CN115801782A - Cross-regional data center cloud resource scheduling system and method - Google Patents
Cross-regional data center cloud resource scheduling system and method Download PDFInfo
- Publication number
- CN115801782A CN115801782A CN202310043278.6A CN202310043278A CN115801782A CN 115801782 A CN115801782 A CN 115801782A CN 202310043278 A CN202310043278 A CN 202310043278A CN 115801782 A CN115801782 A CN 115801782A
- Authority
- CN
- China
- Prior art keywords
- data center
- scheduling
- cross
- regional
- local area
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a cross-regional data center cloud resource scheduling system and a cross-regional data center cloud resource scheduling method, wherein the system comprises a plurality of data centers, and each data center comprises a computing node and a management node; wherein the compute node comprises a virtual switch; wherein the management node comprises an SDN controller; the logic router sends the routing information of the data centers in other areas to the SDN controller through the routing announcement, converts the routing information of the data centers in other areas into an OpenFlow flow table through the SDN controller and sends the OpenFlow flow table to the virtual switch in the area; and the transmission logic switch is respectively connected with the SDN controller and the logic router, forwards the SDN network information of the data center of the area outwards through the logic router, and forwards the information transmitted by other data centers to the SDN network of the data center of the area. The method can realize stable communication between the multi-cloud cross-regional isolation networks of the overlay layer and smooth migration of network security policies between the multi-cloud cross-regional data centers.
Description
Technical Field
The disclosure belongs to the technical field of communication, and particularly relates to a cross-regional data center cloud resource scheduling system and method.
Background
Cloud computing, as a new computing mode, has rapidly developed into a new research and application field, and has gained more and more attention and wide application, and under the wave of digital transformation, all traditional business systems are migrating to the cloud. The high concealment, the high mobility, the high confidentiality and the high security of the application determine that a cloud data center needs to adopt a multi-cloud and cross-region based distributed deployment scheme.
However, the traditional cloud data center resource migration schemes are all established on the basis of the traditional network, the resource scheduling is realized through a service network, and in the practical application development process, service systems under different data centers have different requirements on resources in the aspects of network bandwidth, data scale, CPU efficiency, IO reading and the like, so that the problem that the service resource requirements and the data center resource capacity need to be matched exists in the cross-region multi-data center; secondly, high and frequent service pressure exists in part of data center resources, and the resource load is unbalanced due to the fact that part of the data center resources are idle.
Disclosure of Invention
The present disclosure is made based on the above needs in the prior art, and a technical problem to be solved by the present disclosure is to provide a cross-regional data center cloud resource scheduling system, so as to improve the resource utilization rate and the flexibility of use of multiple clouds and cross-regions.
In order to solve the above problem, the technical solution provided by the present disclosure includes:
a cross-region data center cloud resource scheduling system comprises a plurality of data centers, wherein each data center comprises a computing node and a management node; the computing node comprises a virtual switch and is responsible for forwarding the flow of other virtual switches of the local area data center according to an OpenFlow flow table issued by the local area data center management node; the management node comprises an SDN controller and is connected with the virtual switch network of the local area data center to form an SDN network of the local area data center; the SDN controller issues an OpenFlow flow table to the virtual switch of the local area data center to guide the virtual switch to forward flow; the logic router sends the routing information of the data centers in other areas to the SDN controller through routing announcements, converts the routing information of the data centers in other areas into OpenFlow flow tables through the SDN controller and sends the OpenFlow flow tables to the virtual switch in the area; and the transmission logic switch is respectively connected with the SDN controller and the logic router, forwards the SDN network information of the data center of the area outwards through the logic router, and forwards the information transmitted by other data centers to the SDN network of the data center of the area.
According to the system, stable communication between multi-cloud cross-region isolation networks in an overlay layer and smooth migration of network security policies between multi-cloud cross-region data centers can be achieved through the virtual switch, the SDN controller, the logic router established on the SDN controller and the logic switches established on the virtual switch and the SDN controller.
