CN117240806A - Network resource allocation and scheduling method under super fusion architecture - Google Patents
Network resource allocation and scheduling method under super fusion architecture Download PDFInfo
- Publication number
- CN117240806A CN117240806A CN202311525561.9A CN202311525561A CN117240806A CN 117240806 A CN117240806 A CN 117240806A CN 202311525561 A CN202311525561 A CN 202311525561A CN 117240806 A CN117240806 A CN 117240806A
- Authority
- CN
- China
- Prior art keywords
- node
- resource
- resources
- network
- nodes
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000013468 resource allocation Methods 0.000 title claims abstract description 28
- 230000004927 fusion Effects 0.000 title claims abstract description 21
- 238000010801 machine learning Methods 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims abstract description 4
- 230000002068 genetic effect Effects 0.000 claims description 11
- 238000012546 transfer Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 2
- 238000012216 screening Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Abstract
The invention relates to the technical field of communication, and discloses a network resource allocation and scheduling method under a super fusion architecture, which comprises the following steps: s100, detecting and collecting total resource use conditions of network nodes or resource access amount data of each node in a period of time in a super fusion architecture, and then transmitting the detected and collected data to a central server; s200, predicting the resource access amount and the resource demand of the node in a future period by constructing a model of flow classification such as machine learning or semantic understanding in the super-fusion structure; s300, constructing a time sequence prediction model in a central server to allocate network resources of each node; s400, if the node resource task cannot be met in the resource allocation process, the scheduling model is constructed to reallocate part of the tasks at the node, so that the node resource allocation method has the advantages of being capable of autonomously allocating network resources, spontaneously scheduling the resources of the node which cannot be met, and the like.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a network resource allocation and scheduling method under a super fusion architecture.
Background
Along with the development of science and technology and the improvement of living standard of people, the application occasions of cloud computing and big data technology are wider, the demands for network resources are increased while the scale of a data center is increased, the cloud computing completes the fusion of hardware resources and software resources by virtue of the characteristics of use according to the demands and payment according to the quantity of the cloud computing, a resource pool is formed and a unified scheduling interface is provided for user service, the cost is greatly reduced, the usability is improved, the daily life of a user is greatly facilitated, and how to effectively allocate and schedule the network resources to meet the demands of various applications becomes an important problem.
The cloud computing based on the super-fusion architecture adopts distributed storage, a plurality of resources and technologies such as an information network, information storage, virtualization and the like are simultaneously arranged in one unit device, so that the management difficulty of the server architecture is reduced, the multi-unit device realizes the self-modularization transverse expansion by means of network aggregation, and becomes the main stream of the cloud computing architecture, however, the problem of cloud security of acquiring resources by unauthorized users is brought about a great threat to a cloud platform, the access control technology is an important means for solving the unauthorized override, the current access control model cannot meet the requirements of the cloud platform on fine granularity, dynamic property and easy manageability, and under the super-fusion architecture, the traditional network resource allocation and scheduling method cannot always meet the requirements due to the fact that the interaction and the dependence relationship of various resources are more complex, and a new network resource allocation mode and scheduling method are particularly important.
Disclosure of Invention
Aiming at the defects that the current access control model in the prior art cannot meet the requirements of cloud platform fine granularity, dynamic property and easy manageability, and the conventional network resource allocation and scheduling method cannot meet the requirements due to more complex interaction and dependency relationship of various resources under a super-fusion architecture, the invention provides the network resource allocation and scheduling method under the super-fusion architecture, which has the advantages of autonomously allocating network resources, spontaneously scheduling the resources of nodes which cannot meet the requirements, and the like.
