CN111901145B - Power Internet of things heterogeneous shared resource allocation system and method - Google Patents

Power Internet of things heterogeneous shared resource allocation system and method Download PDF

Info

Publication number
CN111901145B
CN111901145B CN202010582955.8A CN202010582955A CN111901145B CN 111901145 B CN111901145 B CN 111901145B CN 202010582955 A CN202010582955 A CN 202010582955A CN 111901145 B CN111901145 B CN 111901145B
Authority
CN
China
Prior art keywords
service
edge
service request
resource
cloud
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.)
Active
Application number
CN202010582955.8A
Other languages
Chinese (zh)
Other versions
CN111901145A (en
Inventor
刘强
周俊
辛辰
邹明翰
邵苏杰
夏伟栋
王徐延
许洪华
张庆航
沙莉
吴磊
石亚骏
陶欣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Original Assignee
Beijing University of Posts and Telecommunications
Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications, Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd filed Critical Beijing University of Posts and Telecommunications
Priority to CN202010582955.8A priority Critical patent/CN111901145B/en
Publication of CN111901145A publication Critical patent/CN111901145A/en
Application granted granted Critical
Publication of CN111901145B publication Critical patent/CN111901145B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5003—Managing SLA; Interaction between SLA and QoS
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
    • 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
    • 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
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The application discloses a power internet of things heterogeneous shared resource allocation system and method, for a business request, resources are allocated in edge equipment, if the business request cannot be met, different opportunities are selected according to the service level of the business request to enter a cloud service center resource allocation stage, the resources are allocated in the cloud service center, from the perspective of resource service matching integration, service types and levels are distinguished, different cloud edge cooperative resources are allocated according to different service types and levels, processing requirements such as real-time performance corresponding to services are met, and edge resource optimal allocation is guaranteed. An effective technical solution is provided for heterogeneous resource sharing of the power Internet of things.