Preferably, the management node further includes an interconnection module, where the interconnection module is located in the local area data center and the logic switch, and is connected to interconnection modules of other area data centers; and the interactive communication connection between the local area data center virtual switch and the other area data center virtual switches is realized. The interconnection module is responsible for the physical network intercommunication among the cross-region multiple clouds of the multicast layer and provides the communication support of the physical layer for the virtual flow intercommunication.
Preferably, the management node further includes a distribution scheduling manager, which sends the resource scheduling instruction of the local area to the computing node of the local area data center according to the resource information from the local area data center and the scheduling policy of the local area data center; the distribution scheduling manager is also connected with the management nodes of other areas and sends and receives cross-area scheduling instructions to and from the distribution scheduling managers of the data center management nodes of other areas. The dispatching of the resource information of the local area data center and the cross-area data center and the distribution function of the resource information are realized through the distribution dispatching manager.
Preferably, the computing node further includes a resource collection agent module, which is connected to the distribution scheduling manager of the management node of the regional data center, and provides the resource information of the computing node where the resource collection agent module is located to the distribution scheduling manager. The acquisition function of the resource information is realized, and the support service is provided for the management node to acquire the resource information.
Preferably, the data center includes a management node, and the management node manages the plurality of computing nodes. Therefore, the richness of the resource information of the data center in the region is realized.
Preferably, the distribution scheduling manager schedules the computing resources of the local regional data center and the computing resources of the other regional data centers according to a local scheduling policy and a cross-regional scheduling policy. Thereby realizing a flexible resource scheduling strategy.
Preferably, the local scheduling policy is determined according to an influence factor of resource information utilization of a local data center computing node and a weight ratio of the influence factor, where the influence factor at least includes a CPU utilization and a memory utilization. Determining the problem of matching of the service resource requirement of the regional data center with the resource capacity of each computing node of the regional center, and providing a scheduling basis for realizing resource load balance of the regional data center.
Preferably, the cross-region scheduling policy is determined at least according to the resource information utilization rate of each computing node in the cross-region, the influence factor of the network delay and the IO reading performance, the weight proportion of the influence factor, and the local scheduling policy. In a multi-region data center with different network bandwidths, data scales and IO reading performances, the problem that the service resource requirement of the data center of the region is matched with the resource capacity of computing nodes of other region centers is determined, and a scheduling basis is provided for realizing resource load balance of the cross-region data center.
In addition, the present disclosure also provides a cross-regional data center cloud resource scheduling method, including: acquiring resource information of the local area data center computing node; judging whether the resource information meets the scheduling requirement; if yes, distributing and scheduling resources according to a local scheduling strategy; if not, acquiring resource information of the computing nodes of the data centers in other areas; and when the resource information of the computing nodes of the data centers in other areas meets the cross-area scheduling requirement, distributing and scheduling the resources by using a cross-area scheduling strategy.
By the method, the resource integration and unified management scheduling of the multi-cloud cross-regional data center are realized, and the global load balance of the resources is realized.
Preferably, the method further comprises: creating a logical router of a management node of the regional data center according to an SDN controller of the regional data center management node; creating a unique area name of the local area data center in the local area data center management node; network parameters of an underlay layer of a local area data center management node network are configured through an interconnection module, so that intercommunication of internal devices of the local area data center and/or communication between the local area and other area data centers are realized; generating a logic transmission switch by configuring a virtual switch and a remote port switching function of an SDN controller in a local area data center management node; connecting the logic router with the transmission logic switch;
a physical network flow inlet and outlet is appointed on a routing port of the logic router; and setting an automatic routing announcement on the SDN controller of the local area data center.
By the method, stable communication between the multi-cloud cross-regional isolation networks on the overlay layer and smooth migration of network security policies between the multi-cloud cross-regional data centers are achieved.