In order to achieve the above purpose, the present invention provides the following technical solutions: a method for allocating and scheduling network resources under a super fusion architecture, the method comprising:
s100, detecting and collecting total resource use conditions of network nodes or resource access amount data of each node in a period of time in a super fusion architecture, and then transmitting the detected and collected data to a central server;
s200, constructing a model of flow classification such as machine learning or semantic understanding in a super fusion structure, analyzing the access condition of a certain node or resource in a period of time, and then predicting the resource access quantity and the resource demand of the node in a future period of time by using the model;
s300, comprehensively representing the method in a central server by utilizing the access condition of resources in a certain node or factors such as the access quantity, the access time length, the access frequency and the like of the resources, and constructing a time sequence prediction model such as LSTM and the like to allocate network resources of each node in a future period;
s400, if the node resource task cannot be met in the resource allocation process, reallocating part of the tasks at the node by constructing a scheduling model, so that part of the tasks on the node are migrated into other network nodes with sufficient resources until the resource allocation on all the nodes meets the resource requirements of respective nodes, and the central server can schedule part of the resources of the node when the resource utilization condition in each node is fully met and migrate the part of the resources to other nodes with sufficient resources, thereby being convenient for users to use.
Preferably, in the step S100, the implementation is performed by installing a resource monitoring program and a data acquisition module on each network node, and the content of data acquisition includes: optimizing code, adding hardware resources, improving network configuration, carrying out load balancing, improving virtualization efficiency and the like, wherein a resource monitoring program periodically collects resource use conditions of network nodes and sends collected information to a central server, and the monitoring program monitors different time periods of a period of time T: each node at t1, t2, t3 … … tn: resource usage within f1, f2, f3 … … fn: xa, xb, xc … … Xn and other information are detected and collected, and then the detected information is transmitted to a central server, so that the collected information can be used for constructing a prediction algorithm model smoothly in the step S200 to predict the resource use condition of each node in a future period of time, and the normal operation of the allocation and scheduling method is ensured.
Preferably, in the step S200, a time sequence prediction engineering model is constructed in the central server through the resource usage situation of each node transmitted in the step S100 to predict the resource usage situation of each node in a future period of time, and in the central server, according to the transmitted time (T1, T2, … …, ty) and the used resources (Xa, xb, … …, xy) in each node, the rate of the resource X used by each node in the corresponding time T is obtained by generating an objective function and constructing a prediction engineering model, and the resource usage requirement of each node in the future period of time is predicted on the basis of the rate of the resource X used by each node in the corresponding time T, so as to ensure that the central server can primarily perform resource allocation processing on each node.
Preferably, in the step S300, the central server builds a feature engineering model according to the method of comprehensively characterizing the access amount, access duration, access frequency and other factors of the resources in each node detected by the monitoring program, analyzes the feature book data of the resource usage in each node by using a genetic algorithm, then allocates the network resources according to the resource usage features of each node by using the model, builds a genetic algorithm feature engineering model in the central server according to the transmitted resource usage situation of each node f, allocates the network resources X to the f1 node, migrates the network resources X to the f1 node by the genetic feature model in the f1 node for screening, the screened network resources XA are normally allocated to the f1 node for use, and the screened rest of the network resources (X-XA) are migrated to the f2 node for screening the XB resources and so on until the network resources X are all allocated to each node fn, and the central server performs allocation treatment on the network resources X.
Preferably, when the monitoring program detects that the network resource requirement on the node cannot be met in the step S400, a part of network resources at the node are distributed to other nodes with sufficient resources by using a scheduling algorithm and the characteristic model, so that the central server can effectively distribute and schedule the network resources to meet the requirement of each network node, and after all the network resources X are distributed, the monitoring program is utilized to analyze the actual distributed resources XN on each network node and the resource usage XN predicted by the original prediction model of the node fn: if Xn is greater than or equal to XN, and the resources at the node fn are fully used; if Xn is smaller than XN and the resource usage of the node fn cannot be satisfied, then part of the resources at the fn can be migrated to the node with sufficient resources detected by other monitoring programs by the central processor, so that the central server can be ensured to reallocate network resources on the node with unsatisfied resources in time to ensure the normal operation of the node, and the node is convenient for users to use.
Preferably, when the monitoring program detects that the network resource requirement on the node cannot be met, the scheduling algorithm may be used to migrate part of the resources at the node fn to other nodes outside the node fn, then the monitoring program detects the resource requirement of other nodes outside the node fn, if the resource task cannot be met in other nodes, the part of the resources Xn-1 in the node fn-1 is redistributed to all nodes except the previous node fn, and the node fn and other nodes outside the previous node fn-1 are detected again, so that the resource allocation on all nodes meets the resource requirement of each node, so that the central server can effectively allocate and schedule the network resources, and meet the requirement of each network node.
