration and scheduling device, the application orchestration and scheduling device performs the application orchestration and scheduling method based on computing power network in any of the above method embodiments.

[0231] Specifically, executable instructions can be used to cause the application orchestration and scheduling device to perform the following operations:

[0232] The first-level application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request;

[0233] The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, pulls the application image from the image repository of the target data center, and synchronizes the application image from the image repository of the original data center to the image repository of the target data center.

[0234] The application controller in the target data center creates application instances based on the application image.

[0235] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0236] Before the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository in the target data center, the first-level application scheduler determines the application type of the computing power request.

[0237] When the application type is a stateful application, the Level 1 application scheduler decides the data granularity based on the user service level agreement and the computing network information.

[0238] The primary data manager receives the secondary scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center.

[0239] The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center.

[0240] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0241] After the application controller in the target data center receives the first scheduling instruction sent by the first-level application scheduler and pulls the application image from the image repository of the target data center, the application controller in the original data center receives the third scheduling instruction sent by the first-level application scheduler and destroys the original application instance corresponding to the computing power request.

[0242] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0243] After the application controller in the target data center creates an application instance based on the application image, the application controller in the target data center controls the application instance to consume the corresponding business data and obtain the processing result.

[0244] The application controller of the target data center sends the processing results to the primary application scheduler.

[0245] The primary application scheduler receives the processing results and aggregates them.

[0246] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0247] Before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository in the target data center, the image repositories in each data center synchronize the application images of the data center according to the scheduling situation and synchronize the application images to the primary image repository.

[0248] The primary image repository backs up synchronized application images.

[0249] In an alternative approach, the executable instructions can specifically be used to cause the application orchestration and scheduling device to perform the following operations:

[0250] Before the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network perception layer obtains the real-time perception information of each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information.

[0251] After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and network information, obtains computing power orchestration information and network routing information, and sends the computing power orchestration information and network routing information to the primary application scheduler.

[0252] This invention provides a computer program that can be called by a processor to enable an application orchestration and scheduling device to execute the application orchestration and scheduling method based on computing power network in any of the above method embodiments.

[0253] This invention provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions, which, when executed on a computer, cause the computer to perform the application orchestration and scheduling method based on computing power network in any of the above method embodiments.

[0254] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0255] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0256] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim.

[0257] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0258] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. An application orchestration and scheduling method based on computing power networks, characterized in that, The method includes: The primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, obtains the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determines the target data center corresponding to the computing power request; The primary application scheduler determines the application type of the computing power request; When the application type is a stateful application, the first-level application scheduler makes decisions on the granularity of data slicing based on the user service level agreement and the computing network information. The primary data manager receives the second scheduling instruction from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center. The original data center's data controller receives the data slice granularity and target data center information, slices the application data corresponding to the computing power request, and migrates the sliced ​​data to the target data center; The application controller of the target data center receives the first scheduling instruction sent by the first-level application scheduler, and pulls the application image from the image repository of the target data center. The application image is synchronized from the image repository of the original data center to the image repository of the target data center. The application controller of the target data center creates an application instance based on the application image.

2. The method according to claim 1, characterized in that, After the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository of the target data center, the method further includes: The original data center's application controller receives a third scheduling instruction from the primary application scheduler and destroys the original application instance corresponding to the computing power request.

3. The method according to claim 1, characterized in that, After the application controller in the target data center creates an application instance based on the application image, the method further includes: The application controller in the target data center controls the application instance to consume the corresponding business data and obtain the processing result. The application controller of the target data center sends the processing result to the primary application scheduler. The primary application scheduler receives the processing results and performs aggregation processing on the processing results.

4. The method according to any one of claims 1 to 3, characterized in that, Before the application controller in the target data center receives the first scheduling instruction sent by the primary application scheduler and pulls the application image from the image repository of the target data center, the method further includes: Each data center's mirror repository synchronizes the data center's application images according to the scheduling situation, and synchronizes the application images to the primary mirror repository; The primary image repository backs up synchronized application images.

5. The method according to claim 1, characterized in that, Before the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the method further includes: The computing network perception layer acquires real-time perception information from each data center and registers the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information. After the primary application scheduler sends a query request to the computing network orchestration layer based on the computing power request, the method further includes: The computing network orchestration layer performs computing power orchestration and network routing planning based on the computing power information and the network information to obtain computing power orchestration information and network routing information, and sends the computing power orchestration information and the network routing information to the first-level application scheduler.

6. An application orchestration and scheduling device based on a computing power network, characterized in that, The device includes: The first-level application scheduler is used to send a query request to the computing network orchestration layer according to the computing power request, obtain the computing power orchestration information and network routing information sent by the computing network orchestration layer, and determine the target data center corresponding to the computing power request. The application controller of the target data center is used to receive the first scheduling instruction sent by the first-level application scheduler, pull the application image from the image repository of the target data center, and synchronize the application image from the image repository of the original data center to the image repository of the target data center; and create an application instance based on the application image. The primary application scheduler is also used to receive the first scheduling instruction sent by the primary application scheduler at the application controller in the target data center, and to determine the application type of the computing power request before pulling the application image from the image repository of the target data center; in the case of a stateful application, it determines the data slice granularity based on the user service level agreement and computing network information. The primary data manager receives the secondary scheduling instructions from the primary application scheduler and sends the data slice granularity and target data center information to the data controller of the original data center. The original data center's data controller is used to receive data slice granularity and target data center information, slice the application data corresponding to the computing power request, and migrate the sliced ​​data to the target data center.

7. The apparatus according to claim 6, characterized in that, The device further includes: The computing network perception layer is used to acquire real-time perception information from each data center and register the real-time perception information to the computing network orchestration layer. The real-time perception information includes computing power information and network information. The computing network orchestration layer is used to perform computing power orchestration and network routing planning based on the computing power information and the network information after the first-level application scheduler sends a query request to the computing network orchestration layer according to the computing power request, to obtain computing power orchestration information and network routing information, and to send the computing power orchestration information and network routing information to the first-level application scheduler.

8. An application orchestration and scheduling device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation of the application orchestration and scheduling method based on computing power network as described in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on the application orchestration and scheduling device, causes the application orchestration and scheduling device to perform the operation of the application orchestration and scheduling method based on computing power network as described in any one of claims 1-5.

Images