CN114721806A - Task scheduling and executing method and system based on digital twin - Google Patents
Task scheduling and executing method and system based on digital twin Download PDFInfo
- Publication number
- CN114721806A CN114721806A CN202210429702.6A CN202210429702A CN114721806A CN 114721806 A CN114721806 A CN 114721806A CN 202210429702 A CN202210429702 A CN 202210429702A CN 114721806 A CN114721806 A CN 114721806A
- Authority
- CN
- China
- Prior art keywords
- task
- execution
- digital twin
- service
- scheduling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5011—Pool
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5017—Task decomposition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/502—Proximity
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a task scheduling and executing method and system based on a digital twin, wherein a virtual twin entity is constructed for a physical entity by using a digital twin technology, and the operation of the digital twin is borne by a network combined with cloud edges, so that the state perception and control of the physical entity are more convenient and effective, the problem of a physical space is transferred to a digital space mode, the barrier between various types of resources is broken, various types of resources including the digital twin are deeply fused, the task scheduling efficiency is improved, and the execution requirements of various services are better met.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a task scheduling and executing method and system based on a digital twin.
Background
The digital twin is a simulation process integrating multidisciplinary, multi-physical quantity, multi-scale and multi-probability by fully utilizing data such as a physical model, sensor updating, operation history and the like, and mapping is completed in a virtual space, so that the full life cycle process of corresponding entity equipment is reflected. By constructing an interactive loop between the physical space and the digital space, the comprehensive, real-time and accurate digital representation of the physical entity is realized, and the digital twin can achieve the copy and interaction of the physical entity in the virtual space. The information transmission efficiency between a physical space and a network space can be greatly improved through a digital twin technology, and the efficient integration, organization and scheduling of various resources in different scenes are realized, so that the deep cooperative cooperation between the reality and the virtuality is realized, and the task execution efficiency is improved.
The digital twinning technique has gained wide and high attention in recent years, and both the academic community and the industrial community are actively thinking and studying the digital twinning technique and the application thereof. Many well-known enterprises are beginning to explore the application of digital twin technology in product design, manufacturing, and service. But the current research also considers less how to introduce the digital twinning technology in large-scale scenes to solve the practical application problem. Large-scale scenes are complex, involve a large area and many users and devices, and need to consider interaction and cooperation between the users and the devices, so that an interconnected network needs to be constructed, and multi-entity cooperative task execution including network, information resources and physical entities is realized.
In the problem of processing multi-entity cooperative task scheduling by using a digital twin technology, due to the large scene scale, multiple twin bodies and complex internal relation of tasks, in a traditional cloud computing scene, the distance from a terminal to a cloud platform is far, and the real-time interaction between a digital twin body and a corresponding physical entity is difficult to realize. The complex task internal relationship here means that the task to be considered can be divided into a combination of a plurality of services having a certain dependency relationship. The edge computing architecture draws the distance between the equipment or the physical entity and the server, can flexibly distribute various resources, and can effectively reduce the defects of the traditional cloud computing in the aspects of time delay and flexibility. Therefore, a cloud edge (cloud data center and edge computing device) cooperation mode is considered for task scheduling based on the digital twin, in the mode, various resources are dynamically deployed in a distributed mode at all places of a scene, and efficient operation of the system is achieved through a reasonable resource management and task scheduling mechanism. At this time, what needs to be solved is the problem of resource management and task scheduling based on the digital twin body in a complex scene, namely how to reasonably and efficiently schedule various resources and twin bodies, so that the resources of the fusion space can be reasonably configured to realize efficient execution of the responsible tasks, but at present, no related solution exists.
Disclosure of Invention
The invention aims to provide a task scheduling and executing method and system based on a digital twin, which can decompose a task into a plurality of services with finer granularity and certain dependency relationship, improve the task scheduling efficiency and better meet the executing requirements of various services.