The beneficial effects are that:
1. according to the network resource allocation and scheduling method under the super fusion architecture, a characteristic engineering model is built by a central server in the S300 step according to a method for comprehensively representing factors such as the access quantity, the access duration, the access frequency and the like of resources in each node detected by a monitoring program, and characteristic book data of the use of the resources in each node are analyzed by utilizing a genetic algorithm so as to meet the requirement that the central server allocates the network resources according to the resource use characteristics of each node by utilizing the model.
2. According to the network resource allocation and scheduling method under the super fusion architecture, when the monitoring program detects that the network resource requirement on the node cannot be met in the step S400, part of network resources at the node are allocated to other nodes with sufficient resources by using the scheduling algorithm and the characteristic model, so that the central server can effectively allocate and schedule the network resources to meet the requirement of each network node, and the central server can timely reallocate the network resources on the nodes with the unsatisfied resources to ensure normal operation of the network resources, thereby being convenient for users to use.
Drawings
FIG. 1 is a schematic diagram of the overall flow structure of a network resource allocation and scheduling method under a super-converged architecture;
FIG. 2 is a schematic diagram showing the steps of a method S100 and S200 for allocating and scheduling network resources under a super-converged architecture;
FIG. 3 is a schematic diagram illustrating a structure of steps of a method S300 for allocating and scheduling network resources under a super-converged architecture according to the present invention;
fig. 4 is a schematic structural diagram of steps S400 of a network resource allocation and scheduling method under a super-converged architecture according to the present invention;
fig. 5 is a schematic structural diagram of a second embodiment of a network resource allocation and scheduling method under a super-converged architecture according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1-5, a method for allocating and scheduling network resources under a super-converged architecture, the method includes:
s100, detecting and collecting total resource use conditions of network nodes or resource access amount data of each node in a period of time in a super fusion architecture, and then transmitting the detected and collected data to a central server;
s200, constructing a model of flow classification such as machine learning or semantic understanding in a super fusion structure, analyzing the access condition of a certain node or resource in a period of time, and then predicting the resource access quantity and the resource demand of the node in a future period of time by using the model;
s300, comprehensively representing the method in a central server by utilizing the access condition of resources in a certain node or factors such as the access quantity, the access time length, the access frequency and the like of the resources, and constructing a time sequence prediction model such as LSTM and the like to allocate network resources of each node in a future period;
s400, if the node resource task cannot be met in the resource allocation process, reallocating part of the tasks at the node by constructing a scheduling model, so that part of the tasks on the node are migrated into other network nodes with sufficient resources until the resource allocation on all the nodes meets the resource requirements of respective nodes, and the central server can schedule part of the resources of the node when the resource utilization condition in each node is fully met and migrate the part of the resources to other nodes with sufficient resources, thereby being convenient for users to use.
Example two
Referring to fig. 1-5, further, based on the first embodiment, in step S100, the resource monitoring program and the data acquisition module are installed on each network node to implement the data acquisition, where the content of the data acquisition includes: optimizing code, adding hardware resources, improving network configuration, carrying out load balancing, improving virtualization efficiency and the like, wherein a resource monitoring program periodically collects resource use conditions of network nodes and sends collected information to a central server, and the monitoring program monitors different time periods of a period of time T: each node at t1, t2, t3 … … tn: resource usage within f1, f2, f3 … … fn: xa, xb, xc … … Xn and other information are detected and collected, and then the detected information is transmitted to a central server, so that the collected information can be used for constructing a prediction algorithm model smoothly in the step S200 to predict the resource use condition of each node in a future period of time, and the normal operation of the allocation and scheduling method is ensured.
And S200, constructing a time sequence prediction engineering model in the central server through the resource use condition of each node transmitted in the S100 to predict the resource use condition of each node in a future period, and obtaining the rate of the resource X used by each node in the corresponding time T and predicting the resource use requirement of each node in the future period by generating an objective function and constructing the prediction engineering model in the central server according to the transmitted time (T1, T2, … …, ty) in each node and the used resources (Xa, xb, … …, xy) in the central server so as to ensure that the central server can preliminarily perform resource allocation processing on each node.
