CN114710489A - Distributed cloud resource scheduling method and device and distributed cloud - Google Patents
Distributed cloud resource scheduling method and device and distributed cloud Download PDFInfo
- Publication number
- CN114710489A CN114710489A CN202210329988.0A CN202210329988A CN114710489A CN 114710489 A CN114710489 A CN 114710489A CN 202210329988 A CN202210329988 A CN 202210329988A CN 114710489 A CN114710489 A CN 114710489A
- Authority
- CN
- China
- Prior art keywords
- cloud
- center cluster
- designated
- cloud center
- resource
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000015654 memory Effects 0.000 claims description 50
- 238000012216 screening Methods 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multi Processors (AREA)
Abstract
The invention discloses a distributed cloud resource scheduling method and device, wherein a distributed cloud acquires the state of each cloud center cluster when a designated resource needs to run across clouds, and then determines the designated cloud center cluster according to the designated resource, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the designated resource to the designated cloud center cluster, and the designated resource is run by a node of the designated cloud center cluster. According to the method and the system, the designated cloud center cluster is determined according to the state of each cloud center cluster for the resource needing cross-cloud operation, so that the resource is scheduled to the designated cloud center cluster, and cross-cloud scheduling of the distributed cloud universe resource is realized.
Description
Technical Field
The invention relates to the technical field of information, in particular to a distributed cloud resource scheduling method and device. The invention also relates to a distributed cloud.
Background
With the development of cloud computing services, the form of the cloud supports private clouds, public clouds, edge clouds, and the like. The public cloud is a cloud computing service which is managed by each cloud manufacturer, and the cloud manufacturer provides cloud service for users and collects service cost; the private cloud is a cloud computing service which is sold to users by cloud manufacturers and the users use the cloud operation and maintenance service to operate and maintain themselves; the edge cloud is a cloud computing service which is used by cloud manufacturers to operate cloud products in factories and other places which are far away from a cloud data center and have unstable networks and supports high network delay.
The distributed cloud operating system needs to uniformly manage existing public clouds, private clouds and edge clouds, the cloud centers may be distributed in different areas or different data centers, the resources supported by each cloud center are different, some cloud centers support container resources, and some cloud centers support virtual machine resources. Therefore, how to implement scheduling of resources in a distributed cloud and implement cross-cloud management and universe management of resources becomes a technical problem to be solved.
Disclosure of Invention
The invention aims to provide a distributed cloud resource scheduling method and device, which can realize cross-cloud scheduling of distributed cloud global resources. The invention also provides a distributed cloud.
In order to achieve the purpose, the invention provides the following technical scheme:
a distributed cloud resource scheduling method comprises the following steps:
when the designated resources need to run across the clouds, acquiring the state of each cloud center cluster;
and determining an appointed cloud center cluster according to the appointed resources, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the appointed resources to the appointed cloud center cluster, and operating the appointed resources by the nodes of the appointed cloud center cluster.
Preferably, the obtaining the state of each cloud center cluster includes: acquiring the CPU condition and/or the memory condition of each cloud center cluster;
the preset scheduling policy comprises: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories required by the designated resources and the CPU condition and/or the memory condition of each cloud center cluster.
Preferably, the designated resource comprises an application unit, and the application unit comprises a plurality of resources;
scheduling the specified resource to the specified cloud-centric cluster comprises: and scheduling a plurality of resources included by the application unit to the same specified cloud center cluster.
Preferably, the obtaining the state of each cloud center cluster includes: acquiring the CPU condition and/or the memory condition of each cloud center cluster;
the preset scheduling policy comprises: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories required by a plurality of resources and the CPU condition and/or the memory condition of each cloud center cluster, wherein the number of CPUs and/or the memory condition is included in the application unit.
Preferably, the scheduling the specified resource to the specified cloud center cluster, and the operating the specified resource by the node of the specified cloud center cluster includes:
selecting any group from the designated cloud center cluster, and determining a designated node from each node of the selected group so as to run the designated resource by the designated node, wherein the node of the designated cloud center cluster is divided into a plurality of groups.
Preferably, the scheduling the specified resource to the specified cloud center cluster, and the operating the specified resource by the node of the specified cloud center cluster includes:
screening out a plurality of nodes meeting the requirements from the nodes of the designated cloud center cluster according to the labels of the nodes of the designated cloud center cluster;
and determining the designated node from the screened nodes meeting the requirements according to the CPU condition and/or the memory condition and/or the external memory condition of the node.
