CN115879723A - Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform - Google Patents
Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform Download PDFInfo
- Publication number
- CN115879723A CN115879723A CN202211611718.5A CN202211611718A CN115879723A CN 115879723 A CN115879723 A CN 115879723A CN 202211611718 A CN202211611718 A CN 202211611718A CN 115879723 A CN115879723 A CN 115879723A
- Authority
- CN
- China
- Prior art keywords
- computing
- cloud
- terminal
- solution
- edge
- 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
-
- 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 computing resource scheduling system and a scheduling method based on a cloud side-end integrated platform, wherein the system comprises a terminal, a side end connected with the terminal, and a cloud end connected with the side end; the terminal is used for acquiring data of each node in a distributed new energy consumption scene, uploading a calculation task and node data to the side end according to calculation requirements, and receiving a calculation result executed by the side end task; the side terminal is used for receiving the transmission data of the terminal, uploading the information collected in the terminal and the state information of the computing nodes to the cloud terminal, and executing the computing task issued by the cloud terminal; the cloud end receives data information uploaded by the edge end, generates a corresponding computing resource scheduling strategy, issues a computing task request of the edge end based on the terminal to allocate resource conditions of different levels in a distributed new energy consumption scene, and fully utilizes computing capacity and storage capacity of the cloud end, the edge end and the end to realize efficient processing of data and efficient allocation of resources.
Description
Technical Field
The invention relates to power computing resource scheduling, in particular to a computing resource scheduling system based on a cloud edge-end integrated platform, and relates to the technical field of power computing resource scheduling.
Background
Under a novel electric power system scene, a large amount of distributed new energy equipment access electric wire netting, produced massive data and calculation demand from this, had a large amount of demands to cloud computing from this.
In the data transmission process of the traditional cloud computing mode, a large amount of data to be processed by a user needs to be gathered in a cloud data center, and the mode has high time delay and wastes the existing resources and causes great burden on cross-domain link bandwidth.
In the face of a traditional computing mode, a cloud edge collaborative computing system is provided. The core of the cloud edge coordination system is to extend partial services or capabilities (including but not limited to storage, computation, network, AI, big data, security, and the like) of the cloud to the edge node infrastructure, and realize unified resource scheduling, unified resource arrangement, unified deployment, and unified operation and maintenance capabilities between the central cloud and the edge node through the mutual cooperation between the central cloud and the edge node.
Aiming at the characteristics of dynamic change of service peaks and valleys, network connection of different services, time delay difference requirements and the like of power services in a novel power system scene, a reasonable and effective integrated computing power resource scheduling planning model which is suitable for a distributed new energy consumption scene is researched, and computing power planning and dynamic scheduling application are supported more urgently.
Therefore, how to effectively allocate resource conditions of different levels in a distributed new energy consumption scenario is a problem to be solved at present.
Disclosure of Invention
The purpose of the invention is as follows: the computing power resource scheduling system based on the cloud edge-end integrated platform is provided to solve the problems in the prior art.
The technical scheme is as follows: a computing power resource scheduling system based on a cloud edge-end integrated platform comprises:
the system comprises a terminal, an edge end connected with the terminal and a cloud end connected with the edge end; the terminal is used for acquiring data of each node in a distributed new energy consumption scene, uploading a calculation task and node data to the side end according to calculation requirements, and receiving a calculation result executed by the side end task; the side terminal is used for receiving the transmission data of the terminal, uploading the information collected in the terminal and the state information of the computing nodes to the cloud terminal, and executing the computing task issued by the cloud terminal; the cloud end receives the data information uploaded by the side end, then generates a corresponding computing resource scheduling strategy and sends the strategy to the side end; the cloud end, the edge end and the terminal are mutually connected, and the terminal uploads a correspondingly generated calculation task to the edge end according to calculation requirements; the side end uploads the information of the acquisition equipment and the state information of the computing node to the cloud end; the cloud end generates a corresponding computational resource scheduling strategy according to the information from the edge end and issues the strategy to the edge end; the edge terminal executes the strategy to execute the calculation task; the terminal receives the calculation result from the task execution of the side terminal.