The purpose of the invention is realized by the following technical scheme:
a task scheduling and executing method based on a digital twin is applied to a cloud edge collaborative network architecture based on the digital twin, and the method comprises the following steps:
constructing a digital twin body of an object entity, and combining various resources to form a fusion space resource pool;
decomposing each received task request into a series of service combinations with execution sequence, making scheduling decision for each service, distributing corresponding fusion space resources, and finishing task arrangement;
and performing corresponding tasks in a self-adaptive manner according to the task arrangement result, or constructing a pre-execution scene according to the task type, and then pre-executing the corresponding tasks by using the corresponding digital twin bodies.
A digital twin based task scheduling and execution system comprising: a spatial resource pool construction unit and a scheduling center are fused;
the fusion space resource pool construction unit is used for constructing a digital twin body of an object entity and combining various resources to form a fusion space resource pool;
the dispatch center includes: a composer and an actuator; wherein: the orchestrator is used for decomposing each received task request into a series of service combinations with execution sequences, performing scheduling decision for each service, allocating corresponding fusion space resources and finishing task orchestration; the executor is used for executing corresponding tasks in a self-adaptive mode according to task arrangement results or constructing a pre-execution scene according to task types and then pre-executing the corresponding tasks by using corresponding digital twin bodies
According to the technical scheme provided by the invention, the virtual twin entity is constructed for the physical entity by using the digital twin technology, and the operation of the digital twin body is borne by the network combined with the cloud edge, so that the state perception and control of the physical entity are more convenient and effective, the problem of the physical space is transferred to the digital space, the barrier between various types of resources is broken, the various types of resources including the digital twin body are deeply fused, the task scheduling efficiency is improved, and the execution requirements of various services are better met.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is an architecture diagram of a task scheduling and executing method based on digital twin according to an embodiment of the present invention;
FIG. 2 is a flowchart of a digital twin-based task scheduling and executing method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a cloud edge collaborative network architecture based on a digital twin according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating service composition of task 1 and task 2 according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating the optimization of multiple service combinations during execution according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a task scheduling model of an intelligent transportation scenario according to an embodiment of the present invention;
FIG. 7 is a diagram of a digital twin-based task scheduling and executing system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
The terms that may be used herein are first described as follows:
the terms "comprising," "including," "containing," "having," or other similar terms of meaning should be construed as non-exclusive inclusions. For example: including a feature (e.g., material, component, ingredient, carrier, formulation, material, dimension, part, component, mechanism, device, step, process, method, reaction condition, processing condition, parameter, algorithm, signal, data, product, or article, etc.) that is not specifically recited, should be interpreted to include not only the specifically recited feature but also other features not specifically recited and known in the art.
The following describes a task scheduling and executing method and system based on digital twin in detail. Details which are not described in detail in the embodiments of the invention belong to the prior art which is known to the person skilled in the art. Those not specifically mentioned in the examples of the present invention were carried out according to the conventional conditions in the art or conditions suggested by the manufacturer. The reagents and instruments used in the examples of the present invention are not specified by manufacturers, and are conventional products commercially available.
Example one
The embodiment of the invention provides a task scheduling and executing method based on a digital twin, which mainly comprises the following steps:
step 1, constructing a digital twin body of an object entity, and combining various resources to form a fusion space resource pool.
As will be appreciated by those skilled in the art, a physical entity may be understood as an entity that objectively exists in a physical space, such as a vehicle, a pedestrian, etc.
And 2, decomposing each received task request into a series of service combinations with execution sequences, carrying out scheduling decision for each service, distributing corresponding fusion space resources, and finishing task arrangement.
And 3, performing corresponding tasks in a self-adaptive mode according to task arrangement results, or constructing a pre-execution scene according to task types, and then performing the corresponding tasks in a pre-execution mode by using corresponding digital twins.