Preferably, the step of screening out a plurality of nodes meeting the requirement from each node of the designated cloud center cluster according to the label of each node of the designated cloud center cluster comprises: and reserving affinity nodes according to the labels of all nodes of the appointed cloud center cluster, and/or determining whether to reserve taint nodes according to whether the appointed resources tolerate the taint nodes.
Preferably, the method further comprises the following steps: and acquiring the state of each cloud center cluster, and scheduling the resources running on the first cloud center cluster to other cloud center clusters if the state of the first cloud center cluster is abnormal.
A distributed cloud resource scheduling apparatus, comprising:
the acquisition module is used for acquiring the state of each cloud center cluster when the specified resource needs to run across the cloud;
and the scheduling module is used for determining the designated cloud center cluster according to the designated resources, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the designated resources to the designated cloud center cluster, and the designated resources are operated by the nodes of the designated cloud center cluster.
A distributed cloud is applied to the distributed cloud resource scheduling method or comprises the distributed cloud resource scheduling device.
According to the technical scheme, when the designated resource needs to run across clouds, the state of each cloud center cluster is obtained, the designated cloud center cluster is determined according to the designated resource, the state of each cloud center cluster and a preset scheduling strategy, the designated resource is scheduled to the designated cloud center cluster, and the designated resource is run by the node of the designated cloud center cluster. According to the distributed cloud resource scheduling method and device, the designated cloud center cluster is determined according to the state of each cloud center cluster for the resource needing cross-cloud operation, so that the resource is scheduled to the designated cloud center cluster, and cross-cloud scheduling of the universe resource is achieved.
The invention further provides a distributed cloud which can achieve the beneficial effects.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a distributed cloud resource scheduling method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for scheduling a specified resource to a specified cloud-centric cluster for operation of the specified resource by a node of the specified cloud-centric cluster in an embodiment of the present invention;
fig. 3 is a schematic diagram of a distributed cloud resource scheduling apparatus according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating resource scheduling performed by a distributed cloud resource scheduling apparatus according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a distributed cloud according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart of a distributed cloud resource scheduling method provided in this embodiment, and as shown in the figure, the distributed cloud resource scheduling method includes the following steps:
s11: and when the specified resources need to run across the clouds, acquiring the state of each cloud center cluster.
The specified resource belongs to a resource of the distributed cloud. Specifying that resources need to run across clouds refers to specifying that resources need to run on other cloud-centric clusters. The distributed cloud comprises a plurality of cloud center clusters, and the state of the cloud center clusters refers to the operation condition of the cloud center clusters.
S12: and determining an appointed cloud center cluster according to the appointed resources, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the appointed resources to the appointed cloud center cluster, and operating the appointed resources by the nodes of the appointed cloud center cluster.
The preset scheduling strategy describes a method for determining the designated cloud center cluster according to the designated resources and the state of each cloud center cluster.
According to the distributed cloud resource scheduling method, for the resources needing cross-cloud operation, the states of all the cloud center clusters are obtained, the designated cloud center cluster is determined according to the states of all the cloud center clusters, the resources are scheduled to the designated cloud center cluster, and cross-cloud scheduling of the distributed cloud global resources is achieved.
In this embodiment, the manner of obtaining the state of the cloud center cluster is not limited, and obtaining the state of the cloud center cluster may be achieved, for example, an instruction may be issued to the cloud center cluster, and the cloud center cluster returns data reflecting its own state according to the instruction.
In this embodiment, specific contents of the state of the cloud center cluster are not limited, and may be set according to a method for scheduling a specified resource. Optionally, the state of the cloud center cluster may include a CPU condition and/or a memory condition of the cloud center cluster, and accordingly, the obtaining the state of each cloud center cluster may include: and acquiring the CPU condition and/or the memory condition of each cloud center cluster. The corresponding preset scheduling policy may include: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories required by the designated resources and the CPU condition and/or the memory condition of each cloud center cluster.
Optionally, in practical application, the total amount of CPUs, the number of allocated CPUs, or the total amount of memories, and the number of allocated memories of each cloud center cluster may be obtained, and the number of remaining CPUs or the number of remaining memories of each cloud center cluster may be calculated according to the total amount of CPUs and/or the number of memories required by the specified resources, and the number of remaining CPUs or the number of remaining memories of each cloud center cluster, so that the specified cloud center cluster is determined.