In a further embodiment, the cloud includes a resource optimization module and a compute node selection module, and is configured to process a signal transmitted by an edge, generate a scheduling policy of a computational resource, and send the policy to the edge.
In a further embodiment, the terminal comprises a plurality of acquisition devices and a microprocessor unit;
the acquisition equipment comprises a plurality of multi-type sensor equipment and is used for acquiring various state data of each port in a new energy consumption scene, wherein the various state data comprise state quantities such as voltage, current, power and the like;
the micro-processing unit is used for dividing the calculation requirement into a plurality of tasks, uploading the calculation tasks to the side end, enabling data collected by the terminal to have certain calculation requirements, forming corresponding calculation tasks, calculating a part of the tasks at the terminal through a corresponding judgment strategy, and uploading the other part of the calculation tasks to the upper level for calculation when the calculation tasks exceed the calculation capacity in the existing state.
In a further embodiment, the edge is configured to upload a physical location of a state information computing node of the edge, a computing power status of the computing node, and a storage status of the computing node to a cloud;
and receiving a computing power resource scheduling strategy issued by the cloud, reasonably distributing the computing tasks from the terminal to each computing node, and finally issuing the computing results to the terminal.
In a further embodiment, the resource optimization module comprises the following steps:
s1: the total task N is decomposed into N subtasks, and each compute node t is used for computing the subtask i Is a set of m computing hotspots of different computing power, where:
N＝{n 1 ,n 2 ,n 3 ,...n n }；
T i ＝{t i1 ,t i2 ,t i3 ,...t im }；
T i m is a constant for the set of total computational powers;
s2: the attribute set of each compute hotspot and the attribute set of the compute node should satisfy the following formula:
S(t im )＝F(s w ,s r ,s c )；
S(T i )＝F(S w ,S r ,S c )；
calculation cost, S (t) im ) Is a hotspot attribute set, S (T) i ) To compute node attributes, S w For attributes corresponding to the workload, S r For attributes corresponding to turn-around time, S c To calculate the attributes for cost, X best The values are the set of workload, turnaround time and computational expense, and the formula is:
x best ＝{x w ,x r ,x c }，
wherein X best Is a three-dimensional optimal solution;
s3: the probability is determined for the second search for the roulette selection that yields the best solution, which determines the best value for the follow-up search based on the first better value, and is formulated as follows:
wherein S i Calculating the value of the attribute for the i point, S min As a single attribute minimum
S4: setting an iteration cycle, the number of times of single search and the number of tasks to be processed;
s5: at an initial value
The optimal solution is randomly searched for a plurality of times around, and the initial solution X is randomly selected by traversing the upper and lower bounds d0 Then a new feasible solution X is generated after adding the perturbation beta to the new feasible solution X d1 (ii) a Find fit from the fitness value of the point d1 And is combined withBy greedy comparison and initialization>
After the comparison, the @, which was obtained after the first round of search, is updated>
If the point attribute is abnormal, the point is firstly counted into a missing point, and then calculation scheduling resources are considered;
s6: calculating the probability P by the rules of roulette i Selecting a better value which is most suitable for greedy selection, and further searching out an optimal solution in the local part of a better feasible solution; step S2 shows that when the calculated solution and the target value are in a relatively normal condition, and the target value is smaller, the fitness function value will be larger, and the probability selection for selecting the corresponding wheel disk will be larger, which means that the first round of feasible solution search is repeated on the basis of selecting the first comparative better solution, and the two better solution searches obtain the final optimal solution, which is updated to the optimal solution
And then determining a subsequent optimal solution as->
S7: and repeating iterative search under the normal condition of the point, namely randomly searching for a better solution of the new attribute, if the better solution exists, counting the better solution into the optimal value of the missing point, and through a plurality of iterative cycles, obtaining the corresponding optimal resource scheduling value of the attribute of a single calculation task after the set iterative period is finished, and finally outputting a scheduling scheme.