Compared with the prior art, the method and the system have the advantages that stable communication between multi-cloud cross-area isolation networks in an overlay layer and smooth migration of network security policies between multi-cloud cross-area data centers can be achieved through the virtual switch, the interconnection module, the SDN controller, the logic router established on the SDN controller and the logic switches established on the virtual switch and the SDN controller; by the resource acquisition agent and the distribution scheduling manager, the resource integration and unified management scheduling of the multi-cloud cross-regional data center are realized, and the global load balance of resources is realized.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present specification, and other drawings can be obtained by those skilled in the art according to the drawings.
Fig. 1 is a schematic structural diagram of an overall cross-region data center cloud resource scheduling system in this embodiment;
fig. 2 is a flowchart of a cross-regional data center cloud resource scheduling method in this embodiment;
fig. 3 is a schematic diagram of a connection structure between data centers in a cross-regional data center cloud resource scheduling system in this embodiment.
Detailed description of the preferred embodiments
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all 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 application.
To facilitate understanding of the embodiments of the present application, the following detailed description is given with reference to the accompanying drawings, which are not intended to limit the scope of the present application.
The cross-regional data center cloud resource scheduling system and method provided by the specific embodiment are applied to a production environment of a large-scale cloud computing virtual network. The method aims to open two large layers of a plurality of data center networks and enable the virtual networks to be in butt joint with physical networks in the environment of cross-regional data center virtual networks. Especially, compared with the defects that a software defined network monitors flow through an integrated controller and deploys security policies, the system and the method for scheduling cloud resources of a cross-regional data center have limited processing capability and are easy to break down after being attacked, the system and the method for scheduling cloud resources of the cross-regional data center can achieve stable communication between multi-cloud cross-regional isolation networks of an overlay layer and smooth migration of network security policies between the multi-cloud cross-regional data centers.
Specifically, according to a first aspect of the present embodiment, a cross-regional data center cloud resource scheduling system is provided, and an overall framework of the system is shown in fig. 1.
The cross-regional data center cloud resource scheduling system comprises a plurality of data centers, and as shown in fig. 1, for convenience of description, a region a is defined as a local region data center, and a region B is defined as another region data center.
Each data center, as shown in fig. 1, includes compute nodes and management nodes.
Wherein the compute nodes provide resource information storage and forwarding between physical networks. The computing node comprises a virtual switch and is responsible for forwarding the flow of other virtual switches of the local area data center according to an OpenFlow flow table issued by the local area data center management node.
In this embodiment, the virtual switch includes a remote port connection module and a remote port forwarding module, which can be implemented by adding the remote port connection module and the remote port forwarding module on the basis of the existing open source virtual switch. The remote port connection module and the remote port forwarding module are added to realize virtual flow communication of the virtual switch among multiple clouds and cross-regions, and authorize the processing of the flow table. Preferably, the computing node further includes a resource collection agent module, and the resource collection agent module is connected to the management node of the regional data center where the resource collection agent module is located, and provides the resource information of the computing node where the resource collection agent module is located to the management node of the regional data center where the resource collection agent module is located, so as to serve as one of bases for scheduling decisions.
Therefore, the information acquired by the resource acquisition agent module includes but is not limited to the use conditions of CPU, memory, storage, IO, network and the like. The data is closely related to the scheduling decision of cross-region or local region, and the information is provided to help the management node to make a quick and accurate decision.
Preferably, in this embodiment, a data center may include a plurality of computing nodes, each computing node including the virtual switch and the resource collection agent module.
The virtual switch of each computing node is connected with the management node of the area, receives the OpenFlow flow table issued by the data center management node and forwards the flow of other virtual switches of the computing node.
Each resource acquisition agent module is connected with the management node of the data center and sends the resource information of the computing node where the resource acquisition agent module is located to the management node of the data center.
With this arrangement, multiple compute nodes may be provided in a data center in order to provide more compute resources that can be scheduled to provide more compute power.