Optionally, the preset scheduling policy may specifically include: and respectively calculating the scores of the cloud center clusters according to the CPU condition and/or the memory condition of each cloud center cluster, sequencing the cloud center clusters according to the scores of the cloud center clusters, and determining the designated cloud center cluster. For example, the cloud center cluster with the highest score may be used as the designated cloud center cluster, or the designated cloud center cluster may be determined according to the ranking result of the scores of the cloud center clusters and other methods, so that the determined designated cloud center cluster can meet the application requirement of the designated resource. When the score of the cloud center cluster is calculated according to the CPU condition and/or the memory condition of the cloud center cluster, weights can be respectively configured for the CPU condition and the memory condition according to the requirement of the specified resource for calculation, so that the determined specified cloud center cluster can meet the application requirement of the specified resource. In addition, the CPU condition and/or the memory condition of the cloud center cluster refer to a CPU sum condition and/or a memory sum condition of nodes included in the cloud center cluster.
In this embodiment, the form of the designated resource is not limited, and the selectable designated resource may be a container resource, or the designated resource may be a virtual machine resource.
In a preferred embodiment, the specified resource may comprise an application unit comprising a plurality of resources. Correspondingly, after the designated cloud center cluster is determined, scheduling the designated resource to the designated cloud center cluster may include: and scheduling a plurality of resources included by the application unit to the same specified cloud center cluster. In this embodiment, the plurality of resources are collectively scheduled as one unit, and compared with the case where the plurality of resources are separately scheduled, the number of scheduling times is reduced, and the scheduling efficiency can be improved.
Correspondingly, optionally, the preset scheduling policy may include: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories needed by a plurality of resources and the CPU condition and/or the memory condition of each cloud center cluster included in the application unit. Under the condition that a plurality of resources are taken as one unit for scheduling, the appointed cloud center cluster is determined according to the number of CPUs and/or the number of memories needed by the plurality of resources included in the application unit, and each resource of the application unit can be guaranteed to run on the appointed cloud center cluster.
The number of CPUs required by an application unit is the sum of the number of CPUs required by a plurality of resources included in the application unit, and can be calculated according to the following formula:
APPCPU＝∑Ci；
wherein, APPCPUIndicating the number of CPUs required by the application unit, CiThe number of CPUs required for the ith resource included in the application unit is represented, and the application unit includes size resources.
The amount of memory required by the application unit is the sum of the amounts of memory required by a plurality of resources included in the application unit, and can be calculated according to the following formula:
APPmemory device＝∑Mi；
Wherein, APPMemory deviceRepresenting the amount of memory required by the application unit, MiThe memory quantity required by the ith resource included by the application unit is shown, and the application unit includes size resources.
In this embodiment, the time complexity of the application unit performing the one-time scheduling may be represented as o (1/the number of resources included in the application unit), that is, the more the application unit includes resources, the less the time required for performing the one-time scheduling is, the smaller the scheduling delay is, and the higher the scheduling performance is.
In practical applications, the application unit may include a resource list, a resource request upper limit, and the like. The resource request refers to the lower limit of the number of CPUs and/or the number of memories required by the resource, and the upper limit of the resource request refers to the upper limit of the number of CPUs and/or the number of memories required by the resource. In this embodiment, the number of resources included in the application unit is not limited, and may be set according to an application service in actual application. Illustratively, an application unit may comprise a plurality of container resources, or an application unit may comprise a plurality of virtual machine resources.
In the prior art, distributed cloud scheduling resources are assigned one by one, that is, after one resource is scheduled, the next resource can be scheduled, the scheduling times are equal to the number of resources, the scheduling time complexity is o (m), and m represents the number of resources to be scheduled, that is, the more the resources to be scheduled, the longer the resource scheduling time is. Aiming at the scale of tens of millions of nodes in a distributed cloud, resources to be scheduled are in the millions, the scheduling of the millions of resources is completed, the time consumption is about 30 minutes, and the scheduling performance cannot meet the requirement of global resource scheduling. In the method, the plurality of resources are scheduled as one unit, so that the scheduling times are reduced, the scheduling efficiency can be improved, the extremely flexible scheduling capability of the universe of large-scale cluster resources can be supported, and the problem of cloud center cluster resource waste caused by invalid occupation of the cloud center cluster resources can be prevented.