As shown in fig. 1, the whole architecture of the task scheduling and executing method is shown, and in the present invention, a fusion space resource pool is constructed by adding the whole various networks and information resources to the digital twin corresponding to the physical entity. The various resources herein include network bandwidth resources, data resources, storage resources, computing resources, and the like. The dispatching center comprises an orchestrator and an executor. The orchestrator is primarily responsible for analyzing the user's task requirements collected by the edge computing devices. For these tasks, the orchestrator decomposes them into different service combinations and represents them in the form of Directed Acyclic Graph (DAG), and then the orchestrator is responsible for arranging specific execution time and manner for these services, and processes each service based on various resources in the proposed scheduling policy scheduling scenario, so that the services can be executed orderly and efficiently. And the executor will make adjustments to the scheduling process for the specific instance at execution time. The scheduling center can run on the cloud data center and the edge computing device (corresponding to a centralized scheduling scheme and a distributed scheduling scheme), and meets the requirement of task execution by scheduling resources in the fusion space; in the task execution process, the task self-adaptive execution and pre-execution capabilities of adapting to task requirements and environmental state changes are achieved, and efficient operation of the tasks is achieved; in the embodiment of the invention, under the centralized scheduling scheme and the distributed scheduling scheme, the edge computing equipment acquires and submits the task requirements of the user.
In the task scheduling and executing method based on the digital twin provided by the embodiment of the present invention, which is applied to a cloud edge collaborative network architecture based on the digital twin, fig. 2 shows four parts included in the whole method: the method comprises the steps of a digital twin-based cloud edge collaborative network, representation of tasks, arrangement of the tasks and adaptive execution and pre-execution of the tasks. The following is a detailed description of the above four parts.
Firstly, a cloud edge cooperative network architecture based on digital twinning.
Interconnection of the cloud data center and the edge computing device is realized through high-performance computing, storage and network element devices and various communication links, and a digital twin-oriented cloud edge collaborative network infrastructure is constructed, as shown in fig. 3. The physical entity is accessed into the network through the network access equipment, and a digital twin is constructed through a digital twin technology and stored and operated in the cloud data center and the edge computing equipment. Cloud data centers, edge computing devices, have storage and computing capabilities, mainly differing in performance. However, cloud data centers are generally farther from the user side, while edge computing devices are closer to the user side, and therefore have less communication latency with users or physical devices. And forming a fusion space resource pool by the digital twin body and various network resources and information resources, and arranging the fusion space resource pool in a cloud edge cooperative network based on the digital twin body to finish various tasks by scheduling resources.
Centralized task decision scheduling can be performed on a cloud edge cooperative network architecture based on the digital twin, and distributed scheduling of tasks can also be performed. Specifically, the method comprises the following steps: the step 2 is realized by a composer, and the step 3 is realized by an actuator; the orchestrator and the executor form a scheduling center, and the scheduling center runs in a cloud data center or edge computing equipment of a digital twin-based cloud edge collaborative network; when the dispatching center operates in a cloud data center based on a digital twin cloud-side cooperative network, the dispatching center belongs to a centralized task dispatching and executing method; when the dispatching center runs on the edge computing equipment of the digital twin-based cloud edge collaborative network, the method belongs to a distributed task dispatching and executing method.
And II, representing the task.
In the embodiment of the invention, each task request is decomposed into a series of service combinations with different functions, and the series of service combinations are expressed by using a directed acyclic graph; each node in the directed acyclic graph is a service, the direction of edges between adjacent nodes represents the execution sequence between adjacent services, and the parameters of the edges are related data transmitted between adjacent services. Specifically, the method comprises the following steps:
the task representation described herein refers to the understanding, description, and processing of task requirements. As shown in FIG. 4, the present invention decomposes tasks into a series of services and then represents them using a directed acyclic graph. When various task demands are generated by a user, the task demands are firstly analyzed and decomposed into a series of service combinations with different functions. And executing the services according to the incidence relation described by the directed acyclic graph to realize the correct execution of the whole task. A service herein refers to some resource or capability provided by the system for accessing, allocating, scheduling, such as data, function interface, data transmission and processing capability, and executed action, and may also include some function of the physical entity.