In a further embodiment, to ensure that the property sets deployed to the edge computation can meet the criteria, the actual property sets should therefore satisfy the following formula:
wherein T is YD Is an index scheduled to the cloud.
In a further embodiment, X in step S5 d0 The calculation formula of (a) is as follows:
x d0 ＝L+rand (0,1) (U-L)；
wherein X d0 For the initial solution, U represents the upper limit in the allowed solution value interval, L represents the lower limit in the allowed solution value interval, wherein X d0 E (L, U), random function rand (0，1) Represents the interval [0,1]As a random multiple of U-L.
In a further embodiment, increasing the perturbation β yields a new feasible solution X d1 The formula of (1) is as follows:
x d1 ＝x d0 +β|x d0 -x rand |；
wherein X rand To be different from X in all scheduling resource feasible solutions d0 The solution of (1).
In a further embodiment, the formula for further searching the optimal solution in step S6 is:
and P is the selection probability of the feasible solution of the resource which can be determined as the optimal solution, and the greater the probability is, the better the optimal solution is selected by the following formula.
Has the advantages that: the invention discloses a computing power resource scheduling system based on a cloud edge end integrated platform, which can obtain a corresponding optimal resource scheduling value of a single computing task attribute after an iteration period is set through a plurality of iteration cycles, and finally output a scheduling scheme, meanwhile, because a small part of abnormal solutions of the computing attributes are inevitable when fitness is solved, namely a single attribute is smaller than the minimum value of a certain attribute, in order to improve the selection accuracy of the optimal solution, the step of iterative search under the normal condition of the point is repeated, namely a better solution of a new attribute is randomly searched, and if the optimal solution exists, the optimal solution is counted in the optimal value of the missing point, so that the selection accuracy of the optimal solution can be improved; and then based on a computing task request of a terminal, resource conditions of different levels in a distributed new energy consumption scene are effectively allocated, computing power and storage power of the cloud, the edge and the end are fully utilized, and efficient processing of data and efficient allocation of resources are achieved.
Drawings
Fig. 1 is a schematic structural diagram of a computing power resource scheduling framework based on a cloud edge-side integrated platform in an embodiment of the present invention.
FIG. 2 is a flow diagram of a resource optimization module in an embodiment of the invention.
Detailed Description
Through research and analysis of the applicant, the reason for the problem (the resource conditions of different levels in a distributed new energy consumption scene cannot be effectively allocated) is that a large amount of data to be processed by a user needs to be gathered in a cloud data center in the data transmission process of the traditional cloud computing mode, and the method is high in time delay, wastes the existing resources and causes great burden on cross-domain link bandwidth; in the face of a traditional computing mode, a computing system with cloud edge cooperation is provided. The core of the cloud edge cooperative system is to expand partial services or capabilities (including but not limited to storage, calculation, network, AI, big data, safety and the like) of the cloud onto edge node infrastructure, and realize unified resource scheduling, unified resource arrangement, unified deployment and unified operation and maintenance capabilities between the central cloud and the edge nodes through the mutual cooperation between the central cloud and the edge nodes; aiming at the characteristics of dynamic variation of service peaks and valleys, network connection of different services, time delay difference requirements and the like of power services in a novel power system scene, a reasonable and effective integrated computing power resource scheduling planning model which is suitable for a distributed new energy consumption scene is researched, and computing power planning and dynamic scheduling application are supported more urgently; and then based on a computing task request of a terminal, resource conditions of different levels in a distributed new energy consumption scene are effectively allocated, computing power and storage power of the cloud, the edge and the end are fully utilized, and efficient processing of data and efficient allocation of resources are achieved.