Description

Power Internet of things heterogeneous shared resource allocation system and method
Technical Field
The invention belongs to the technical field of power Internet of things, and relates to a power Internet of things heterogeneous shared resource allocation system and method based on cloud edge cooperation.
Background
With the comprehensive construction of the power internet of things integrating edge computing and the deepened development and updating of service application, the problems of insufficient processing capacity of edge equipment, insufficient utilization of distributed edge resources and the like under multi-service access become increasingly prominent, and the high-quality service provision of the power internet of things is influenced. Therefore, the resource allocation of the power internet of things needs to be further optimized, the service space-time distribution and the service access characteristics are considered, the service demand and the resource allocation are accurately matched, the use of network resources and the carrying capacity of the service are optimized, the resource utilization rate is improved, the influence on the service caused by insufficient resources at the edge side, unbalanced resource distribution and insufficient utilization is reduced, and the overall performance of the service is further improved.
However, at present, the power internet of things still faces a series of problems, influences brought by service type differentiation in resource allocation are not considered, and meanwhile, cooperation of edge resources is not sufficiently utilized. The method is based on the resource service matching integration, service categories are distinguished, different cloud-side cooperative resources are distributed according to different service providing types, the processing requirements of the service such as corresponding real-time performance are met, and edge resource optimal distribution is guaranteed. The method has the characteristic of distinguishing the service types, so that the method has more flexibility and expansibility in the resource allocation decision process, and an effective solution is provided for dynamic resource allocation.
In order to understand the development situation of the existing cloud-edge cooperative resource sharing technology, the existing papers and patents are searched, compared and analyzed, and the following technical information with high relevance to the invention is screened out:
the technical scheme 1: a patent of "a method and an apparatus for resource allocation based on edge computing", publication number CN110048882A, relates to a method and an apparatus for resource allocation based on edge computing, the method for resource allocation is characterized in that, according to the coupling degree between different service modules in an application system, all application systems in a server are decomposed into a plurality of meta service units, and a user experience Qo E index parameter value of each meta service unit is obtained; clustering a plurality of the meta-service units by using a clustering algorithm based on the Qo E index parameter value of each meta-service unit; and allocating resources for each cluster according to the Qo E index parameter values of all the meta service units in each cluster, so that all the meta service units in each cluster share the allocated resources. The resource allocation device is mainly completed by three steps, namely, an initialization module for decomposing all application systems in a server into a plurality of meta-service units according to the coupling degrees between different service modules in the application systems and acquiring a user experience Qo E index parameter value of each meta-service unit; secondly, a clustering module, configured to cluster a plurality of meta-service units by using a clustering algorithm based on a qoe index parameter value of each meta-service unit; thirdly, allocating resources for each of the clusters according to the qoe index parameter values of all the meta service units in each of the clusters, so that all the meta service units in each of the clusters share the allocated resources.
The method and the device for resource allocation based on edge computing provided by the technical scheme 1 comprise the steps of decomposing all application systems in a server into a plurality of meta-service units according to the coupling degrees among different service modules in the application systems, and acquiring a user experience QoE index parameter value of each meta-service unit; clustering a plurality of meta-service units by using a clustering algorithm based on the QoE index parameter value of each meta-service unit; and allocating resources for each cluster according to the QoE index parameter values of all the meta service units in each cluster, so that all the meta service units in each cluster share the allocated resources. However, the scheme does not consider the problem that a plurality of edge devices jointly carry one service, which may cause the utilization rate of edge resources to be low.
The technical scheme 2 is as follows: a patent of 'an edge calculation task allocation method based on a branch-and-bound method' with publication number CN110048882A, which relates to a task allocation method based on a branch-and-bound method, and is mainly completed through four steps, firstly, an edge calculation scene is arranged, and the scene is composed of a network model composed of a user terminal and a plurality of edge servers; secondly, describing a task initiated by the user terminal by using a DAG task graph G = (T, P), wherein the task has priority, and a subsequent task needs to start processing after all predecessor tasks of the subsequent task are completed; third, a network model formed by a plurality of edge servers connected to each other is described by a mesh network N = (a, D); fourthly, distributing the tasks in the second DAG task graph to a network model formed by connecting a third plurality of edge servers, in order to minimize the energy consumption of the edge servers for executing all the tasks and meet the constraint conditions, solving a task distribution matrix through a branch-and-bound method, wherein each element in the matrix represents whether the tasks are executed on the edge servers, if the tasks are executed on the edge servers, the value of the element is 1, otherwise, the value of the element is 0.
The edge computing task allocation method based on the branch-and-bound method provided by the technical scheme 2 minimizes the total energy consumption of task allocation on the premise of considering constraints such as task completion time and DAG parallel system reliability requirements. Firstly, relaxing the optimization problem, and solving a temporary solution by using an interior point method; and then, taking a discrete value as a decision variable of a first task in the temporary solution and meeting the constraint that one task can only be executed on one edge server, namely only one decision variable is taken as 1, and the others are all 0, traversing the value taking condition from the first edge server to a last edge server, continuously adopting an interior point method to solve the rest tasks and calculate energy consumption values of different solutions, taking a solution with the minimum energy consumption value, and repeating the step until the last task. However, the scheme needs to traverse all the edge servers, so that the total energy consumption of task allocation is increased, the service classification of the tasks is omitted, and the resource utilization rate is low.
The technical scheme 3 is as follows: a patent of "resource allocation algorithm for multi-access edge calculation" with publication number CN110856215A, which relates to a resource allocation algorithm for multi-access edge calculation. The method is mainly completed through three steps: firstly, calculating the energy consumption and the time consumption of a calculation task to be unloaded; secondly, the total cost consumption of the computing tasks to be unloaded is jointly calculated; third, selecting an appropriate calculator based on the local computing power of the user and the priority parameter of the delay requirement, the calculator comprising: the invention relates to a local terminal, an MEC server and a core network, which set priority for each user according to local computing capacity and delay requirement, and the user with high priority can unload and select the channel with better transmission condition in priority to obtain the minimum total cost.