For a directed acyclic graph, each node represents a service with a certain function, and the direction of an edge and the parameters of the edge are the execution sequence of the service in a task and relevant data needing to be transferred between the services respectively. The task execution time is the sum of the time spent on executing all the services in the service combination according to the described association relationship, and can be regarded as the time difference from the task request sent by the user to the task execution completed by the system. The execution target of the task is to complete the execution of the task within a limited time, or the total execution time of the task is minimum, or on the basis of completing the execution of the task, the resource consumption is minimum, and the like.
And thirdly, arranging tasks.
In the embodiment of the invention, the representation of the tasks and the arrangement of the tasks are realized by the orchestrator. The task arrangement of the invention refers to making a scheduling decision of the service and arranging corresponding fusion resources for the service so as to ensure the smooth execution of the service. After the representation of the tasks, each task will be described as a combination of services represented by directed acyclic graphs. When a plurality of task requests are continuously received, the tasks need to be quickly and efficiently processed to ensure the on-demand execution of the tasks or improve the execution efficiency of the tasks.
As shown in fig. 1, after decomposing the task requirements of the user and generating the directed acyclic graph of the service combination, the orchestrator assumes the task of scheduling the converged resource for the service combination. The orchestrator makes scheduling decisions according to the information, and reasonably distributes all resources of each service. Generally, scheduling decisions (applicable to both centralized and distributed scheduling schemes) can be made according to the service type, for example, for a computing type service, edge computing resources are allocated by merging spatial resource pools; for services of a storage category, distributing edge storage resources through a fusion space resource pool; for the service of the transmission class, the network bandwidth resource is distributed by merging the space resource pools. The index for judging the completion of the scheduling job may be various, and an important one of them is to ensure that all services are processed within a specified time, and further include the energy consumption condition and the like.
In the process of scheduling tasks, the method of the invention adopts a centralized type or a distributed type, and can flexibly select different scenes and tasks or cooperate with the scenes and the tasks. When the centralized type is adopted, the arrangement is carried out by the cloud data center, and when the distributed type is adopted, the arrangement is carried out by the edge computing equipment.
And fourthly, performing adaptive execution and pre-execution of the task.
On the basis of task arrangement, the system needs to have self-adaptive capacity when the task is executed. Adaptive refers to deciding how to adjust without user control or with only minimal control by the user. The system is endowed with self-adaptive capacity, so that the effect of task execution is better. The executor is responsible for realizing self-adaptive execution and pre-execution.
1. The main body of adaptive execution appears in the following three aspects:
1) and adjusting the arrangement scheme according to the network state. The network state and the resource distribution condition are monitored in real time, and if the network state and the resource distribution condition fluctuate (change greatly), the task execution scheme scheduled by the scheduler is not an optimal scheduling method and even cannot be successfully executed, so that the task scheduling scheme needs to be adaptively adjusted according to the conditions.
In this case, the sequence of the service and the resources allocated to the service are mainly adjusted according to the change of the network state and the task execution situation to meet the requirement of completing the task execution, so the adaptive adjustment task scheduling scheme includes the adjustment of the service sequence and the adjustment of the service resources.
2) Adaptive decision making based on perception. As shown in fig. 1, the acquisition and prediction of the service operation flow and the operation state are introduced in the whole process of representing and scheduling the task and executing various services by the resource. Acquiring all-element information in service execution including resource information, environment information and network state, and monitoring and analyzing the running state of the service; the obtained full-factor information and the operation state of the service obtained by analysis are combined, and self-adaptive adjustment is carried out according to the actual state of the scene when the service is executed; during operation, the physical entity and the twin body are synchronized in real time, and the physical entity really executes operation.
For example: the automatic driving vehicle sends a task request containing a series of services, a certain service is about to time out due to the change of the network state, the service sequence is adaptively adjusted at the moment, the service which is about to time out is preferentially executed, and more resources are allocated to the service.