As shown in fig. 1 to 2, the system includes a terminal, an edge connected to the terminal, and a cloud connected to the edge; the terminal is used for acquiring data of each node in a distributed new energy consumption scene, uploading a calculation task and node data to the edge terminal according to calculation requirements, and receiving a calculation result executed by the edge terminal task; the side terminal is used for receiving the transmission data of the terminal, uploading the information collected in the terminal and the state information of the computing nodes to the cloud terminal, and executing the computing task issued by the cloud terminal; the cloud end receives the data information uploaded by the side end, then generates a corresponding computing resource scheduling strategy and sends the strategy to the side end; the cloud end, the edge end and the terminal are mutually connected, and the terminal uploads a correspondingly generated calculation task to the edge end according to calculation requirements; the method comprises the steps that the side end uploads information of acquisition equipment and state information of a computing node to a cloud end; the cloud end generates a corresponding computational resource scheduling strategy according to the information from the edge end and issues the strategy to the edge end; the edge terminal executes the strategy to execute the calculation task; the terminal receives the calculation result from the task execution of the side terminal.
The cloud end comprises a resource optimization module and a computing node selection module, and is used for processing signals transmitted by the edge end, generating a scheduling strategy of computing resources and issuing the strategy to the edge end;
the terminal comprises a plurality of acquisition devices and a micro-processing unit; the acquisition equipment comprises a plurality of multi-type sensor equipment and is used for acquiring various state data of each port in a new energy consumption scene, wherein the various state data comprise state quantities such as voltage, current, power and the like; the micro-processing unit is used for dividing the calculation requirement into a plurality of tasks, uploading the calculation tasks to the side end, enabling data collected by the terminal to have certain calculation requirements, forming corresponding calculation tasks, calculating a part of the tasks at the terminal through a corresponding judgment strategy, and uploading the other part of the calculation tasks to the upper level for calculation when the calculation tasks exceed the calculation capacity in the existing state.
The side end is used for uploading the physical position of the state information computing node, the computing power condition of the computing node and the storage condition of the computing node to the cloud end; and receiving a computing power resource scheduling strategy issued by the cloud, reasonably distributing the computing tasks from the terminal to each computing node, and finally issuing the computing results to the terminal.
In a further embodiment, the resource optimization module comprises the steps of:
s1: the total task N is decomposed into N subtasks, and each compute node t is used for computing the subtask i Is a set of m computing hotspots of different computing power, where:
N＝{n 1 ,n 2 ,n 3 ,...n n }；
T i ＝{t i1 ,t i2 ,t i3 ,...t im }；
T i m is a constant for the set of total computational powers;
s2: the attribute set of each computing hotspot and the attribute set of the computing node should satisfy the following formula:
S(t im )＝F(s w ,s r ,s c )；
S(T i )＝F(S w ,S r ,S c )；
wherein j and m are constants, m is the number of calculation hot spots, w is the workload, r is the turnaround time, c is the calculation cost, S (t) im ) Is a hotspot attribute set, S (T) i ) To compute node attributes, S w To workAttribute corresponding to load, S r For attributes corresponding to turn-around time, S c To calculate the attributes corresponding to the cost, X best The values are the set of workload, turnaround time and computational expense, and the formula is:
x best ＝{x w ,x r ,x c }，
wherein X best Is a three-dimensional optimal solution;
s3: to determine the probability of a roulette selection for the second search for the best solution, it is decided to select the follow-up search for the best value on the basis of the first better value, the formula is as follows:
wherein S i For the calculated attribute value of point i, S min As a single attribute minimum
S4: setting an iteration period, the number of times of single search and the number of tasks to be processed;
s5: at an initial value
The optimal solution is randomly searched for many times around, and the initial solution X is randomly selected by traversing the upper and lower bounds d0 Then a new feasible solution X is generated after adding the perturbation beta to the new feasible solution X d1 (ii) a Find fit from the fitness value of the point d1 And by greedy comparison and initialization>
After the comparison, the value obtained after the first search is updated>
If the point attribute is abnormal, the point is firstly counted into a missing point, and then calculation scheduling resources are considered;
s6: calculating the probability P by the rules of roulette i Selecting a better value which is most suitable for greedy selection, and further searching the part of a better feasible solutionObtaining an optimal solution; step S2 shows that when the calculated solution and the target value are in a relatively normal condition, and the target value is smaller, the fitness function value will be larger, and the probability selection for selecting the corresponding wheel disk will be larger, which means that the first round of feasible solution search is repeated on the basis of selecting the first comparative better solution, and the two better solution searches obtain the final optimal solution, which is updated to the optimal solution
Thereby determining a subsequent optimal solution as->
/>
S7: and repeating iterative search under the normal repeated condition at the point, namely randomly searching a better solution of the new attribute, if the better solution exists, counting the optimal value of the missing point, and through a plurality of iterative cycles, obtaining the corresponding optimal resource scheduling value of the attribute of a single calculation task after the set iterative period is finished, and finally outputting a scheduling scheme.