The resource allocation algorithm for multi-access edge calculation provided by the technical scheme 3 has the advantages that the consideration factor is single, the total cost consumption is only considered in the optimization goal, and the deep consideration of the influence on the reasonable allocation of task resources caused by different task grades and service types is lacked. Efficient utilization of edge resources is not fully considered.
Disclosure of Invention
In order to solve the defects in the prior art, the application provides a power internet of things heterogeneous shared resource allocation system and a power internet of things heterogeneous shared resource allocation method.
In order to achieve the above object, the first invention of the present application adopts the following technical solutions:
the utility model provides a heterogeneous shared resource distribution system of electric power thing networking, includes terminal layer, edge layer, cloud computing layer and M service provider, its characterized in that:
the terminal layer comprises a plurality of electric power Internet of things service terminals, the service terminals generate service requests VB of different service types, and the VB attribute is a binary set { R, beta }, wherein R represents the quantity of resources required by the service requests, and beta represents the time delay threshold of the service requests;
the edge layer comprises N edge regions;
one edge Area comprises Q edge devices, and the edge Area uses the set Area i ＝{ED 1 ,ED 2 ,…,ED Q Denotes that 1. Ltoreq. I.ltoreq.N, the jth edge device in the ith edge region is denoted ED i,j ，1≤j≤Q；
An edge device accesses a plurality of service terminals and processes service requests obtained from the service terminals;
the cloud computing layer assists the edge layer resource processing service request by using a cloud service center;
the cloud service center comprises E cloud service resource providers, and each cloud service resource provider has n e A cloud service physical data center;
each cloud service resource provider manages the resource state of the cloud service physical data center managed by the cloud service resource provider through a uniform cloud service resource provider control unit, and all the cloud service resource providers support all types of service request services;
each service provider supports a class of service request types, the service request types being SP' s k Denotes, k =1,2,3, \ 8230;, M;
SP k class of service request EG k ，EG k ∈{1,2,3}，EG k ＝1、EG k =2 and EG k =3 denotes real time, quasi real time and non real time traffic, respectively.
The application also discloses another invention, namely a power internet of things heterogeneous shared resource allocation method, which is based on the power internet of things heterogeneous shared resource allocation system and comprises the following steps:
step 1: a service terminal sends a service request;
and 2, step: the edge device accessed by the service terminal receives the service request and judges the service type of the service request; if the service type of the service request is matched with the edge equipment accessed by the service terminal and the edge equipment meets the requirement of the service request, the edge equipment accessed by the service terminal distributes resources required by the service request and processes the service request on the edge equipment, otherwise, step 3 is executed;
and 3, step 3: the edge device accessed by the service terminal judges the grade of the service request, if the grade is real-time service or quasi-real-time service, the step 4 is executed; otherwise, turning to step 6;
and 4, step 4: judging whether the edge equipment set which can execute the service in the edge area of the service terminal meets the requirement of the service request, if so, processing the service request by matching the edge equipment with the service request in the edge area of the service terminal, otherwise, executing the step 5;
and 5: judging whether edge equipment in an adjacent edge area of the edge area where the service terminal is located meets the requirement of the service request, if so, screening the optimal adjacent edge area to match the edge equipment with the service request for processing, and otherwise, executing the step 6;
step 6: and the service request is sent to the cloud service center, and the cloud service center performs shared resource allocation and service request processing.
The invention further comprises the following preferred embodiments:
preferably, in step 2, the condition for determining whether the edge device accessed by the service terminal meets the service request requirement is as follows:
Figure BDA0002553709210000051
if the conditions are met, the edge device ED i,j The requirement of a business request VB is met;
wherein,
Figure BDA0002553709210000052
indicating ED i,j The amount of currently occupied resource, CED i,j Indicating ED i,j R denotes the number of resources required for the service request VB, T EX (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j The execution time of (1);
Figure BDA0002553709210000053
wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
Preferably, in step 4, the condition for determining whether the edge device set capable of executing the service in the edge region where the service terminal is located meets the service request requirement is as follows:
Figure BDA0002553709210000054
if the condition is satisfied, it indicates that Area is in the edge Area i The requirement of a business request VB can be met;
wherein,
Figure BDA0002553709210000055
Figure BDA0002553709210000056
indicating ED i,j The amount of currently occupied resources, CED i,j Indicating ED i,j The total number of resources of (a) is,
Figure BDA0002553709210000057
Q k representation collection
Figure BDA0002553709210000058
The size of (a) is (b),
Figure BDA0002553709210000059
representing the set of edge devices in the ith area which can execute the kth class service;
Figure BDA00025537092100000510
Figure BDA00025537092100000511
indicating edge devices ED i,j Whether or not to support service SP k ，
Figure BDA00025537092100000512
The indication is that it is supported,
Figure BDA00025537092100000513
indicating no support.
Preferably, in step 5, the condition for determining whether the edge device in the adjacent edge area of the edge area where the service terminal is located meets the service request requirement is as follows:
Figure BDA00025537092100000514
if the conditions are met, the edge equipment in the adjacent edge area of the edge area where the service terminal is located can meet the requirement of the service request VB;
wherein,
Figure BDA0002553709210000061
Figure BDA0002553709210000062
indicating ED i,j The amount of currently occupied resources, CED i,j Indicating ED i,j The total number of resources of (a) is,
Figure BDA0002553709210000063
Q k representation collection
Figure BDA0002553709210000064
The size of (a) is (b),
Figure BDA0002553709210000065
representing the set of edge devices in the ith area which can execute the kth class service;
Figure BDA0002553709210000066
Figure BDA0002553709210000067
indicating edge devices ED i,j Whether or not to support service SP k ，
Figure BDA0002553709210000068
The indication is that it is supported,
Figure BDA0002553709210000069
indicating no support.
Preferably, in step 5, if the edge device in the edge area adjacent to the edge area where the service terminal is located is an edge area set ASP for executing the kth class service k non-Area in (1) i Satisfies equation (9), equation (8) is applied to ASP k Each of which is not Area i Is judged, and a non-condition satisfying the formula (8)Area i Selecting the edge area closest to the edge area where the service terminal is located as an optimal adjacent edge area, wherein the optimal adjacent edge area is a service request matching edge device, and otherwise, executing the step 6;
preferably, the edge area where the service terminal is located in step 4 is the service request matching edge device, and the optimal adjacent edge area in step 5 is the service request matching edge device, and the matching method includes:
step a: calculating the available capacity of the edge equipment in the edge area and screening the edge equipment of which the available capacity meets the resource requirement of the service request;
step b: further screening edge devices which meet the service request resource requirements in terms of available capacity;
step c: and c, calculating the edge equipment obtained by screening in the step b according to an objective function of the shared resource distribution of the edge equipment to obtain the edge equipment for bearing the service request.