And S300, constructing a characteristic engineering model by a central server according to a method for comprehensively characterizing factors such as the access quantity, access time length and access frequency of resources in each node detected by a monitoring program, analyzing characteristic book data of resource use in each node by utilizing a genetic algorithm, distributing network resources according to the resource use characteristics of each node by utilizing the model, constructing a genetic algorithm characteristic engineering model in the central server according to the transmitted resource use condition of each node f, distributing the network resources X to the f1 node by the central server, screening the network resources X to the f1 node by the genetic characteristic model in the f1 node, normally distributing the screened network resources XA to the f1 node for use, screening the screened rest of network resources (X-XA) to the f2 node, and so on until the network resources X are all distributed to each node fn, and distributing the network resources X by the central server.
When the monitoring program detects that the network resource requirement on the node cannot be met in the step S400, a part of network resources at the node are distributed to other nodes with sufficient resources by utilizing a scheduling algorithm and the characteristic model, so that the central server can effectively distribute and schedule the network resources to meet the requirement of each network node, and after all the network resources X are distributed, the monitoring program is utilized to analyze the actual distributed resources XN on each network node and the resource use Xn predicted by the original prediction model of the node fn by the central processor: if Xn is greater than or equal to XN, and the resources at the node fn are fully used; if Xn is smaller than XN and the resource usage of the node fn cannot be satisfied, then part of the resources at the fn can be migrated to the node with sufficient resources detected by other monitoring programs by the central processor, so that the central server can be ensured to reallocate network resources on the node with unsatisfied resources in time to ensure the normal operation of the node, and the node is convenient for users to use.
Example III
Referring to fig. 1-5, further, based on the second embodiment, when the monitoring procedure detects that the network resource requirement on the node cannot be met, the scheduling algorithm may be used to migrate part of the resources at the node fn to other nodes outside the node fn, then the monitoring procedure detects the resource requirement of other nodes outside the node fn, if the resource task cannot be met in other nodes, then the part of the resources Xn-1 in the node fn-1 is redistributed to all nodes except the previous node fn, and the detection is performed again on the node fn and other nodes outside the previous node fn-1, so that the resource allocation on all nodes meets the resource requirement of each node, so that the central server can effectively allocate and schedule the network resource to meet the requirement of each network node.
It should be noted that, in the second embodiment, the scheduling of the resources is completed by transferring the node resources which cannot be met by the resources to the nodes with sufficient resources of other nodes, so that the allocation and allocation efficiency of the central server is higher, while in the third embodiment, the node resources which cannot be met by the resources are transferred to all nodes except the node, and the node is excluded in the scheduling range of the central server, then the central server and the monitoring program detect and analyze other nodes except the node until the resources are completely allocated, so that the allocation of the resources of each node by the central server is more accurate, and the two embodiments can be replaced mutually.
Working principle: and S100, the step is realized by installing a resource monitoring program and a data acquisition module on each network node, and the content of data acquisition comprises the following steps: optimizing code, adding hardware resources, improving network configuration, carrying out load balancing, improving virtualization efficiency and the like, wherein a resource monitoring program periodically collects resource use conditions of network nodes and sends collected information to a central server, and the monitoring program monitors different time periods of a period of time T: each node at t1, t2, t3 … … tn: resource usage within f1, f2, f3 … … fn: xa, xb, xc … … Xn and other information are detected and collected, and then the detected information is transmitted to a central server, so that the collected information can be used for constructing a prediction algorithm model smoothly in the step S200 to predict the resource use condition of each node in a future period of time, and the normal operation of the allocation and scheduling method is ensured.
And S200, constructing a time sequence prediction engineering model in the central server through the resource use condition of each node transmitted in the S100 to predict the resource use condition of each node in a future period, and obtaining the rate of the resource X used by each node in the corresponding time T and predicting the resource use requirement of each node in the future period by generating an objective function and constructing the prediction engineering model in the central server according to the transmitted time (T1, T2, … …, ty) in each node and the used resources (Xa, xb, … …, xy) in the central server so as to ensure that the central server can preliminarily perform resource allocation processing on each node.