The management node provides resource scheduling and distribution, and forwarding and intercommunication functions of the physical network and the virtual network. The structure of the data center is shown in fig. 1 and 3, and usually only one management node needs to be provided in one regional data center, for example, in fig. 1, the a regional data center has one management node, and the B regional data center also has only one management node. The management node is configured to manage resource scheduling of the local area computing node and implement cross-area resource scheduling, for example, between an area a data center and an area B data center in fig. 1.
Specifically, in this embodiment, the management node includes an SDN controller, and the SDN controller is connected to the virtual switch network of the local area data center to form an SDN network of the local area data center. In the SDN network, the SDN controller is used for issuing an OpenFlow flow table to the virtual switch of a local area data center to guide the virtual switch to forward flow; as described above, the virtual switch is responsible for receiving flow tables from the SDN controller and forwarding the flow to other virtual switch traffic of the local area data center.
A specific implementation manner of the SDN control may include adding a routing forwarding function to an OVN SDN controller to implement the software router based on an OpenFlow flow table, where the software router is responsible for forwarding and learning a routing table between multiple data centers across a region, and converting routing information into an OpenFlow table and issuing the OpenFlow table to a virtual switch of a corresponding computing node in the region, so as to implement intercommunication of virtual isolation networks between data centers across the region.
The management node further comprises a logic router, and the logic router sends the routing information of the data centers in other areas to the SDN controller in a routing announcement mode, so that the routing information of the data centers in other areas is converted into an OpenFlow flow table through the SDN controller and is issued to the virtual switch in the area.
In this specific embodiment, the management node further includes a transmission logic switch, which is connected to the SDN controller and the logic router, respectively, and forwards information of the SDN network of the local data center to the outside through the logic router, and forwards information transmitted by other data centers to the SDN network of the local data center.
In this embodiment, one way to implement the logical switch includes adding a remote port connection module and a remote port forwarding module to the OVN SDN controller and the virtual switch, so that a transport logical switch can be conveniently generated, virtual traffic communication between multiple clouds and cross-regions of the virtual machine is implemented, and processing of a flow table is authorized.
Preferably, the management node further includes an interconnection module, where the interconnection module is connected to the logic switch in the local area data center, and is further connected to interconnection modules of other area data centers; and the interactive communication connection between the local area data center virtual switch and the other area data center virtual switches is realized.
Further, the interconnect module may communicate with a northbound database of the SDN controller via 6645/TCP and with a southbound database of the SDN controller via 6646/TCP. The method and the device can provide the physical network intercommunication among the cross-region multiple clouds of the multicast layer and provide the communication support of the physical layer for the virtual flow intercommunication.
In this embodiment, the management node further includes a distribution scheduling manager. The distribution scheduling manager is connected with the resource information acquisition module in the management node to acquire the resource information of each computing node in the regional data center, and is also connected with the scheduling policy manager to receive the scheduling policy from the scheduling policy manager. In the data center of the region, the distribution scheduling manager is also connected with each computing node and schedules the resources of each computing node. Besides, the resource scheduling manager is connected with the resource region scheduling managers of other region data centers to send and receive cross-region scheduling instructions to and from the distribution scheduling managers of the data center management nodes of other regions. The dispatching of the resource information of the local area data center and the cross-area data center and the distribution function of the resource information are realized through the distribution dispatching manager. The resource scheduling of the local region and the resource scheduling of the cross region are realized through the structure.
Further preferably, when the distribution scheduling manager performs resource scheduling, the resource information collected by the resource information collection module is firstly used as a precondition for distribution scheduling, and subsequent scheduling policy execution can be performed only when resource requirements are met. And then the dispatching strategy provided by the dispatching strategy manager is used as a guide to distribute and dispatch the resources. The method comprises the steps that resources managed by a local data center cannot meet resource requirements, resource information of other regional data centers is collected in a trans-regional mode through an external service interface in a management node, so that preconditions are provided for the trans-regional data center scheduling, and then a trans-regional scheduling strategy which is provided by a scheduling strategy manager and combined with network delay factors and IO reading performance influence factors is used as guidance to distribute and schedule trans-regional resources. Through the structure, the computing resources of the local area data center and the computing resources of other area data centers can be scheduled through the distribution scheduling manager according to the local scheduling strategy and the cross-area scheduling strategy. Thereby realizing a flexible resource scheduling strategy.