Preferably, the objective function of the edge device shared resource allocation in step c is:
RAgoal(VB,ED i,j )＝mincost VB,i,j
s.t. T_C(VB,ED i,j )≤β (6)
wherein, cost VB,i,j At the edge device ED for service requests VB i,j Cost of execution, T _ C (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j The completion time of (c).
Preferably, the cost VB,i,j The calculation formula is as follows:
cost VB,i,j ＝T_EX(VB,ED i,j )*P i,j (5)
wherein, P i,j Indicating edge devices ED i,j A price per unit time of the resource;
T_C(VB,ED i,j ) The calculation formula is as follows:
T_C(VB,ED i,j )＝T_W(VB,ED i,j )+T_EX(VB,ED i,j ) (3)
T_W(VB,ED i,j ) Indicating service request VB at edge device ED i,j Wait time of (c), T _ EX (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j Execution time of;
T_EX(VB,ED i,j ) The calculation formula is as follows:
Figure BDA0002553709210000071
wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
Preferably, in step 6, the cloud service center meets the formula through calculation and screening
Figure BDA0002553709210000072
Figure BDA0002553709210000073
All required cloud service resource providers obtain cloud service resource providers capable of receiving service requests; CC (challenge collapsar) e Representing the total resource amount of the E-th cloud service resource provider, E is more than or equal to 1 and less than or equal to E,
Figure BDA0002553709210000074
representing the number of resources currently occupied by the e-th cloud service resource provider;
for real-time or quasi-real-time business, the cloud service center further calculates the distance d between the control unit of the cloud service resource provider capable of receiving the business request and the cloud service physical data center managed by the cloud service resource provider e,g (ii) a Screening satisfies distance threshold constraints: d is a radical of e,g ≤ε VB And d is e,g And epsilon VB The cloud service physical data center with the minimum difference value serves as an optimal data center to distribute resources, epsilon, needed by service requests VB The method comprises the steps of representing conversion of service request delay constraint into a distance threshold value from a control unit capable of receiving a service request to a data center managed by the control unit;
for the non-real-time service, the cloud service center further screens the cloud service resource provider with the minimum capacity in the cloud service resource providers capable of receiving the service request as the optimal cloud service resource provider to receive the service request.
The beneficial effect that this application reached:
1. according to the method, the quality of service provision of the power Internet of things is improved through resource optimization allocation. According to the method, by utilizing cloud-edge cooperation and service providing ideas, from the perspective of resource service matching integration, service categories are distinguished, different cloud-edge cooperation resources are distributed according to different service providing types, processing requirements of corresponding services such as instantaneity are met, edge resource optimal distribution is guaranteed, and the resource utilization rate and service supporting capacity of the power internet of things are improved. The characteristic of distinguishing the service types in the method enables flexibility and expansibility in the resource allocation decision process.
2. The method is used for distributing the cloud-edge shared resources based on service division and mainly comprises the following two parts: the method comprises the steps of edge device shared resource allocation and cloud service center shared resource allocation. For the service request, resources are firstly distributed in the edge equipment, if the service request cannot be met, the resources are distributed in the cloud service center, and the time of entering the resource distribution stage of the cloud service center is distinguished according to different service grades, so that the problems that the processing capacity of the edge equipment in resource distribution is insufficient, the distributed edge resources are not sufficiently utilized, the influence brought by service type differentiation in resource distribution is lack of consideration, the cooperative utilization of the edge resources is not sufficiently utilized and the like in the power internet of things scene can be solved.
3. According to the method and the device, on the basis of ensuring that the service request resource demand is fully met, efficient utilization of the edge resources is preferentially ensured, and the purpose of improving the quality of service provision of the power internet of things is achieved.
Drawings
Fig. 1 is a schematic structural diagram of a power internet of things heterogeneous shared resource allocation system according to the present application;
fig. 2 is a flowchart of a power internet of things heterogeneous shared resource allocation method according to the present application;
fig. 3 shows service request resource allocation satisfaction rates for different service request numbers in the embodiment of the present application;
fig. 4 illustrates the number of edge devices determining service requests for different service request numbers in the embodiment of the present application;
fig. 5 shows service request resource allocation satisfaction rates for different numbers of edge devices in this embodiment.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
As shown in fig. 1, the power internet of things heterogeneous shared resource allocation system of the present application includes a terminal layer, an edge layer, a cloud computing layer, and M service providers;
the terminal layer comprises a plurality of electric power Internet of things service terminals, the service terminals generate service requests VB of different service types, and the VB attribute is a binary set { R, beta }, wherein R represents the quantity of resources required by the service requests, and beta represents the time delay threshold of the service requests; for example: the service request VB may be: and service data of service types such as video monitoring, state monitoring, charging pile and the like.
The edge layer comprises N edge regions;
one edge Area comprises Q edge devices, and the edge Area is used for collecting Area i ＝ {ED 1 ,ED 2 ,…,ED Q Denotes that 1 ≦ i ≦ N, and the jth edge device in the ith edge region is denoted ED i,j ，1≤j≤Q；
An edge device accesses a plurality of service terminals and processes service requests obtained from the service terminals;
the cloud computing layer provides data processing resources and service capacity by using a cloud service center. Considering that the capability of the edge device in the edge layer is limited and the edge device is enabled to process the service with high real-time requirement as much as possible, if the resource of the edge layer is insufficient or the real-time requirement of the service is not high, the corresponding service can be sent to the cloud service center for processing, and the use of the resource of the edge layer is optimized.
The cloud service center comprises E cloud service resource providers, and each cloud service resource provider has n e A cloud service physical data center;
each cloud service resource provider manages the resource state of the cloud service physical data center managed by the cloud service resource provider through a unified cloud service resource provider control unit, and all the cloud service resource providers support all types of service request services;
each service provider supports a class of service request types, SP being used for service request types k Denotes, k =1,2,3, \ 8230;, M;
in order to improve the resource allocation efficiency and ensure that the service request resource requirement is fully met, the service is allocated to the edge device supporting the type of service, and the edge region where the service terminal is located has priority. And (4) grading the n types of service terminal services by considering the differentiated resource requirements of the calculation data intensive service, the storage type service and the time delay sensitive service. The application classifies the received service of the request resource into three grades: real-time traffic, quasi-real-time traffic, and non-real-time traffic. Namely SP k Class of service request EG k ， EG k ∈{1,2,3}，EG k ＝1、EG k =2 and EG k =3 denotes real time, quasi real time and non real time traffic, respectively. The classification is generally related to service scenes and service attributes, and service requests with the real-time requirement of less than or equal to 1 second can be regarded as real-time services in a certain sense; the service request with the real-time requirement of 1 second-1 minute is a quasi-real-time service; the service request with the real-time requirement of more than or equal to 1 minute is a non-real-time service.