3) Optimization of multiple service combinations. When the same service exists among different tasks and various parameters of the service are the same, the same service is executed only once, and the execution result is shared by corresponding different tasks. As shown in FIG. 5, the directed acyclic graph of task 1 and task 2 intersect at a "query services" node. That is, the node is a common service for task 1 and task 2. If the services required to be executed are the same and other parameters of the services are consistent, the service combination can be further optimized: the query service only needs to be executed once, and the two tasks share the service operation result. The system is endowed with the self-adaptive optimization capability in the execution service, so that the waste of resources can be further reduced, and the execution time and the cost can be reduced.
2. Pre-execution of tasks and services.
The invention also enables pre-execution of tasks and services, that is to say that these tasks and services can be performed in advance on a virtual space digital twin. Firstly, state prediction of individuals, environments and the like is carried out according to requirements, then a pre-execution scene is constructed according to task types (for example, some tasks with safety risks), relevant services in the tasks are executed in the pre-execution scene in advance by using corresponding digital twins, wherein during pre-execution, various data and results of an execution process are recorded, and effect evaluation is carried out after execution; the physical entities are not synchronized with the corresponding digital twin when pre-executed.
Since the digital twin is an exact mapping of the physical entities, the task execution results in all virtual spaces are the execution results in the real physical space. The tasks are executed in the virtual space in advance and the effect is observed, so that the fault tolerance rate of the task and service execution is improved, the safety of the physical space task execution is greatly improved, and the subsequent prediction work can be conveniently carried out.
For ease of understanding, the present invention is further described below with an example of an intelligent traffic scenario.
Fig. 6 is a schematic diagram of a task scheduling model when the present invention is applied to an intelligent transportation scenario. In an intelligent traffic scene, the components of the physical space are participants of the traffic scene, including environmental information of different vehicles, non-vehicles, pedestrians and the whole traffic; all the participants belong to physical entities, and firstly, digital twins corresponding to all object examples need to be constructed in a digital space. For a group of motor vehicles, various data in the driving process of the vehicles can be synchronously transmitted to a digital space through the internet of vehicles technology, wherein the data comprises the speed, the direction and the position information of the vehicles, various requirements of passengers in the vehicles and the like. For non-motor vehicles and pedestrians, the behavior tracks and states can be captured in real time through the drive test camera and the motor vehicle-mounted camera, and then the real-time behavior tracks and states are sent to the digital space. The whole traffic big environment comprises traffic signal lamp states, weather conditions, map information and the like. In establishing digital twins for physical entities, the emphasis and difficulty is on dynamic traffic participants.
After establishing the virtual-real mapping of the physical entity and the digital twin, the digital twin becomes a full-right agent of the physical entity, and the digital twin cooperates with various resources in the edge server to jointly form a fusion space resource pool.
After the construction of the fusion space resource pool is completed, the service scheduling request of the user can be processed. In the prior art, the relation between various service requirements generated in a complex scene is complicated, and the intelligent traffic scene also has the problem. When there are services proposed by the user in the scene, the edge server in the scene will receive these requirements and pass them to the orchestrator. The composer in the figure is an important component of the scheduling system. The orchestrator preprocesses the service requirements: and optimizing the directed acyclic graph path of the service combination, and transmitting the optimized service sequence to the executor.
At the core of the service executor is a task scheduling and executing module with intelligent processing and decision-making capability, and input data of the task scheduling and executing module is related information of service combination representation of directed acyclic graph representation of the tasks which are preprocessed. The executor makes the optimal decision of the fusion resource scheduling and feeds back the decision result to each edge server. Meanwhile, the decision result can also be used as feedback information to participate in the training and updating of the model so as to enhance the decision capability of the model. The edge server allocates various resources or mobilizes the digital twin to perform tasks according to the decision result. Meanwhile, the twin body and the physical entity keep real-time synchronization, so that the physical entity can perform actual operation.
In an intelligent traffic scene, the scheduling method provided by the invention collects the service requirements of different users in the scene through the edge server, and finally reasonably schedules resources to meet and solve various tasks and service requirements of the users through the continuous cooperation of the orchestrator and the executor. In the process, the execution of the service, the state of the environment, the scheduling condition of the resource and the like are monitored all the time, such as the running information of all vehicles, the signal light information of the crossroad, the real-time state of the road environment, the use condition of the resource in the server and the like.