And S300, constructing a characteristic engineering model by a central server according to a method for comprehensively characterizing factors such as the access quantity, access time length and access frequency of resources in each node detected by a monitoring program, analyzing characteristic book data of resource use in each node by utilizing a genetic algorithm, distributing network resources according to the resource use characteristics of each node by utilizing the model, constructing a genetic algorithm characteristic engineering model in the central server according to the transmitted resource use condition of each node f, distributing the network resources X to the f1 node by the central server, screening the network resources X to the f1 node by the genetic characteristic model in the f1 node, normally distributing the screened network resources XA to the f1 node for use, screening the screened rest of network resources (X-XA) to the f2 node, and so on until the network resources X are all distributed to each node fn, and distributing the network resources X by the central server.
When the monitoring program detects that the network resource requirement on the node cannot be met in the step S400, a part of network resources at the node are distributed to other nodes with sufficient resources by utilizing a scheduling algorithm and the characteristic model, so that the central server can effectively distribute and schedule the network resources to meet the requirement of each network node, and after all the network resources X are distributed, the monitoring program is utilized to analyze the actual distributed resources XN on each network node and the resource use Xn predicted by the original prediction model of the node fn by the central processor: if Xn is greater than or equal to XN, and the resources at the node fn are fully used; if Xn is smaller than XN and the resource usage of the node fn cannot be satisfied, then part of the resources at the fn can be migrated to the node with sufficient resources detected by other monitoring programs by the central processor, so that the central server can be ensured to reallocate network resources on the node with unsatisfied resources in time to ensure the normal operation of the node, and the node is convenient for users to use.
When the monitoring program detects that the network resource requirement on the node cannot be met, the scheduling algorithm can be utilized to transfer part of the resources at the node fn to other nodes outside the node fn one by one, then the monitoring program detects the resource requirement of the other nodes outside the node fn, if the condition that the resource task cannot be met occurs in the other nodes, the part of the resources Xn-1 in the node fn-1 are redistributed to all the nodes except the last node fn, and the node fn and the other nodes except the last node fn-1 are detected again, so that the scheduling algorithm can be used for pushing until the resource allocation on all the nodes meets the resource requirement of each node, and the central server can effectively allocate and schedule the network resources to meet the requirement of each network node.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. A network resource allocation and scheduling method under a super fusion architecture comprises the following steps: the characteristic model and the monitoring program are characterized in that: the method comprises the following steps:
s100, detecting and collecting total resource use conditions of network nodes or resource access amount data of each node in a period of time in a super fusion architecture, and then transmitting the detected and collected data to a central server;
s200, constructing a model of flow classification such as machine learning or semantic understanding in a super fusion structure, analyzing the access condition of a certain node or resource in a period of time, and then predicting the resource access quantity and the resource demand of the node in a future period of time by using the model;
s300, comprehensively representing the method in a central server by utilizing the access condition of resources in a certain node or factors such as the access quantity, the access time length, the access frequency and the like of the resources, and constructing a time sequence prediction model such as LSTM and the like to allocate network resources of each node in a future period;
s400, in the process of allocating resources, if the node resource task cannot be met, reallocating part of tasks at the node by constructing a scheduling model, so that part of tasks on the node are migrated into other network nodes with sufficient resources until the resource allocation on all the nodes meets the resource requirements of the respective nodes.
2. The method for allocating and scheduling network resources under a super fusion architecture according to claim 1, wherein the method is characterized in that: the step S100 is realized by installing a resource monitoring program and a data acquisition module on each network node, and the content of data acquisition comprises the following steps: optimizing codes, adding hardware resources, improving network configuration, carrying out load balancing, improving virtualization efficiency and the like, and periodically collecting resource usage conditions of network nodes by a resource monitoring program and sending the collected information to a central server so as to ensure that a prediction algorithm model can be constructed by smoothly utilizing the collected information in the step S200 to predict the resource usage conditions of each node in a future period of time.