For real-time services and quasi-real-time services, preferentially distributing the services to the edge area where the service terminal is located, distributing the services to the adjacent edge area when the edge area where the service terminal is located cannot meet the requirements of resources and services, and finally matching edge equipment for the service request by the edge area distributed to the service request for processing; if the edge area where the service terminal is located and the adjacent edge area can not meet the resource and service requirements, sending the service terminal to a cloud service center, and performing resource sharing allocation and processing by the cloud service center;
for non-real-time services, when the edge equipment accessed by the service terminal cannot meet the resource and service requirements, the service resource request is directly sent to the cloud service center, and the cloud service center allocates and processes resources.
As shown in fig. 2, according to the power internet of things heterogeneous shared resource allocation method, based on the power internet of things heterogeneous shared resource allocation system, when an edge device ED is used i,j When receiving a service request VB sent by a service terminal and performing resource allocation, firstly judging the service type SP of the VB k Then according to SP k Performing corresponding processing on the class service request grade, specifically:
the method comprises the following steps:
step 1: a service terminal sends a service request;
and 2, step: the method comprises the steps that edge equipment accessed by a service terminal receives a service request and judges the service type of the service request; if the service type of the service request is matched with the edge equipment accessed by the service terminal and the edge equipment meets the requirement of the service request, the edge equipment accessed by the service terminal distributes resources required by the service request and processes the service request on the edge equipment, otherwise, step 3 is executed;
in the embodiment of the present application, the condition for determining whether the edge device accessed by the service terminal meets the service request requirement is as follows:
Figure BDA0002553709210000101
if the conditions are met, the edge device ED i,j The requirement of a business request VB is met; ED (electronic device) i,j Allocating the needed resources to VB and in ED i,j VB is processed.
Wherein,
Figure BDA0002553709210000102
indicating ED i,j The amount of currently occupied resource, CED i,j Indicating ED i,j R denotes the number of resources required for the service request VB, T EX (VB, ED) i,j ) Indicating service request VB at edge device ED i,j Execution time of;
Figure BDA0002553709210000103
wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
And 3, step 3: the edge device accessed by the service terminal judges the grade of the service request, if the grade is real-time service or quasi-real-time service, the step 4 is executed; otherwise, turning to step 6;
and 4, step 4: judging whether the edge equipment set which can execute the service in the edge area of the service terminal meets the requirement of the service request, if so, processing the service request by matching the edge equipment with the service request by the edge area of the service terminal, otherwise, executing the step 5;
in the embodiment of the present application, the condition for determining whether the edge device set capable of executing the service in the edge region where the service terminal is located meets the service request requirement is as follows:
Figure BDA0002553709210000111
the judgment condition includes a condition that a plurality of edge devices commonly meet the resources required by VB, and if the condition is met, the condition indicates that the Area is in the edge Area i Can meet the requirement of a business request VB, namely
Figure BDA0002553709210000112
One or more edge devices within can satisfy the resource requests required by the VB.
Wherein,
Figure BDA0002553709210000113
Figure BDA0002553709210000114
indicating ED i,j The amount of currently occupied resources, CED i,j Indicating ED i,j The total number of resources of (a) is,
Figure BDA0002553709210000115
Q k representation collection
Figure BDA0002553709210000116
The size of (a) is (b),
Figure BDA0002553709210000117
representing the set of edge devices in the ith area which can execute the kth class service;
Figure BDA0002553709210000118
Figure BDA0002553709210000119
indicating edge devices ED i,j Whether to support service SP k ，
Figure BDA00025537092100001110
The indication is that it is supported,
Figure BDA00025537092100001111
indicating no support.
When the conditions shown in (8) are met, the edge area where the service terminal is located matches the edge device for the service request, that is, the edge device set
Figure BDA00025537092100001112
The method for finding out the edge device which specifically confirms to bear the service request comprises the following steps:
step a: calculating the available capacity of the edge equipment in the edge area and screening the edge equipment of which the available capacity meets the resource requirement of the service request;
the total capacity of one edge device minus the used capacity is the available capacity of the edge device.
Step b: further screening edge devices which meet the service request resource requirements in terms of available capacity;
step c: and c, calculating the edge equipment obtained by screening in the step b according to an objective function of the shared resource distribution of the edge equipment to obtain the edge equipment for bearing the service request.
In the edge device shared resource allocation, in addition to considering the service level, the support situation of the edge device for different types of services needs to be further considered.
Let Q k Representation collection
Figure BDA00025537092100001113
The size of (a) is (b),
Figure BDA00025537092100001114
is Q k ×Q k The matrix is a matrix of a plurality of matrices,
Figure BDA00025537092100001115
to represent
Figure BDA00025537092100001116
The transmission time delay from the middle a edge device to the b edge device is that a is more than or equal to 1 and Q is more than or equal to b k 。
The set of edge regions that can perform class k services is represented as:
Figure BDA0002553709210000121
let N be k Representation set ASP k The size of (a) is smaller than (b),
Figure BDA0002553709210000122
is N k ×N k The matrix is a matrix of a plurality of pixels,
Figure BDA0002553709210000123
express ASP k The transmission delay from the c-th edge area to the d-th edge area is more than or equal to 1 and less than or equal to c, and d is less than or equal to N k 。
The attribute of a service request VB of a service terminal is defined as a binary set { R, beta }, wherein R represents the quantity of resources required by the service request, and beta represents the delay threshold of the service request. T _ C (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j Wherein the calculation formula can be described as:
T_C(VB,ED i,j )＝T_W(VB,ED i,j )+T_EX(VB,ED i,j ) (3)
T_W(VB,ED i,j ) Indicating resource requests VB at the edge device ED i,j Or (c) is detected. T _ EX (VB, ED) i,j ) Indicating resource request VB at edge device ED i,j The execution time of (c).
Figure BDA0002553709210000124
Wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
Service request VB at edge device ED i,j The cost of execution can be expressed as:
cost VB,i,j ＝T_EX(VB,ED i,j )*P i,j (5)
wherein P is i,j Indicating edge devices ED i,j Price per unit time of the resource.
For real-time or quasi-real-time services, the minimum resource cost needs to be ensured as far as possible on the basis of meeting time constraints. For non-real-time services, only the resource cost needs to be considered to be minimum.
The objective function of the edge device shared resource allocation can be expressed as:
RAgoal(VB,ED i,j )＝mincost VB,i,j
s.t. T_C(VB,ED i,j )≤β (6)
the method searches and determines the optimal edge device set to bear the service request, and completes the corresponding resource allocation. At this time, if the number of selected edge devices is 1, the device cannot be ED i,j ED may be included if multiple edge devices are selected i,j 。