Still further, the system provides the ability to pre-execute on virtual space digital twins for tasks and services that may be risky when executed in physical space. The following are exemplary: when a user driving in a scene meets an emergency, a temporary lane change and side parking requirement is generated. When the system dispatches and fuses space resources to meet the task requirement, due to the fact that certain potential safety hazards exist in service execution, previewing can be conducted in a digital twin pre-execution mode in a virtual space, actual operation is conducted in a physical space after previewing is successful, and risks are reduced. The digital twin space is mapped from the real physical space one by one, so that the real-time traffic condition of the road surface can be reflected really, the twin body is scheduled in the virtual space to execute the task, the physical entity does not execute the task really, and the effect of executing the task in the physical space can be achieved. This ensures that the security risk is minimised when the service is actually being performed.
The mode scheme provided by the embodiment of the invention greatly improves the information transmission efficiency between the physical space and the network space, and realizes the efficient integration, organization and scheduling of various resources in a scene, thereby achieving the deep cooperative cooperation between the virtual space and the real space and improving the task execution efficiency. The scheduling method designed by the invention has two task execution modes of distributed type and centralized type, and can be flexibly selected or used in a matched way aiming at different scenes and tasks. The method has the capability of self-adaptively executing the task, so that the method is more intelligent and reliable in the actual operation process. At the same time, by virtue of the digital twin-based pre-execution capability, security risks in task execution can be minimized.
Example two
The invention also provides a task scheduling and executing system based on digital twin, which is implemented mainly based on the method provided by the first embodiment, as shown in fig. 7, the system mainly includes: fusing a space resource pool construction unit and a scheduling center;
the fusion space resource pool construction unit is used for constructing a digital twin body of an object entity and combining various resources to form a fusion space resource pool;
the dispatch center includes: a composer and an actuator; wherein: the orchestrator is used for decomposing each received task request into a series of service combinations with execution sequences, performing scheduling decision for each service, allocating corresponding fusion space resources and finishing task orchestration; the executor is used for executing corresponding tasks in a self-adaptive mode according to task arrangement results, or constructing a pre-execution scene according to task types, and then pre-executing the corresponding tasks by using corresponding digital twin bodies.
It is obvious to those skilled in the art that, for convenience and simplicity of description, the above division of each functional module is only used for illustration, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the internal structure of the system is divided into different functional modules to complete all or part of the above described functions.
The principle of each part in the above system has been described in detail in the first embodiment, and thus is not described again.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A task scheduling and executing method based on a digital twin is characterized in that in a cloud edge collaborative network architecture based on the digital twin, the method comprises the following steps:
constructing a digital twin body of an object entity, and combining various resources to form a fusion space resource pool;
decomposing each received task request into a series of service combinations with execution sequence, making scheduling decision for each service, distributing corresponding fusion space resources, and finishing task arrangement;
and performing corresponding tasks in a self-adaptive manner according to the task arrangement result, or constructing a pre-execution scene according to the task type, and then pre-executing the corresponding tasks by using the corresponding digital twin bodies.
2. The method according to claim 1, wherein the constructing the digital twin of the object entity and combining various types of resources to form a fusion space resource pool comprises:
the method comprises the following steps of accessing a physical entity into a network through network access equipment, constructing a digital twin body through a digital twin technology, storing and operating the digital twin body in equipment based on a digital twin cloud edge cooperative network, wherein the equipment based on the digital twin cloud edge cooperative network comprises the following steps: a cloud data center and an edge computing device;
and forming a fusion space resource pool by the digital twin body and various network resources and information resources, and arranging the fusion space resource pool in a cloud edge cooperative network based on the digital twin body.
3. The method of claim 1, wherein the decomposing each received task request into a series of service combinations with an execution order comprises:
decomposing each task request into a combination of a series of services with different functions, and representing the combination of the series of services by using a directed acyclic graph; each node in the directed acyclic graph is a service, the direction of edges between adjacent nodes represents the execution sequence between adjacent services, and the parameters of the edges are related data transmitted between adjacent services.