In a further embodiment, to ensure that the attribute set deployed to the edge computation can meet the criteria, the actual attribute set should therefore satisfy the following formula:
wherein T is YD Is an index scheduled to the cloud.
X in step S5 d0 The calculation formula of (a) is as follows:
x d0 ＝L+rand (0,1) (U-L)；
wherein X d0 For the initial solution, U and L represent the upper and lower bounds within the range of allowable solution values, where X d0 E (L, U), random function rand (0，1) Represents the interval [0,1]As a random multiple of U-L.
Generating new feasible solution X after increasing disturbance beta d1 The formula of (1) is as follows:
x d1 ＝x d0 +β|x d0 -x rand |；
wherein X rand To be different from X in all scheduling resource feasible solutions d0 The solution of (1).
The formula for further searching the optimal solution in step S6 is:
and P is the selection probability of the feasible solution of the resource which can be determined as the optimal solution, and the greater the probability is, the better the optimal solution is selected by the following formula.
Description of the working principle: through a plurality of iteration cycles, the corresponding optimal resource scheduling value of a single calculation task attribute can be obtained after the iteration period is set, and a scheduling scheme is finally output.
The preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, however, the present invention is not limited to the specific details of the embodiments, and various equivalent changes can be made to the technical solution of the present invention within the technical idea of the present invention, and these equivalent changes are within the protection scope of the present invention.
Claims (9)
1. A computing power resource scheduling system based on a cloud edge-side integrated platform is characterized by comprising:
the system comprises a terminal, an edge end connected with the terminal and a cloud end connected with the edge end;
the terminal is used for acquiring data of each node in a distributed new energy consumption scene, uploading a calculation task and node data to the side end according to calculation requirements, and receiving a calculation result executed by the side end task;
the side terminal is used for receiving the transmission data of the terminal, uploading the information collected in the terminal and the state information of the computing nodes to the cloud terminal, and executing the computing task issued by the cloud terminal;
and the cloud end receives the data information uploaded by the side end, then generates a corresponding computing resource scheduling strategy and issues the strategy to the side end.
2. The computing power resource scheduling system based on the cloud edge-end integrated platform as claimed in claim 1, wherein: the cloud comprises a resource optimization module and a computing node selection module, and is used for processing signals transmitted by the edge terminals, generating a scheduling strategy of computing resources and issuing the strategy to the edge terminals.
3. The computing resource scheduling system based on the cloud edge-side integrated platform is characterized in that the terminal comprises a plurality of acquisition devices and a micro-processing unit;
the acquisition equipment comprises a plurality of multi-type sensor equipment and is used for acquiring various state data of each port in a new energy consumption scene;
the micro-processing unit is used for dividing the calculation requirement into a plurality of tasks and uploading the calculation tasks to the side end.
4. The computing resource scheduling system based on the cloud edge integrated platform as claimed in claim 1, wherein the edge is configured to upload a physical location of a state information computing node of the edge, a computing power status of the computing node, and a storage status of the computing node to a cloud;
and receiving a computing power resource scheduling strategy issued by the cloud, reasonably distributing the computing tasks from the terminal to each computing node, and finally issuing the computing results to the terminal.