And 5: judging whether edge equipment in an adjacent edge area of the edge area where the service terminal is located meets the requirement of the service request, if so, screening the optimal adjacent edge area to match the edge equipment with the service request for processing, and otherwise, executing the step 6;
in the embodiment of the present application, the condition for determining whether the edge device in the edge area adjacent to the edge area where the service terminal is located meets the service request requirement is as follows:
Figure BDA0002553709210000131
if the condition is satisfied, ED i,j Area of the edge Area i Cannot satisfy the resource request required by the VB, but other edge devices in the edge area adjacent to the edge area capable of executing the kth type of service can satisfy the resource request required by the VB, i.e. ASP k non-Area in (1) i One or more edge devices within the other edge region(s) may satisfy the resource request required by the VB.
If the condition (9) is satisfied, the condition (8) is applied to ASP k Each of which is not Area i The edge regions are judged, the adjacent edge regions meeting the condition (8) are screened, the edge region closest to the edge region where the service terminal is located is selected as the optimal adjacent edge region, and the optimal edge device set in the optimal adjacent edge region is found out by utilizing the steps a-c to meet the resource requirement of VB. Otherwise, step 6 is executed.
And 6: and the service request is sent to the cloud service center, and the cloud service center performs shared resource allocation and service request processing.
In the cloud service center shared resource allocation, the strong fusion computing capacity of cloud computing is considered, only the service level is considered, and all the cloud service resource providers support all types of service request services by default. The cloud service center has a plurality of cloud service resource providers, and each cloud service resource provider manages the resource state of the cloud service physical data center managed by the cloud service resource provider through a unified cloud service resource provider control unit. In view of the fact that each cloud service resource provider may operate a plurality of geographically dispersed cloud service physical data centers, in the cloud service shared resource allocation process, when allocating resources to real-time or quasi-real-time services, in addition to meeting resource requirements, time delay factors caused by transmission distances between the control unit and the physical data centers need to be considered, and non-real-time services do not need to be considered.
In order to optimize resource allocation as much as possible, for real-time or quasi-real-time services, service request processing delay needs to be increased, and service allocation is further optimized in consideration of the distance difference and the resource occupation state difference between each cloud service resource provider control unit and a specific cloud service physical data center managed by the cloud service resource provider control unit.
Specifically, the method comprises the following steps:
for real-time or quasi-real-time business, the cloud service center meets the formula through calculation and screening
Figure BDA0002553709210000132
Figure BDA0002553709210000141
All required cloud service resource providers obtain cloud service resource providers capable of receiving service requests;
e cloud service resource providers exist in the cloud service center, and the CC is ordered e Representing the total resource amount of the E-th cloud service resource provider, E is more than or equal to 1 and less than or equal to E,
Figure BDA0002553709210000142
indicating the amount of resources currently occupied by the e-th cloud service resource provider. N per cloud service resource provider e Physical data center of cloud service, and data center to cloud service resource provider controlThe distance of the system unit can be represented by D _ cud.
Figure BDA0002553709210000143
Figure BDA0002553709210000144
And d is e,g ∈D_cud
Figure BDA0002553709210000145
By finding that the distance threshold constraint is satisfied: d is a radical of e,g ≤ε VB And satisfy d e,g ＝argmin h [δ 1,h -ε VB ]The data center (VB) is used as an optimal data center to distribute the resources required by the service request VB and confirm to process the VB. argmin h [δ 1,h -ε VB ]Indicating the distance d of h control units to the data center e,g With a distance threshold epsilon VB The minimum difference of (c).
For non-real-time business, the cloud service center meets the formula through calculation and screening
Figure BDA0002553709210000146
And all the required cloud service resource providers obtain the cloud service resource providers capable of receiving the service request, and then, the cloud service resource provider with the minimum capacity in the cloud service resource providers capable of receiving the service request is further screened as the optimal cloud service resource provider to receive the service request.
The feasibility and the high efficiency of the proposed scheme are verified through simulation numerical results, and the simulation results show that compared with a baseline method which is only executed on the edge side, namely Full-local and DTMO, the resource allocation efficiency of the edge computing network is greatly improved.
Assume that the number of edge devices is 100, the service type is 10, and there are 10 edge devices per edge device zone. The number of real-time and quasi-real-time service requests accounts for 60% of the total number of service requests per time slot. The number of non-real time service requests accounts for 40% of the total number of service requests per time slot.
As the number of service requests increases, full-local, DTMO and the present application compare the resource allocation based on service partition with respect to the number of edge devices for confirming service requests and the service request resource allocation satisfaction rate as shown in fig. 3 and 4, it can be known from fig. 3 that executing all service requests at the edge layer does not guarantee that the resource requirements of all requests are confirmed to be executed, and the service request resource allocation satisfaction rate of Full-local is lower than that of the DTMO and the resource allocation method based on service partition. And the service request resource allocation satisfaction rate is highest based on the resource allocation of the service division.
Fig. 5 shows the comparison result of Full-local, DTMO and resource allocation based on service division with respect to the service request resource allocation satisfaction rate when the number of edge devices increases from 50 to 500 in the case of a fixed number of service requests. The results show that the Full-local efficiency is greatly affected as edge devices grow. In contrast, the service classification is carried out on the service requests, the distributed heterogeneous resource allocation is carried out according to the service types, the resource allocation is carried out based on the service division, the performance of the distributed heterogeneous resource allocation method is superior to that of Full-local and DTMO, the service requests are effectively shunted, certain decision flexibility is achieved, and resources are effectively allocated. As can be seen from the results, as the number of edge devices increases, the service request resource allocation degree is higher, because the number of edge devices is more, the number of edge resources that can provide the completion request is more, and the efficiency of service request resource allocation and the utilization rate of the edge resources are significantly improved. Therefore, the method and the device provide an effective technical solution for heterogeneous resource sharing of the power internet of things.
Abbreviation:
the DTMO is a framework for dynamic tracking, monitoring and editing of cloud resources, namely an A frame for dynamic tracking, monitoring and editing of closed resources, which is called DTMO for short.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are only preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for the purpose of limiting the scope of the present invention, and on the contrary, any modifications or modifications based on the spirit of the present invention should fall within the scope of the present invention.