Preferably, when the designated cloud center cluster is determined, the designated resource is scheduled to the designated cloud center cluster, and the running of the designated resource by the node of the designated cloud center cluster may include: and selecting any group from the designated cloud center cluster, and determining a designated node from each node of the selected group so as to run the designated resource by the designated node, wherein the nodes of the designated cloud center cluster are divided into a plurality of groups. In this embodiment, a manner of dividing the nodes of the designated cloud center cluster into a plurality of groups is not limited, and the number of groups into which the nodes of the designated cloud center cluster are divided and the number of nodes included in each group are not limited.
Optionally, each node of the cloud center cluster may be sequentially divided into a plurality of groups, one group may be selected from the groups when one scheduling is performed, and another group may be selected from the groups when the next scheduling is performed. For example, the node group selected by one-time scheduling can be represented as: [ (Index)i+1)，(NodeSize/Shard+Indexi)]And wherein NodeSize represents the number of nodes included in the specified cloud center cluster, and Shard represents the number of the nodes of the specified cloud center cluster divided into groups. i initial value 0, IndexiThe initial value is 0. First scheduling selected node grouping listShown as [1, (NodeSize/Shard)]And adding 1 to the maximum subscript of the node during each scheduling to serve as the subscript of the starting node for the next scheduling.
In the method, the nodes of the cloud center cluster are grouped, and the nodes of the operating resources are determined from the node groups, so that the time complexity of primary resource scheduling is represented as o (n/s), n represents the number of the nodes of the cloud center cluster, s represents the number of the nodes of the cloud center cluster divided into the groups, and the larger s is, the smaller the time complexity is, the smaller the primary resource scheduling delay is, and the higher the primary resource scheduling performance is. In the prior art, when resource scheduling is performed, all nodes need to be traversed to schedule each resource, a designated node is selected according to the conditions of all nodes, the number of nodes of a cloud center cluster may be thousands of nodes, the time consumption is too long due to too many nodes in the process, the time complexity is o (n), n represents the number of nodes of the cloud center cluster, that is, the number of nodes is more, the scheduling time consumption is longer, for example, in a cluster with 10000 node scales, about 10 minutes is required for completing one scheduling, and the application requirements cannot be met. Compared with the method, the time consumption of scheduling can be reduced, and the scheduling efficiency is improved.
Further preferably, the scheduling the specified resource to the specified cloud center cluster, and the running the specified resource by the node of the specified cloud center cluster may include the following processes, please refer to fig. 2, where fig. 2 is a flowchart of a method for scheduling the specified resource to the specified cloud center cluster and running the specified resource by the node of the specified cloud center cluster in this embodiment, and includes the following steps:
s121: and screening a plurality of nodes meeting the requirements from the nodes of the appointed cloud center cluster according to the label of each node of the appointed cloud center cluster.
The label of the node reflects the performance or the attribute of the node, and the specific content of the label of the node is not limited in this embodiment, for example, the node may be a computing node, the node is a storage node, or a specific problem exists in the node. And screening out a plurality of nodes meeting the requirements according to the requirements of the specified resources and the labels of the nodes of the specified cloud center cluster. The filtering process of the nodes is carried out in the step, and the efficiency of determining the designated nodes from the cloud center cluster is improved.
Optionally, affinity nodes may be reserved according to labels of each node of the designated cloud center cluster, and/or whether a taint node is reserved may be determined according to whether the designated resource tolerates the taint node. An affinity node may be understood as a node that is considered satisfactory based on its label. For example, if the designated resource corresponds to a storage service and a certain node is a storage node, the node is to be reserved for the designated resource as an affinity node. A taint node may be understood as a node whose one or more properties or attributes are deemed not to be in accordance with the needs of a given resource based on the node's label. If the designated resources do not tolerate the taint node, directly removing the taint node; if the designated resources allow tolerance for the taint node, the taint node can be reserved for subsequent screening processes, so that more nodes can be selected in the subsequent screening processes.
S122: and determining the designated node from the screened nodes meeting the requirements according to the CPU condition and/or the memory condition and/or the external memory condition of the node.