And the resource acquisition agent module in the computing node is connected with the resource information acquisition module in the management node so as to send the resource information of the computing node where the resource acquisition agent module is located, which is acquired by the resource acquisition agent module, to the resource information acquisition module of the management node.
The dispatching strategy communication between the cross-regional data centers can be realized through an external resource service interface arranged on the management node, the external resource service interface is connected with the distribution dispatching manager of the region in the region data center and is connected with the resource dispatching managers of other region data centers through a network, and therefore the service interface for providing resource information acquisition and cross-regional resource dispatching support for the outside is provided.
The scheduling policy manager is deployed on the management node, serves as a policy arranging and managing tool, provides a basis for distribution scheduling for the distribution scheduling manager, and is responsible for managing scheduling policies. In this embodiment, the degree policy preferably includes determining the scheduling policy by using the product of the impact factor and the weight ratio, for example. Preferably, the scheduling policy manager may further set the scheduling policy to be executed when the CPU usage or the memory usage reaches a threshold value and lasts for a limited time.
Preferably, the local scheduling policy is determined according to an influence factor of resource information utilization of the local data center computing node and a weight ratio of the influence factor, where the influence factor at least includes CPU utilization and memory utilization. Determining the problem of matching of the service resource requirement of the regional data center with the resource capacity of each computing node of the regional center, and providing a scheduling basis for realizing resource load balance of the regional data center.
The cross-region scheduling strategy is determined at least according to the resource information utilization rate of each computing node in the cross-region, the influence factors of network delay and IO reading performance, the weight proportion of the influence factors and the local scheduling strategy. Therefore, in multi-region data centers with different network bandwidths, data scales and IO reading performances, the problem that the service resource requirement of the data center of the region is matched with the resource capacity of the computing nodes of other region centers is determined, and a scheduling basis is provided for realizing resource load balance of the cross-region data center.
According to another aspect of the present embodiment, there is further provided a cross-regional data center cloud resource scheduling method, where the method is implemented by using the system described above, and the method includes:
acquiring resource information of the local area data center computing node;
judging whether the resource information meets the scheduling requirement; if yes, distributing and scheduling resources according to a local scheduling strategy; if not, acquiring resource information of the computing nodes of the data centers in other areas; and when the resource information of the computing nodes of the data centers in other areas meets the cross-area scheduling requirement, distributing and scheduling the resources by using a cross-area scheduling strategy.
Specifically, a flow of the cross-regional data center cloud resource scheduling method is shown in fig. 2, and the method includes the following steps:
step 1: and the resource information acquisition agent modules of the computing nodes in the regional data center acquire resource information and send the resource information to the resource information acquisition module of the management node.
Step 2: and the distribution scheduling manager takes the acquired resource information as a precondition to execute the resource scheduling policy execution requirement evaluation provided by the scheduling policy manager.
And step 3: the local data center resource meets the scheduling requirement, and the distribution scheduling manager distributes and schedules the resource according to the local scheduling strategy provided by the scheduling strategy manager.
And 4, step 4: the local area data center resources cannot meet the scheduling requirement, and a distribution scheduling manager of the management node acquires resource information of other area data centers in a cross-area mode through an external resource service interface.
And 5: and taking the cross-region acquisition resource information as a precondition, and executing the cross-region scheduling policy execution requirement evaluation provided by the scheduling policy manager by the distribution scheduling manager.
Step 6: and the distribution scheduling manager distributes and schedules resources according to the cross-region scheduling strategy provided by the scheduling strategy manager.