3. The method for allocating and scheduling network resources under a super fusion architecture according to claim 1, wherein the method is characterized in that: and in the step S200, a time sequence prediction engineering model is built in the central server through the resource use condition of each node transmitted in the step S100, and the resource use condition of each node in a future period of time is predicted.
4. The method for allocating and scheduling network resources under a super fusion architecture according to claim 1, wherein the method is characterized in that: and step S300, the central server builds a characteristic engineering model according to the method for comprehensively characterizing the factors such as the access quantity, the access time length, the access frequency and the like of the resources in each node detected by the monitoring program, analyzes the characteristic book data of the resource usage in each node by utilizing a genetic algorithm, and then distributes the network resources according to the resource usage characteristics of each node by utilizing the model.
5. The method for allocating and scheduling network resources under a super fusion architecture according to claim 1, wherein the method is characterized in that: when the monitoring program detects that the network resource requirement on the node cannot be met in the step S400, the part of network resources at the node are distributed to other nodes with sufficient resources by using a scheduling algorithm and the characteristic model, so that the central server can effectively distribute and schedule the network resources to meet the requirement of each network node.
6. The method for allocating and scheduling network resources under a super fusion architecture according to claim 5, wherein the method comprises the following steps: when the monitoring program detects that the network resource requirements on the node cannot be met, the scheduling algorithm can be utilized to transfer part of the resources at the node to other nodes outside the node one by one, then the monitoring program detects the resource requirements of the other nodes outside the node, if the condition that the resource tasks cannot be met occurs in the other nodes, the part of the resources in the node are redistributed to all the nodes except the previous node, and the other nodes except the previous node are detected again, so that the scheduling algorithm can be utilized until the resource allocation on all the nodes meets the resource requirements of the respective nodes, and the central server can effectively allocate and schedule the network resources to meet the requirements of each network node.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311525561.9A CN117240806B (en) | 2023-11-16 | 2023-11-16 | Network resource allocation and scheduling method under super fusion architecture |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311525561.9A CN117240806B (en) | 2023-11-16 | 2023-11-16 | Network resource allocation and scheduling method under super fusion architecture |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117240806A true CN117240806A (en) | 2023-12-15 |
CN117240806B CN117240806B (en) | 2024-02-06 |
Family
ID=89091628
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311525561.9A Active CN117240806B (en) | 2023-11-16 | 2023-11-16 | Network resource allocation and scheduling method under super fusion architecture |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117240806B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190213027A1 (en) * | 2018-01-10 | 2019-07-11 | Vmware, Inc. | Predictive allocation of virtual desktop infrastructure computing resources |
US10489215B1 (en) * | 2016-11-02 | 2019-11-26 | Nutanix, Inc. | Long-range distributed resource planning using workload modeling in hyperconverged computing clusters |
US20200019841A1 (en) * | 2018-07-12 | 2020-01-16 | Vmware, Inc. | Neural network model for predicting usage in a hyper-converged infrastructure |
CN110781125A (en) * | 2019-09-12 | 2020-02-11 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Management method and system for complete cabinet super-fusion server |
CN112000421A (en) * | 2020-07-15 | 2020-11-27 | 北京计算机技术及应用研究所 | Management scheduling technology based on super-fusion architecture |
CN112052072A (en) * | 2020-09-10 | 2020-12-08 | 华云数据控股集团有限公司 | Scheduling strategy and super-fusion system of virtual machine |
US20230123303A1 (en) * | 2021-10-20 | 2023-04-20 | International Business Machines Corporation | Adjusting resources within a hyperconverged infrastructure system based on environmental information |
US20230196182A1 (en) * | 2021-12-21 | 2023-06-22 | International Business Machines Corporation | Database resource management using predictive models |