In this embodiment, the specific method for determining the designated node according to the CPU condition and/or the memory condition and/or the external memory condition of the node is not limited. Optionally, the scores of the nodes may be respectively calculated according to the screened CPU condition and/or memory condition and/or external memory condition of each node that meets the requirement, and the nodes are sorted according to the scores of the nodes, so as to determine the designated node. For example, the node with the highest score may be designated as the designated node. When the score of the node is calculated according to the CPU condition and/or the memory condition and/or the external memory condition of the node, the CPU condition, the memory condition and the external memory condition can be respectively configured with weights according to the requirement of the specified resource for calculation, so that the determined specified node can meet the requirement of the specified resource.
Further preferably, the distributed cloud resource scheduling method of this embodiment may further include: and acquiring the state of each cloud center cluster, and scheduling the resources running on the first cloud center cluster to other cloud center clusters if the state of the first cloud center cluster is abnormal. The states of the cloud center clusters can comprise the running states of the cloud center clusters and the running states of loads on the cloud center clusters, and by monitoring the running states of the cloud center clusters and the running states of the loads on the cloud center clusters, when one cloud center cluster is abnormal, resources of the cloud center cluster can be scheduled to other normal cloud center clusters, so that normal running of application services can be guaranteed.
Accordingly, referring to fig. 3, fig. 3 is a schematic diagram of a distributed cloud resource scheduling device provided in this embodiment, as shown in the figure, the distributed cloud resource scheduling device includes:
the acquisition module 21 is configured to acquire a state of each cloud center cluster when a specified resource needs to run across clouds;
and the scheduling module 22 is configured to determine an appointed cloud center cluster according to the appointed resource, the state of each cloud center cluster and a preset scheduling policy, so as to schedule the appointed resource to the appointed cloud center cluster, and allow a node of the appointed cloud center cluster to operate the appointed resource.
The distributed cloud resource scheduling device of this embodiment obtains the state of each cloud center cluster for a resource that needs to run across clouds to determine an appointed cloud center cluster according to the state of each cloud center cluster, so as to schedule the resource to the appointed cloud center cluster, and can implement the cross-cloud scheduling of the distributed cloud global resource.
In this embodiment, the obtaining module 21 obtains the state of each cloud center cluster, and the scheduling module 22 determines, according to the specified resource, the state of each cloud center cluster, and the preset scheduling policy, an implementation manner of the specified cloud center cluster to schedule the specified resource to the specified cloud center cluster, which may refer to the implementation manner described in the above embodiment of the distributed cloud resource scheduling method and will not be described herein again.
Fig. 4 may exemplarily refer to fig. 4, and fig. 4 is a schematic diagram of resource scheduling performed by a distributed cloud resource scheduling apparatus according to an embodiment. As shown in the figure, what needs to be scheduled across the cloud may be a container resource, a virtual machine resource, or an application unit, and the application unit may be a unit formed by a plurality of container resources. Dispatching container resources or virtual machine resources to corresponding cloud center clusters; and scheduling the container resources included by the application units to the same cloud center cluster.
The scheduling device 20 may specify, according to the condition of the resource to be scheduled, that the current resource operates in a specified cloud center cluster in a container load manner, and support the operation in a container manner; the current resources can also be appointed to run in the appointed cloud center cluster in a virtual machine load mode, and the virtual machine running mode is supported.
The embodiment also provides a distributed cloud, which applies the above-mentioned distributed cloud resource scheduling method, or comprises the above-mentioned distributed cloud resource scheduling device.
In the distributed cloud of this embodiment, when the designated resource needs to run across clouds, the state of each cloud center cluster is obtained, and then the designated cloud center cluster is determined according to the designated resource, the state of each cloud center cluster and a preset scheduling policy, so that the designated resource is scheduled to the designated cloud center cluster, and the designated resource is run by the node of the designated cloud center cluster. According to the distributed cloud, the appointed cloud center cluster is determined according to the state of each cloud center cluster for the resource needing cross-cloud operation, so that the resource is scheduled to the appointed cloud center cluster, and cross-cloud scheduling of the universe resource is achieved.
Referring to fig. 5, fig. 5 is a schematic diagram of a distributed cloud according to an embodiment, where the distributed cloud may include a public cloud, a private cloud, an edge cloud, a central cloud, or a local cloud. The scheduling device 20 may implement cross-cloud scheduling of distributed cloud universe resources.