Claims (9)

1. The utility model provides a heterogeneous shared resource distribution system of electric power thing networking, includes terminal layer, edge layer, cloud computing layer and M service provider, its characterized in that:
the terminal layer comprises a plurality of electric power Internet of things service terminals, the service terminals generate service requests VB of different service types, and the VB attribute is a binary set { R, beta }, wherein R represents the quantity of resources required by the service requests, and beta represents the time delay threshold of the service requests;
the edge layer comprises N edge regions;
one edge Area comprises Q edge devices, and the edge Area is used for collecting Area i ＝{ED 1 ,ED 2 ,…,ED Q Denotes that 1 ≦ i ≦ N, and the jth edge device in the ith edge region is denoted ED i,j ，1≤j≤Q；
An edge device accesses a plurality of service terminals and processes service requests obtained from the service terminals;
the cloud computing layer assists the edge layer resource in processing the service request by utilizing the cloud service center;
the cloud service center comprises E cloud service resource providers, and each cloud service resource provider has n e A cloud service physical data center;
each cloud service resource provider manages the resource state of the cloud service physical data center managed by the cloud service resource provider through a unified cloud service resource provider control unit, and all the cloud service resource providers support all types of service request services;
each service provider supports a class of service request types, the service request types being SP' s k Denotes, k =1,2,3, \ 8230;, M;
SP k class of service request EG k ，EG k ∈{1,2,3}，EG k ＝1、EG k =2 and EG k =3 real-time, quasi-real-time and non-real-time traffic, respectively;
when the edge equipment set which can execute real-time or quasi-real-time service in the edge area of the service terminal meets the requirement of the service request, the edge area of the service terminal matches the edge equipment for the service request to process; when the edge equipment in the adjacent edge area of the edge area where the service terminal is located meets the requirement of the service request, screening the optimal adjacent edge area to match the edge equipment for the service request for processing; the specific matching mode is as follows:
calculating the available capacity of the edge equipment in the edge area and screening the edge equipment of which the available capacity meets the resource requirement of the service request;
further screening edge devices which meet the service request resource requirements in terms of available capacity;
and calculating the edge equipment which accords with the service request time delay threshold value according to an objective function of the shared resource distribution of the edge equipment to obtain the edge equipment for bearing the service request.
2. An electric power internet of things heterogeneous shared resource allocation method based on the electric power internet of things heterogeneous shared resource allocation system of claim 1, characterized in that:
the method comprises the following steps:
step 1: a service terminal sends a service request;
step 2: the edge device accessed by the service terminal receives the service request and judges the service type of the service request; if the service type of the service request is matched with the edge equipment accessed by the service terminal and the edge equipment meets the requirement of the service request, the edge equipment accessed by the service terminal allocates resources required by the service request and processes the service request on the edge equipment, otherwise, the step 3 is executed;
and 3, step 3: the edge device accessed by the service terminal judges the grade of the service request, if the grade is real-time service or quasi-real-time service, the step 4 is executed; otherwise, turning to step 6;
and 4, step 4: judging whether the edge equipment set which can execute real-time or quasi-real-time service in the edge area of the service terminal meets the requirement of the service request, if so, processing the service request by matching the edge equipment with the edge area of the service terminal, otherwise, executing the step 5;
and 5: judging whether edge equipment in an adjacent edge area of the edge area where the service terminal is positioned meets the requirement of the service request, if so, screening the optimal adjacent edge area as the service request to be matched with the edge equipment for processing, otherwise, executing the step 6;
step 6: and the service request is sent to the cloud service center, and the cloud service center performs shared resource allocation and service request processing.
3. The power internet of things heterogeneous shared resource allocation method according to claim 2, characterized in that:
in step 2, the condition for judging whether the edge device accessed by the service terminal meets the service request requirement is as follows:
Figure FDA0003802373130000021
if the condition is satisfied, the edge device ED i,j The requirement of a business request VB is met;
wherein,
Figure FDA0003802373130000022
indicating ED i,j The amount of currently occupied resources, CED i,j Indicating ED i,j R denotes the number of resources required for the service request VB, T _ EX (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j Execution time of;
Figure FDA0003802373130000031
wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
4. The power internet of things heterogeneous shared resource allocation method according to claim 2, characterized in that:
in step 4, the condition for judging whether the edge device set capable of executing real-time or quasi-real-time service in the edge region where the service terminal is located meets the service request requirement is as follows:
Figure FDA0003802373130000032
if the condition is satisfied, the Area in the edge Area is indicated i The requirement of a business request VB can be met;
wherein,
Figure FDA0003802373130000033
Figure FDA0003802373130000034
indicating ED i,j The amount of currently occupied resources, CED i,j Indicating ED i,j The total number of resources of (a) is,
Figure FDA0003802373130000035
Q k representation collection
Figure FDA0003802373130000036
The size of (a) is (b),
Figure FDA0003802373130000037
indicating a set of edge devices in the ith area which can execute the kth class service;
Figure FDA0003802373130000038
Figure FDA0003802373130000039
indicating edge devices ED i,j Whether or not to support service SP k ，
Figure FDA00038023731300000310
The indication is that it is supported,
Figure FDA00038023731300000311
indicating no support.
5. The power internet of things heterogeneous shared resource allocation method according to claim 4, wherein the power internet of things heterogeneous shared resource allocation method comprises the following steps:
in step 5, the condition for determining whether the edge device in the adjacent edge area of the edge area where the service terminal is located meets the service request requirement is as follows:
Figure FDA00038023731300000312
if the conditions are met, the edge devices in the adjacent edge areas of the edge areas where the service terminals are located can meet the requirements of the service request VB;
wherein,
Figure FDA00038023731300000313
Figure FDA00038023731300000314
indicating ED i,j The amount of currently occupied resource, CED i,j Indicating ED i,j The total number of resources of (a) is,
Figure FDA00038023731300000315
Q k representation collection
Figure FDA00038023731300000316
The size of (a) is (b),
Figure FDA00038023731300000317
representing the set of edge devices in the ith area which can execute the kth class service;
Figure FDA0003802373130000041
Figure FDA0003802373130000042
indicating edge devices ED i,j Whether to support service SP k ，
Figure FDA0003802373130000043
The support is represented by a representation of,
Figure FDA0003802373130000044
indicating no support.
6. The power internet of things heterogeneous shared resource allocation method according to claim 5, wherein:
in step 5, if the edge device in the adjacent edge region of the edge region where the service terminal is located, the edge region set ASP executing the kth class service k non-Area in (1) i If one or more edge devices in the edge region satisfy equation (9), equation (8) is applied to ASP k Each of which is not Area i The edge Area of (2) is determined, and the non-Area satisfying the formula (8) is determined i The edge area of the service terminal is selected as the optimal adjacent edge area, the optimal adjacent edge area is the edge equipment matched with the service request, and if not, the step 6 is executed.
7. The power internet of things heterogeneous shared resource allocation method according to claim 2, characterized in that:
the objective function of the edge device shared resource allocation is as follows:
RAgoal(VB,ED i,j )＝mincost VB,i,j
s.t.T_C(VB,ED i,j )≤β (6)
wherein, cost VB,i,j At the edge device ED for service requests VB i,j Cost of execution, T _ C (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j The completion time of (c).
8. The method for allocating the heterogeneous shared resources of the power internet of things according to claim 7, wherein:
the cost VB,i,j The calculation formula is as follows:
cost VB,i,j ＝T_EX(VB,ED i,j )*P i,j (5)
wherein, P i,j Indicating edge devices ED i,j A price per unit time of the resource;
T_C(VB,ED i,j ) The calculation formula is as follows:
T_C(VB,ED i,j )＝T_W(VB,ED i,j )+T_EX(VB,ED i,j ) (3)
T_W(VB,ED i,j ) Indicating service request VB at edge device ED i,j Wait time of (c), T _ EX (VB, ED) i,j ) Indicating service requests VB at the edge device ED i,j Execution time of;
T_EX(VB,ED i,j ) The calculation formula is as follows:
Figure FDA0003802373130000051
wherein, abt (ED) i,j ) Indicating edge devices ED i,j CR (VB) represents the computational resource requirements of the service request VB.
9. The power internet of things heterogeneous shared resource allocation method according to claim 2, wherein:
in step 6, the cloud service center meets the formula through calculation and screening
Figure FDA0003802373130000052
All required cloud service resource providers obtain cloud service resource providers capable of receiving service requests; CC (challenge collapsar) e Representing the total resource amount of the E-th cloud service resource provider, E is more than or equal to 1 and less than or equal to E,
Figure FDA0003802373130000053
representing the number of resources currently occupied by the e-th cloud service resource provider;
for real-time or quasi-real-time business, the cloud service center further calculates the distance d between the control unit of the cloud service resource provider capable of receiving the business request and the cloud service physical data center managed by the cloud service resource provider e,g (ii) a Screening satisfies distance threshold constraints: d is a radical of e,g ≤ε VB And d is d e,g And epsilon VB The cloud service physical data center with the minimum difference value serves as an optimal data center to distribute resources, epsilon, needed by service requests VB The method comprises the steps of representing conversion of service request delay constraint into a distance threshold value from a control unit capable of receiving a service request to a data center managed by the control unit;
for the non-real-time service, the cloud service center further screens the cloud service resource provider with the minimum capacity in the cloud service resource providers capable of receiving the service request as the optimal cloud service resource provider to receive the service request.
CN202010582955.8A 2020-06-23 2020-06-23 Power Internet of things heterogeneous shared resource allocation system and method Active CN111901145B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010582955.8A CN111901145B (en) 2020-06-23 2020-06-23 Power Internet of things heterogeneous shared resource allocation system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010582955.8A CN111901145B (en) 2020-06-23 2020-06-23 Power Internet of things heterogeneous shared resource allocation system and method