And 7: and if the data center resources in other areas cannot meet the scheduling requirements, scheduling fails, and the shortage of resources is prompted.
Each regional data center is executed according to steps 1-7. By the method, the multi-cloud cross-regional data center resource integration and unified management scheduling are realized, and the global load balance of resources is realized.
Preferably, the method for scheduling cloud resources of a cross-regional data center provided by the present embodiment further includes the following steps:
creating a logic router of a management node of the regional data center according to an SDN controller of the management node of the regional data center; creating a unique area name of the local area data center in the local area data center management node; network parameters of an underlay layer of a local area data center management node network are configured through an interconnection module, so that intercommunication of internal devices of the local area data center and/or communication between the local area and other area data centers are realized; generating a logic transmission switch by configuring a virtual switch and a remote port switching function of an SDN controller in a local area data center management node; connecting the logic router with the transmission logic switch; appointing a physical network flow inlet and outlet on a routing port of the logic router; and setting an automatic routing announcement on an SDN controller of the local area data center.
Specifically, the method for scheduling cloud resources in a cross-regional data center according to the specific embodiment further includes:
a logic router is established in the local area data center through an SDN controller; creating a unique area name for cross-area multi-cloud interconnection in the local area data center; configuring network parameters of other regional data centers which need to be communicated with the regional data center on an underlay level through an interconnection module in the regional data center; configuring a virtual switch and a remote port connection module and a remote port forwarding module of an SDN controller in a local area data center to generate a transmission logic switch; a physical network flow inlet and outlet is appointed on a routing port of a logic router on the regional data center; setting an automatic routing announcement on an SDN controller of a local area data center; and repeating the steps in the data centers of other areas.
Through the steps, stable communication between the multi-cloud cross-regional isolation networks on the overlay layer and smooth migration of network security policies between the multi-cloud cross-regional data centers are achieved.
Compared with the prior art, the method and the system have the advantages that stable communication between multi-cloud cross-area isolation networks in an overlay layer and smooth migration of network security policies between multi-cloud cross-area data centers can be achieved through the virtual switch, the interconnection module, the SDN controller, the logic router established on the SDN controller and the logic switches established on the virtual switch and the SDN controller; through the resource acquisition agent and the distribution scheduling manager, the multi-cloud cross-regional data center resource integration and unified management scheduling are realized, and the global load balance of resources is realized.
The above-mentioned embodiments, objects, technical solutions and advantages of the present application are described in further detail, it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present application, and are not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present application should be included in the scope of the present application.
Claims (10)
1. A cross-regional data center cloud resource scheduling system is characterized by comprising:
the system comprises a plurality of data centers, a plurality of data centers and a plurality of management nodes, wherein each data center comprises a computing node and a management node;
the computing node comprises a virtual switch and is responsible for forwarding the flow of other virtual switches of the local area data center according to an OpenFlow flow table issued by the local area data center management node;
the management node comprises an SDN controller and is connected with the virtual switch network of the local area data center to form an SDN network of the local area data center; the SDN controller issues an OpenFlow flow table to the virtual switch of the local area data center to guide the virtual switch to forward flow;
the logic router sends the routing information of the data centers in other areas to the SDN controller through routing announcements, converts the routing information of the data centers in other areas into OpenFlow flow tables through the SDN controller and sends the OpenFlow flow tables to the virtual switch in the area;
and the transmission logic switch is respectively connected with the SDN controller and the logic router, forwards the SDN network information of the data center of the area outwards through the logic router, and forwards the information transmitted by other data centers to the SDN network of the data center of the area.
2. The cross-regional data center cloud resource scheduling system of claim 1, wherein the management node further comprises an interconnection module, the interconnection module is located in the local regional data center and the logic switch, and is connected to interconnection modules of other regional data centers; and the interactive communication connection between the local area data center virtual switch and the other area data center virtual switches is realized.