The distributed cloud resource scheduling method and device provided by the invention and the distributed cloud are introduced in detail above. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (10)
1. A distributed cloud resource scheduling method is characterized by comprising the following steps:
when the designated resources need to run across the clouds, acquiring the state of each cloud center cluster;
and determining an appointed cloud center cluster according to the appointed resources, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the appointed resources to the appointed cloud center cluster, and operating the appointed resources by the nodes of the appointed cloud center cluster.
2. The distributed cloud resource scheduling method of claim 1, wherein obtaining the status of each cloud center cluster comprises: acquiring the CPU condition and/or the memory condition of each cloud center cluster;
the preset scheduling policy comprises: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories required by the designated resources and the CPU condition and/or the memory condition of each cloud center cluster.
3. The distributed cloud resource scheduling method of claim 1 wherein the specified resource comprises an application unit, the application unit comprising a plurality of resources;
scheduling the specified resource to the specified cloud-centric cluster comprises: and scheduling a plurality of resources included by the application unit to the same specified cloud center cluster.
4. The distributed cloud resource scheduling method of claim 3, wherein obtaining the state of each cloud center cluster comprises: acquiring the CPU condition and/or the memory condition of each cloud center cluster;
the preset scheduling policy comprises: and determining the designated cloud center cluster according to the number of CPUs and/or the number of memories needed by a plurality of resources and the CPU condition and/or the memory condition of each cloud center cluster included in the application unit.
5. The distributed cloud resource scheduling method of claim 1, wherein scheduling the specified resource to the specified cloud-centric cluster, the running of the specified resource by a node of the specified cloud-centric cluster comprises:
selecting any group from the designated cloud center cluster, and determining a designated node from each node of the selected group so as to run the designated resource by the designated node, wherein the node of the designated cloud center cluster is divided into a plurality of groups.
6. The distributed cloud resource scheduling method of claim 1, wherein scheduling the specified resource to the specified cloud-centric cluster, the running of the specified resource by a node of the specified cloud-centric cluster comprises:
screening out a plurality of nodes meeting the requirements from the nodes of the designated cloud center cluster according to the labels of the nodes of the designated cloud center cluster;
and determining the designated node from the screened nodes meeting the requirements according to the CPU condition and/or the memory condition and/or the external memory condition of the node.
7. The distributed cloud resource scheduling method of claim 6, wherein the screening of the nodes of the designated cloud center cluster from the nodes of the designated cloud center cluster according to the label of the nodes of the designated cloud center cluster comprises: and reserving affinity nodes according to the labels of all nodes of the appointed cloud center cluster, and/or determining whether to reserve taint nodes according to whether the appointed resources tolerate the taint nodes.
8. The distributed cloud resource scheduling method of any of claims 1-7, further comprising: and acquiring the state of each cloud center cluster, and scheduling the resources running on the first cloud center cluster to other cloud center clusters if the state of the first cloud center cluster is abnormal.
9. A distributed cloud resource scheduling device, comprising:
the acquisition module is used for acquiring the state of each cloud center cluster when the specified resource needs to run across the cloud;
and the scheduling module is used for determining the designated cloud center cluster according to the designated resources, the state of each cloud center cluster and a preset scheduling strategy so as to schedule the designated resources to the designated cloud center cluster, and the designated resources are operated by the nodes of the designated cloud center cluster.