5. The scheduling method of the computing power resource scheduling system based on the cloud edge-side integrated platform is characterized in that the resource optimizing module comprises the following steps:
s1: the total task N is decomposed into N subtasks, and each compute node t is used for computing the subtask i Is a set of m computing hotspots of different computing power, where:
N＝{n 1 ,n 2 ,n 3 ,...n n }；
T i ＝{t i1 ,t i2 ,t i3 ,...t im }；
T i m is a constant for the set of total computational powers;
s2: the attribute set of each compute hotspot and the attribute set of the compute node should satisfy the following formula:
S(t im )＝F(s w ,s r ,s c )；
S(T i )＝F(S w ,S r ,S c )；
wherein j and m are constants, m is the number of calculation hot spots, w is the workload, r is the turnaround time, c is the calculation cost, S (t) im ) Is a hotspot attribute set, S (T) i ) To compute node attributes, S w For attributes corresponding to the workload, S r For attributes corresponding to turn-around time, S c To calculate the attributes corresponding to the cost, X best The values are the set of workload, turnaround time and computational expense, and the formula is:
x best ＝{x w ,x r ,x c }，
wherein X best Is a three-dimensional optimal solution;
s3: the probability is determined for the second search for the roulette selection that yields the best solution, which determines the best value for the follow-up search based on the first better value, and is formulated as follows:
wherein S i For the calculated attribute value of point i, S min As a single attribute minimum
S4: setting an iteration cycle, the number of times of single search and the number of tasks to be processed;
s5: at an initial value
The optimal solution is randomly searched for a plurality of times around, and the initial solution X is randomly selected by traversing the upper and lower bounds d0 Then a new feasible solution X is generated after adding the perturbation beta to the new feasible solution X d1 (ii) a Find fit from the fitness value of the point d1 And by greedy comparison and initialization>
After the comparison, the @, which was obtained after the first round of search, is updated>
If the attribute of the point is abnormal, the point is firstly counted into a missing point, and then the calculation scheduling resource is considered;
s6: calculating probability P by rules of roulette i Selecting a better value which is most suitable for greedy selection, and further searching out an optimal solution in the part of a better feasible solution; step S2 shows that when the calculated solution and the target value are in a relatively normal condition, and the target value is smaller, the fitness function value will be larger, and the probability selection for selecting the corresponding wheel disk will be larger, which means that the first round of feasible solution search is repeated on the basis of selecting the first comparative better solution, and the two better solution searches obtain the final optimal solution, which is updated to the optimal solution
And then determining a subsequent optimal solution as->
S7: and repeating iterative search under the normal condition of the point, namely randomly searching for a better solution of the new attribute, if the better solution exists, counting the better solution into the optimal value of the missing point, and through a plurality of iterative cycles, obtaining the corresponding optimal resource scheduling value of the attribute of a single calculation task after the set iterative period is finished, and finally outputting a scheduling scheme.
6. The scheduling method of the computing resource scheduling system based on the cloud edge-side integrated platform according to claim 3, wherein in order to ensure that the attribute set deployed to the edge computing can meet the standard, the actual attribute set should satisfy the following formula:
wherein T is YD Is an index scheduled to the cloud.
7. The scheduling method of the computing power resource scheduling system based on the cloud edge-end integrated platform as claimed in claim 3, wherein X in step S5 is d0 The calculation formula of (a) is as follows:
x d0 ＝L+rand (0,1) (U-L)；
wherein X d0 For the initial solution, U and L represent the upper and lower bounds within the range of allowable solution values, where X d0 E (L, U), random function rand (0，1) Represents the interval [0,1]As a random multiple of U-L.