3. The cross-regional data center cloud resource scheduling system of claim 1, wherein the management node further comprises,
the distribution scheduling manager sends a resource scheduling instruction of the local area to the computing node of the local area data center according to the resource information from the local area data center and the scheduling strategy of the local area data center; the distribution scheduling manager is also connected with the management nodes of other areas and sends and receives cross-area scheduling instructions to and from the distribution scheduling managers of the data center management nodes of other areas.
4. The cross-regional data center cloud resource scheduling system of claim 3, wherein the computing node further comprises a resource collection agent module, which is connected to a distribution scheduling manager of a management node of the regional data center, and provides resource information of the computing node where the distribution scheduling manager is located to the distribution scheduling manager.
5. The cross-regional data center cloud resource scheduling system of claim 4, wherein the data center comprises a management node, and wherein the management node manages the plurality of computing nodes.
6. The cross-regional data center cloud resource scheduling system of claim 3, wherein the distribution scheduling manager schedules the computing resources of the local regional data center and the other regional data centers according to a local scheduling policy and a cross-regional scheduling policy.
7. The cross-regional data center cloud resource scheduling system of claim 6, wherein the local scheduling policy is determined according to an influence factor of resource information utilization of each computing node in a region and a weight ratio of the influence factor, and the influence factor at least includes a CPU utilization and a memory utilization.
8. The cross-regional data center cloud resource scheduling system of claim 6, wherein the cross-regional scheduling policy is determined at least according to an influence factor of resource information usage, network delay and IO reading performance of each computing node in a cross-region, a weight ratio of the influence factor, and a local scheduling policy.
9. A cross-regional data center cloud resource scheduling method, wherein the method uses the cross-regional data center cloud resource scheduling system according to any one of claims 1 to 8 for resource scheduling, and the method comprises:
acquiring resource information of the local area data center computing node;
judging whether the resource information meets the scheduling requirement; if yes, distributing and scheduling resources according to a local scheduling strategy; if not, acquiring resource information of the computing nodes of the data centers in other areas; and when the resource information of the computing nodes of the data centers in other areas meets the cross-area scheduling requirement, distributing and scheduling the resources by using a cross-area scheduling strategy.
10. The cross-regional data center cloud resource scheduling method according to claim 9, wherein the method comprises:
creating a logic router of a management node of the regional data center according to an SDN controller of the management node of the regional data center;
creating a unique area name of the local area data center in the local area data center management node;
network parameters of an underlay layer of a local area data center management node network are configured through an interconnection module, so that intercommunication of internal devices of the local area data center and/or communication between the local area and other area data centers are realized;
generating a logic transmission switch by configuring a virtual switch and a remote port switching function of an SDN controller in a local area data center management node; connecting the logic router with the transmission logic switch;
a physical network flow inlet and outlet is appointed on a routing port of the logic router;
and setting an automatic routing announcement on the SDN controller of the local area data center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310043278.6A CN115801782B (en) | 2023-01-29 | 2023-01-29 | Cross-regional data center cloud resource scheduling system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310043278.6A CN115801782B (en) | 2023-01-29 | 2023-01-29 | Cross-regional data center cloud resource scheduling system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115801782A true CN115801782A (en) | 2023-03-14 |
| CN115801782B CN115801782B (en) | 2023-05-16 |
Family