CN116662010A (en) * | 2023-06-14 | 2023-08-29 | 肇庆学院 | Dynamic resource allocation method and system based on distributed system environment |
US20230342174A1 (en) * | 2022-04-25 | 2023-10-26 | Vmware, Inc. | Intelligent capacity planning for storage in a hyperconverged infrastructure |
-
2023
- 2023-11-16 CN CN202311525561.9A patent/CN117240806B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10489215B1 (en) * | 2016-11-02 | 2019-11-26 | Nutanix, Inc. | Long-range distributed resource planning using workload modeling in hyperconverged computing clusters |
US20190213027A1 (en) * | 2018-01-10 | 2019-07-11 | Vmware, Inc. | Predictive allocation of virtual desktop infrastructure computing resources |
US20200019841A1 (en) * | 2018-07-12 | 2020-01-16 | Vmware, Inc. | Neural network model for predicting usage in a hyper-converged infrastructure |
CN110781125A (en) * | 2019-09-12 | 2020-02-11 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Management method and system for complete cabinet super-fusion server |
CN112000421A (en) * | 2020-07-15 | 2020-11-27 | 北京计算机技术及应用研究所 | Management scheduling technology based on super-fusion architecture |
CN112052072A (en) * | 2020-09-10 | 2020-12-08 | 华云数据控股集团有限公司 | Scheduling strategy and super-fusion system of virtual machine |
US20230123303A1 (en) * | 2021-10-20 | 2023-04-20 | International Business Machines Corporation | Adjusting resources within a hyperconverged infrastructure system based on environmental information |
US20230196182A1 (en) * | 2021-12-21 | 2023-06-22 | International Business Machines Corporation | Database resource management using predictive models |
US20230342174A1 (en) * | 2022-04-25 | 2023-10-26 | Vmware, Inc. | Intelligent capacity planning for storage in a hyperconverged infrastructure |
CN116662010A (en) * | 2023-06-14 | 2023-08-29 | 肇庆学院 | Dynamic resource allocation method and system based on distributed system environment |
Non-Patent Citations (1)
Title |
---|
李廷伟等: "异构多云环境下的算力资源调度技术研究", 《电子技术与软件工程(总第236期)》 * |
Also Published As
Publication number | Publication date |
---|---|
CN117240806B (en) | 2024-02-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Shyam et al. | Virtual resource prediction in cloud environment: a Bayesian approach | |
Selvarani et al. | Improved cost-based algorithm for task scheduling in cloud computing | |
Jung et al. | Synchronous parallel processing of big-data analytics services to optimize performance in federated clouds | |
Hashem et al. | MapReduce scheduling algorithms: a review | |
Patel et al. | Survey of load balancing techniques for grid | |
CN102667724A (en) | Goal oriented performance management of workload utilizing accelerators | |
CN103701886A (en) | Hierarchic scheduling method for service and resources in cloud computation environment | |
Ding et al. | Adaptive virtual machine consolidation framework based on performance-to-power ratio in cloud data centers | |
Jiang et al. | Characterizing co-located workloads in alibaba cloud datacenters | |
Ali et al. | Optimizing inference serving on serverless platforms | |
CN112181620A (en) | Big data workflow scheduling method for sensing service capability of virtual machine in cloud environment | |
Zhang et al. | A spark scheduling strategy for heterogeneous cluster | |
Fareghzadeh et al. | Toward holistic performance management in clouds: taxonomy, challenges and opportunities | |
Ullah et al. | LSTPD: least slack time-based preemptive deadline constraint scheduler for Hadoop clusters | |
KR20210041295A (en) | Virtualized resource distribution system in cloud computing environment | |
Singh et al. | Combining malleability and i/o control mechanisms to enhance the execution of multiple applications | |
CN117240806B (en) | Network resource allocation and scheduling method under super fusion architecture | |
Piao et al. | Computing resource prediction for mapreduce applications using decision tree | |
Ray et al. | Is high performance computing (HPC) ready to handle big data? | |
Zacheilas et al. | A Pareto-based scheduler for exploring cost-performance trade-offs for MapReduce workloads | |
Talia et al. | The grid backfilling: a multi-site scheduling architecture with data mining prediction techniques | |
US20230161620A1 (en) | Pull mode and push mode combined resource management and job scheduling method and system, and medium | |
Patel et al. | Analysis of workloads for cloud infrastructure capacity planning | |
Fernández-Cerero et al. | Quality of cloud services determined by the dynamic management of scheduling models for complex heterogeneous workloads | |
Ramachandra et al. | Task Clustering and Scheduling in Fault Tolerant Cloud Using Dense Neural Network |
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 |