10. A distributed cloud, characterized in that the distributed cloud resource scheduling method of any one of claims 1 to 8 is applied, or comprises the distributed cloud resource scheduling apparatus of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210329988.0A CN114710489A (en) | 2022-03-31 | 2022-03-31 | Distributed cloud resource scheduling method and device and distributed cloud |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210329988.0A CN114710489A (en) | 2022-03-31 | 2022-03-31 | Distributed cloud resource scheduling method and device and distributed cloud |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114710489A true CN114710489A (en) | 2022-07-05 |
Family
ID=82170833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210329988.0A Pending CN114710489A (en) | 2022-03-31 | 2022-03-31 | Distributed cloud resource scheduling method and device and distributed cloud |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114710489A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108400898A (en) * | 2018-05-30 | 2018-08-14 | 郑州云海信息技术有限公司 | The management method and device of resource in cloud data management platform |
| CN109451056A (en) * | 2018-12-20 | 2019-03-08 | 中国软件与技术服务股份有限公司 | Server dynamic allocation method and system between more clusters |
| CN112350855A (en) * | 2020-10-26 | 2021-02-09 | 浪潮云信息技术股份公司 | Configuration-based cloud center management method |
| US20210149737A1 (en) * | 2019-11-14 | 2021-05-20 | Korea Electronics Technology Institute | Method for fast scheduling for balanced resource allocation in distributed and collaborative container platform environment |
| CN112882790A (en) * | 2020-12-31 | 2021-06-01 | 华数云科技有限公司 | Cloud edge cooperative management method based on distributed cloud platform |
| KR20220006490A (en) * | 2021-12-29 | 2022-01-17 | 케이웨어 (주) | Hybrid cloud resource allocation method for workload dynamic resource placement and optimization performance management |
-
2022
- 2022-03-31 CN CN202210329988.0A patent/CN114710489A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108400898A (en) * | 2018-05-30 | 2018-08-14 | 郑州云海信息技术有限公司 | The management method and device of resource in cloud data management platform |
| CN109451056A (en) * | 2018-12-20 | 2019-03-08 | 中国软件与技术服务股份有限公司 | Server dynamic allocation method and system between more clusters |
| US20210149737A1 (en) * | 2019-11-14 | 2021-05-20 | Korea Electronics Technology Institute | Method for fast scheduling for balanced resource allocation in distributed and collaborative container platform environment |
| CN112350855A (en) * | 2020-10-26 | 2021-02-09 | 浪潮云信息技术股份公司 | Configuration-based cloud center management method |
| CN112882790A (en) * | 2020-12-31 | 2021-06-01 | 华数云科技有限公司 | Cloud edge cooperative management method based on distributed cloud platform |
| KR20220006490A (en) * | 2021-12-29 | 2022-01-17 | 케이웨어 (주) | Hybrid cloud resource allocation method for workload dynamic resource placement and optimization performance management |
Non-Patent Citations (1)
| Title |
|---|
| 彭丽苹;吕晓丹;蒋朝惠;彭成辉;: "基于Docker的云资源弹性调度策略", 计算机应用, no. 02 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111966500B (en) | Resource scheduling method and device, electronic equipment and storage medium | |
| US7856572B2 (en) | Information processing device, program thereof, modular type system operation management system, and component selection method | |
| US20050132379A1 (en) | Method, system and software for allocating information handling system resources in response to high availability cluster fail-over events | |
| CN110221920B (en) | Deployment method, device, storage medium and system | |
| CN114356587B (en) | Calculation power task cross-region scheduling method, system and equipment | |
| CN113946431B (en) | Resource scheduling method, system, medium and computing device | |
| CN112685153A (en) | Micro-service scheduling method and device and electronic equipment | |
| CN109525410A (en) | The method, apparatus and distributed memory system of distributed memory system updating and management | |
| CN116541134B (en) | Method and device for deploying containers in multi-architecture cluster | |
| CN110162396A (en) | Method for recovering internal storage, device, system and storage medium | |
| CN112395269B (en) | MySQL high availability group building method and device | |
| CN111930493A (en) | NodeManager state management method and device in cluster and computing equipment | |
| CN112099917B (en) | Regulation and control system containerized application operation management method, system, equipment and medium | |
| CN114844791B (en) | Cloud service automatic management and distribution method and system based on big data and storage medium | |
| CN113674131A (en) | Hardware accelerator equipment management method and device, electronic equipment and storage medium | |
| CN111796933A (en) | Resource scheduling method, device, storage medium and electronic equipment | |
| CN115658311A (en) | Resource scheduling method, device, equipment and medium | |
| CN107203256B (en) | Energy-saving distribution method and device under network function virtualization scene | |
| CN112650449B (en) | Method and system for releasing cache space, electronic device and storage medium | |
| CN112130927A (en) | Reliability-enhanced mobile edge computing task unloading method | |
| CN108681578B (en) | Business data storage method, device, server and storage medium | |
| CN114710489A (en) | Distributed cloud resource scheduling method and device and distributed cloud | |
| CN109408230A (en) | Docker container dispositions method and system based on energy optimization | |
| CN109558214B (en) | Host machine resource management method and device in heterogeneous environment and storage medium | |
| CN112612579A (en) | Virtual machine deployment method, storage medium, and computer device |
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 |