Publications (2)

Publication Number Publication Date
CN111901145A CN111901145A (en) 2020-11-06
CN111901145B true CN111901145B (en) 2022-11-08

Family

ID=73206959

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010582955.8A Active CN111901145B (en) 2020-06-23 2020-06-23 Power Internet of things heterogeneous shared resource allocation system and method

Country Status (1)

Country Link
CN (1) CN111901145B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112671852A (en) * 2020-12-14 2021-04-16 北京邮电大学 Internet of things equipment computing unloading framework based on in-network computing platform
CN112969107B (en) * 2021-02-01 2023-02-28 广东浩云长盛网络股份有限公司 Resource optimization method for multidimensional collaboration of data center optical network
CN113065785B (en) * 2021-04-13 2024-02-20 国网江苏省电力有限公司信息通信分公司 Dynamic resource expansion method for electric power Internet of things management platform
CN113556380A (en) * 2021-06-07 2021-10-26 广东东华发思特软件有限公司 Edge distributed multi-copy processing method, device and medium of Internet of things equipment
CN114595049A (en) * 2022-03-15 2022-06-07 国网智能电网研究院有限公司 Cloud-edge cooperative task scheduling method and device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108990159B (en) * 2018-07-12 2023-04-07 东南大学 Combined resource allocation method based on hierarchical game in mobile edge computing system
CN109218414A (en) * 2018-08-27 2019-01-15 杭州中恒云能源互联网技术有限公司 A kind of distributed computing method of smart grid-oriented hybrid network framework
CN111045828B (en) * 2019-12-27 2021-08-27 南方电网电力科技股份有限公司 Distributed edge calculation method based on distribution network area terminal and related device

Also Published As

Publication number Publication date
CN111901145A (en) 2020-11-06

Similar Documents

Publication Publication Date Title
CN111901145B (en) Power Internet of things heterogeneous shared resource allocation system and method
CN108076486B (en) Dynamic wireless resource allocation algorithm based on load balancing
CN110996393B (en) Single-edge computing server and multi-user cooperative computing unloading and resource allocation method
CN111277437A (en) Network slice resource allocation method for smart power grid
CN110505644B (en) User task unloading and resource allocation joint optimization method
CN113918240B (en) Task unloading method and device
CN111614754B (en) Fog-calculation-oriented cost-efficiency optimized dynamic self-adaptive task scheduling method
CN108901075B (en) GS algorithm-based resource allocation method
CN104619029B (en) It is a kind of centralization cellular network architecture under baseband pool resource allocation methods and device
CN105446817B (en) A kind of federated resource configuration of reservations algorithm based on robust optimization in mobile cloud computing
CN103401938A (en) Resource distribution system based on service features under distributed cloud architecture and method thereof
CN113891477A (en) Resource allocation method based on MEC calculation task unloading in Internet of vehicles
CN111915142A (en) Unmanned aerial vehicle auxiliary resource allocation method based on deep reinforcement learning
CN116708446B (en) Network performance comprehensive weight decision-based computing network scheduling service method and system
EP4055787A1 (en) Cdn optimization platform
CN114124955A (en) Calculation unloading method based on multi-agent game
CN116916386A (en) Large model auxiliary edge task unloading method considering user competition and load
Sun et al. A resource allocation scheme for edge computing network in smart city based on attention mechanism
CN113747450A (en) Service deployment method and device in mobile network and electronic equipment
CN117176733A (en) Multi-service resource management method and system for power distribution network
CN115080249B (en) Vehicle networking multidimensional resource allocation method and system based on federal learning
CN115756772A (en) Dynamic arrangement and task scheduling method and system for edge cloud mixed operation
CN111738582B (en) Dynamic allocation method, device and equipment for banking website tasks based on community
CN113055854A (en) NOMA-based vehicle edge computing network optimization method, system, medium and application
CN110996396A (en) Mobile edge network resource allocation method based on online combined auction

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
GR01 Patent grant
GR01 Patent grant