ID=85429067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310043278.6A Active CN115801782B (en) | 2023-01-29 | 2023-01-29 | Cross-regional data center cloud resource scheduling system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115801782B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104301391A (en) * | 2014-09-19 | 2015-01-21 | 北京邮电大学 | Multi-domain optical network data center resource virtualization mapping method |
| WO2015192584A1 (en) * | 2014-06-18 | 2015-12-23 | 中兴通讯股份有限公司 | Virtual routing system and method |
| CN109218053A (en) * | 2017-07-03 | 2019-01-15 | 中兴通讯股份有限公司 | Implementation method, system and the storage medium of virtual data center |
| US10567288B1 (en) * | 2017-11-28 | 2020-02-18 | Juniper Networks, Inc. | Automatic discovery, scaling, and load balancing of multiple cloud data centers in a software-defined network environment |
| CN113515350A (en) * | 2021-07-29 | 2021-10-19 | 广州高维网络科技有限公司 | Hybrid cloud computing management platform |
| CN113992577A (en) * | 2021-09-24 | 2022-01-28 | 广东浪潮智慧计算技术有限公司 | Network intercommunication method, device, equipment and medium |
-
2023
- 2023-01-29 CN CN202310043278.6A patent/CN115801782B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015192584A1 (en) * | 2014-06-18 | 2015-12-23 | 中兴通讯股份有限公司 | Virtual routing system and method |
| CN104301391A (en) * | 2014-09-19 | 2015-01-21 | 北京邮电大学 | Multi-domain optical network data center resource virtualization mapping method |
| CN109218053A (en) * | 2017-07-03 | 2019-01-15 | 中兴通讯股份有限公司 | Implementation method, system and the storage medium of virtual data center |
| US10567288B1 (en) * | 2017-11-28 | 2020-02-18 | Juniper Networks, Inc. | Automatic discovery, scaling, and load balancing of multiple cloud data centers in a software-defined network environment |
| CN113515350A (en) * | 2021-07-29 | 2021-10-19 | 广州高维网络科技有限公司 | Hybrid cloud computing management platform |
| CN113992577A (en) * | 2021-09-24 | 2022-01-28 | 广东浪潮智慧计算技术有限公司 | Network intercommunication method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115801782B (en) | 2023-05-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111953526B (en) | Hierarchical computational power network arrangement method, device and storage medium | |
| CN109240821B (en) | Distributed cross-domain collaborative computing and service system and method based on edge computing | |
| Zeng et al. | Orchestrating tree-type VNF forwarding graphs in inter-DC elastic optical networks | |
| CN113448721A (en) | Network system for computing power processing and computing power processing method | |
| CN109257091B (en) | Global load balancing satellite-ground cooperative network networking device and method | |
| CN112600717B (en) | Satellite network management and control protocol semi-physical test device based on SDN | |
| CN107276662A (en) | A kind of software definition Information Network multi-controller dynamic deployment method | |
| CN106790436B (en) | Traffic system monitoring method based on cloud architecture and control center cloud server | |
| CN109196828A (en) | A kind of method for managing resource and device of network slice | |
| CN104468688A (en) | Method and apparatus for network virtualization | |
| CN109660466A (en) | A kind of more live load balance realizing methods towards cloud data center tenant | |
| CN106953744A (en) | A kind of SDN cluster controllers High Availabitity architecture design method | |
| CN113055232B (en) | Network configuration deployment method, device and equipment | |
| CN104301391A (en) | Multi-domain optical network data center resource virtualization mapping method | |
| CN103888360A (en) | Method for integrating covering method to obtain service node in SDN based on greedy algorithm | |
| Mouradian et al. | Application component placement in NFV-based hybrid cloud/fog systems | |
| CN113645146B (en) | New stream density-based software defined network controller load balancing method and system | |
| Liao et al. | Survivable provisioning for multicast service oriented virtual network requests in cloud-based data centers | |
| CN103067287A (en) | Method achieving virtual programmable router under framework of forwarding and control separation | |
| CN105072037A (en) | Dynamic flow generating method for distributed SDN controllers | |
| CN113300952B (en) | Distributed drainage system for cloud security resource pool and drainage method thereof | |
| Saravanan et al. | Improving map reduce task scheduling and micro-partitioning mechanism for mobile cloud multimedia services | |
| CN110535673A (en) | Construction method, equipment and the storage medium of virtual subnet | |
| CN115801782B (en) | Cross-regional data center cloud resource scheduling system and method | |
| CN105847428A (en) | Mobile cloud platform |
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 |