8. The scheduling method of the computing power resource scheduling system based on the cloud edge-end integrated platform as claimed in claim 3, wherein a new feasible solution X is generated after the disturbance beta is increased d1 The formula of (1) is as follows:
x d1 ＝x d0 +β|x d0 -x rand |；
wherein X rand Is at the same timeDifferent from X in all scheduling resource feasible solutions d0 The solution of (1).
9. The scheduling method of the computational resource scheduling system based on the cloud edge-side integrated platform according to claim 3, wherein a formula for further searching for an optimal solution in step S6 is as follows:
and P is the selection probability of the feasible solution of the resource which can be determined as the optimal solution, and the greater the probability is, the better the optimal solution is selected by the following formula.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211611718.5A CN115879723A (en) | 2022-12-14 | 2022-12-14 | Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211611718.5A CN115879723A (en) | 2022-12-14 | 2022-12-14 | Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115879723A true CN115879723A (en) | 2023-03-31 |
Family
ID=85767607
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211611718.5A Pending CN115879723A (en) | 2022-12-14 | 2022-12-14 | Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115879723A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662021A (en) * | 2023-08-01 | 2023-08-29 | 鹏城实验室 | Collaborative scheduling system and method based on end-edge cloud architecture |
| CN117272838A (en) * | 2023-11-17 | 2023-12-22 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
-
2022
- 2022-12-14 CN CN202211611718.5A patent/CN115879723A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662021A (en) * | 2023-08-01 | 2023-08-29 | 鹏城实验室 | Collaborative scheduling system and method based on end-edge cloud architecture |
| CN117272838A (en) * | 2023-11-17 | 2023-12-22 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
| CN117272838B (en) * | 2023-11-17 | 2024-02-02 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115879723A (en) | Computing resource scheduling system and scheduling method based on cloud edge-end integrated platform | |
| Kumar et al. | Vehicular delay-tolerant networks for smart grid data management using mobile edge computing | |
| WO2017036244A1 (en) | Digital simulation system of power distribution network | |
| CN102473116B (en) | A method and system for job scheduling in distributed data processing system with identification of optimal network topology | |
| CN111200528B (en) | Intelligent linkage method for smart city with edge cloud cooperation | |
| CN113810233B (en) | Distributed computation unloading method based on computation network cooperation in random network | |
| CN111045828A (en) | Distributed edge calculation method based on distribution network area terminal and related device | |
| CN113703969A (en) | Power distribution Internet of things system capable of achieving multi-source data processing based on edge computing | |
| CN114567598A (en) | Load balancing method and device based on deep learning and cross-domain cooperation | |
| JP6779800B2 (en) | Distributed control system, distributed control method, distributed control system of power system and control method of power resources | |
| CN116662010B (en) | Dynamic resource allocation method and system based on distributed system environment | |
| Ashraf et al. | TOPSIS-based service arbitration for autonomic internet of things | |
| CN102339233A (en) | Cloud computing centralized management platform | |
| CN116700993B (en) | Load balancing method, device, equipment and readable storage medium | |
| CN113315659B (en) | Task collaborative planning method and system for intelligent factory | |
| Chen et al. | Robust geographical load balancing for sustainable data centers | |
| CN116385857A (en) | Calculation power distribution method based on AI intelligent scheduling | |
| CN116684472A (en) | Service deployment system and service deployment method for terminal-side computing network | |
| Xia et al. | Distributed resource management and admission control of stream processing systems with max utility | |
| Zeng et al. | Economical behavior modeling and analyses for data collection in edge Internet of Things networks | |
| CN113872673A (en) | Satellite constellation simulation container cloud resource scheduling system and scheduling method thereof | |
| Kargahi et al. | Utility accrual dynamic routing in real-time parallel systems | |
| CN112367275A (en) | Multi-service resource allocation method, system and equipment for power grid data acquisition system | |
| Ma et al. | Research on the on-demand scheduling algorithm of intelligent routing load based on SDN | |
| CN117724853B (en) | Data processing method and device based on artificial intelligence |
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 |