4. The method as claimed in claim 1, wherein the step of making a scheduling decision for each service and allocating the corresponding converged spatial resource comprises:
performing scheduling decision according to the service type; for the service of the calculation type, distributing edge calculation resources by fusing a space resource pool; for services of a storage category, distributing edge storage resources through a fusion space resource pool; and for the service of the transmission class, allocating network bandwidth resources by fusing the space resource pool.
5. The method for task scheduling and execution based on digital twin as claimed in claim 1, wherein the task execution adaptive according to task arrangement result comprises:
and monitoring the network state and the resource distribution condition in real time, and if the current network state and the resource distribution condition are transmitted and fluctuated and the task scheduling result is not an optimal execution mode, adaptively adjusting the task scheduling scheme.
6. The method for task scheduling and execution based on digital twin as claimed in claim 1, wherein the task execution adaptive according to task arrangement result comprises:
acquiring all-element information in service execution including resource information, environment information and network state, and monitoring and analyzing the running state of the service;
combining the acquired full-factor information and the operation state of the service acquired by analysis, and performing self-adaptive adjustment according to the actual state of the scene during service execution; during operation, the physical entity and the twin body are synchronized in real time, and the physical entity really executes operation.
7. The method for task scheduling and execution based on digital twin as claimed in claim 1, wherein the task execution adaptive according to task arrangement result comprises:
when the same service exists among different tasks and various parameters of the service are the same, the same service is executed only once, and the execution result is shared by corresponding different tasks.
8. The method as claimed in claim 1, wherein the constructing of the pre-execution scenario according to task types and the re-performing of the corresponding task by using the corresponding digital twin comprises:
constructing a pre-execution scene according to the task type, and pre-executing related services in the task in the pre-execution scene by using the corresponding digital twin body, wherein during pre-execution, various data and results of an execution process are recorded, and effect evaluation is performed after execution; upon pre-execution, the physical entity does not remain synchronized with the corresponding digital twin.
9. A method for digital twin based task scheduling and execution according to claim 1,
decomposing each received task request into a series of service combinations with execution sequence, making scheduling decision for each service, distributing corresponding fusion space resources, and completing task arrangement by a scheduler;
adaptively executing corresponding tasks according to task arrangement results, or constructing a pre-execution scene according to task types, and pre-executing the corresponding tasks by using corresponding digital twin bodies through an actuator;
the orchestrator and the executor form a scheduling center, and the scheduling center runs in a cloud data center or edge computing equipment of a digital twin-based cloud edge collaborative network; when the dispatching center operates in a cloud data center based on a digital twin cloud edge collaborative network, the dispatching center belongs to a centralized task dispatching and executing method; when the dispatching center runs on the edge computing equipment of the digital twin-based cloud edge collaborative network, the method belongs to a distributed task dispatching and executing method.
10. A digital twin-based task scheduling and execution system, realized based on the method of any one of claims 1 to 9, the system comprising: fusing a space resource pool construction unit and a scheduling center;
the fusion space resource pool building unit is used for building a digital twin body of an object entity and combining various resources to form a fusion space resource pool;
the dispatch center includes: a composer and an actuator; wherein: the orchestrator is used for decomposing each received task request into a series of service combinations with execution sequences, performing scheduling decision for each service, allocating corresponding fusion space resources and finishing task orchestration; the executor is used for executing corresponding tasks in a self-adaptive mode according to task arrangement results, or constructing a pre-execution scene according to task types, and then pre-executing the corresponding tasks by using corresponding digital twin bodies.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210429702.6A CN114721806A (en) | 2022-04-22 | 2022-04-22 | Task scheduling and executing method and system based on digital twin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210429702.6A CN114721806A (en) | 2022-04-22 | 2022-04-22 | Task scheduling and executing method and system based on digital twin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114721806A true CN114721806A (en) | 2022-07-08 |
Family
ID=82245342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210429702.6A Pending CN114721806A (en) | 2022-04-22 | 2022-04-22 | Task scheduling and executing method and system based on digital twin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114721806A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116321036A (en) * | 2023-03-27 | 2023-06-23 | 南京邮电大学 | System and control method based on data twinning for green wireless resource management |
| CN116501478A (en) * | 2023-06-28 | 2023-07-28 | 中国电信股份有限公司 | Task allocation method, device, equipment, medium and digital twin system |
| CN117255087A (en) * | 2023-11-20 | 2023-12-19 | 中国科学技术大学 | Network twinning realization method and system supporting access of mass heterogeneous devices |
-
2022
- 2022-04-22 CN CN202210429702.6A patent/CN114721806A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116321036A (en) * | 2023-03-27 | 2023-06-23 | 南京邮电大学 | System and control method based on data twinning for green wireless resource management |
| CN116501478A (en) * | 2023-06-28 | 2023-07-28 | 中国电信股份有限公司 | Task allocation method, device, equipment, medium and digital twin system |
| CN117255087A (en) * | 2023-11-20 | 2023-12-19 | 中国科学技术大学 | Network twinning realization method and system supporting access of mass heterogeneous devices |
| CN117255087B (en) * | 2023-11-20 | 2024-02-23 | 中国科学技术大学 | Network twinning realization method and system supporting access of mass heterogeneous devices |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114721806A (en) | Task scheduling and executing method and system based on digital twin | |
| Zhou et al. | Exploiting moving intelligence: Delay-optimized computation offloading in vehicular fog networks | |
| Qi et al. | Scalable parallel task scheduling for autonomous driving using multi-task deep reinforcement learning | |
| CN113965568B (en) | Edge computing system for urban road C-V2X network | |
| Fellir et al. | A multi-Agent based model for task scheduling in cloud-fog computing platform | |
| Tang et al. | A container based edge offloading framework for autonomous driving | |
| Ivaschenko | Multi-agent solution for business processes management of 5PL transportation provider | |
| CN111311038A (en) | Evaluation method of traffic management and control service index | |
| CN113284370A (en) | Intelligent management system of city sharing unmanned aerial vehicle based on block chain technology | |
| Dai et al. | A software-defined-networking-enabled approach for edge-cloud computing in the internet of things | |
| Sharma et al. | AdPS: Adaptive priority scheduling for data services in heterogeneous vehicular networks | |
| Li et al. | A new fuzzy-based method for energy-aware resource allocation in vehicular cloud computing using a nature-inspired algorithm | |
| CN116755867B (en) | Satellite cloud-oriented computing resource scheduling system, method and storage medium | |
| Ouarnoughi et al. | Simulating multi-agent-based computation offloading for autonomous cars | |
| CN115208892B (en) | Vehicle-road collaborative online task scheduling method and system based on dynamic resource demand | |
| CN114444700A (en) | Quantum cloud computing platform job scheduling and resource allocation method | |
| CN114035919A (en) | Task scheduling system and method based on power distribution network layered distribution characteristics | |
| Xiao et al. | Mobile-edge-platooning cloud: a lightweight cloud in vehicular networks | |
| Grislin-Le Strugeon et al. | A multi-agent approach for vehicle-to-fog fair computation offloading | |
| CN113377503A (en) | Task scheduling method, device and system for collaborative AI (artificial intelligence) | |
| Ribeiro Jr et al. | HARMONIC: Shapley values in market games for resource allocation in vehicular clouds | |
| Tan et al. | Digital Twin-based Cloud-native Vehicular Networks Architecture for Intelligent Driving | |
| Sahoo et al. | AALB: Application Aware Load Balancing Algorithm for Road Side Units | |
| Jia et al. | A Survey on Task Scheduling Schemes in Mobile Edge Computing | |
| Filipe et al. | Cooperative, Connected and Autonomous Mobility: A Simulation Framework for Evaluating Coordination Mechanisms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |