CN119172236A - A multi-level quota management system, method and device - Google Patents

Abstract

The application relates to the field of quota management, in particular to a system, a method and a device for multi-level quota management. The system comprises a multi-level quota management node, when a cloud service wants to apply for new resources, the cloud service sends a resource request to a cloud service deployment module, the cloud service deployment module firstly finds a first quota management node corresponding to the cloud service and sends the resource request to the first quota management node, the first quota management node also sends the resource request to an upper quota management node, and when the first quota management node receives an agreement request of the upper quota management node, the first quota management node sends an agreement deployment instruction to the cloud service deployment module.

Description

Multi-level quota management system, method and device

Technical Field

The application relates to the field of resource management, in particular to a multi-level quota management system, method and device.

Background

With the development of computer science and technology, various manufacturers provide various services for users, wherein some services can open some physical or virtual resources, resources or charge free or for charge for users.

Therefore, various manufacturers can develop similar resource quota management systems, so that services can apply to the quota management systems before acquiring resources, corresponding request resources can be acquired after the systems agree, and the traditional quota management systems use a unified quota management module, so that single-point faults are easy to occur, and overload is easy to cause.

Disclosure of Invention

The embodiment of the application provides a multi-level quota management system, a management method and a device. Before each quota management node returns resource feedback to the cloud service deployment module, resources are applied to the superior parent node, so that the cloud service or the cloud service deployment module can obtain agreements of a plurality of quota management nodes only by interacting with one quota management node.

In a first aspect, a method for managing resources is provided, where a quota management system manages cloud computing resources for deploying cloud services, where the cloud computing resources are provided by a plurality of servers, and where the quota management system includes a multi-level quota management node, where the multi-level quota management node includes a plurality of first quota management nodes and parent nodes of the first quota management nodes, and where the method includes:

The cloud service deployment module receives a request for deploying cloud service;

The cloud service deployment module determines an Internet Protocol (IP) address of the first quota management node;

The cloud service deployment module sends a resource application request to the first quota management node according to the IP address of the first quota management node, wherein the resource application request carries a target resource quota of cloud computing resources required for deploying the cloud service;

The first quota management node applies for cloud computing resources of the target resource quota to a parent node of the first quota management node;

the first quota management node receives feedback of the father node;

If the feedback is a consent application, the first quota management node provides cloud computing resources of the target resource quota to the cloud service deployment module;

the cloud service deployment module deploys the cloud service by utilizing cloud computing resources of the target resource quota.

In the implementation mode, by arranging quota management nodes in a layered manner, the cloud service or the cloud service deployment module can simultaneously contact all the upper nodes by only connecting with one leaf node (a first quota management limit) and transmitting information to the upper nodes by the leaf node, so that the number of interfaces of the service is reduced, the development difficulty is reduced, and the development efficiency is improved.

In one possible implementation manner, if the feedback is a disagreement application, the first quota management node feeds back an application result to the cloud service deployment module.

In this implementation, if the feedback result of the upper node is an instruction such as disagreement or refusal, it indicates that although the first quota node may have enough quota, the quota of the upper node is insufficient, and thus related resources should not be opened for the cloud service.

In one possible implementation, the quota limit of at least one quota managing node is smaller than the sum of the quota limits of all its child nodes.

In the implementation mode, the quota management system adopts a super quota management mode, the super quota is a core quota mode in the quota management system, and the use efficiency of the quota management system on resources can be improved by using the super quota management mode, so that the operation cost is reduced.

In one possible implementation, the resource application includes a resource usage application and/or a resource reservation application.

In the implementation mode, besides the use application can be provided when the actual use of the resources is needed, the resource reservation request can be sent out in advance for the resources which are possibly used by the important business in the future, and the resources are prevented from being preempted.

In one possible implementation, the quota managing node and the service are configured with corresponding cache layers.

In the implementation mode, each quota management node and service can achieve the effect of reducing the request and inquiry times to the father node by caching related information in the cache layer, so that the operation pressure of the whole quota management system is reduced, and the availability of the system is improved.

In one possible implementation, the cache layer of the service contains address information of its corresponding quota-managing leaf node.

In this implementation, the service may query the buffer layer for address information of the corresponding quota-managed leaf node, and after determining the address information of the corresponding quota-managed leaf node, may send a corresponding request to the leaf node. Only after the address information of the corresponding leaf node is not queried in the cache layer, the query request of the leaf node address information is sent to the root node.

In one possible implementation manner, the cache layer of the quota management node includes a quota amount returned by a quota management father node of the quota management node.

In such an embodiment, if the cache layer node of the quota managing node contains the quota amount returned by its parent node,

In this implementation manner, the cloud service deployment module may first send a query request to the root node, obtain a next-layer node corresponding to the root node and the cloud service, and then send the query request to the next-layer node, so as to perform a round trip, and finally, be able to find a first quota management node directly corresponding to the cloud service.

In a second aspect, a resource management system is provided, where the system is implemented by the system in the first aspect, and details are not repeated herein.

In a third aspect, a cluster of computing devices is provided, the cluster of computing devices comprising at least one computing device, each computing device comprising a processor and a memory;

the processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method of any of claims 1-7.

In a fourth aspect, a computer program product comprising instructions is provided, the computer program product comprising instructions which, when executed by a cluster of computing devices, cause the cluster of computing devices to perform the method of any of claims 1-7.

This implementation manner is a computer program product implementation manner of the second aspect and the third aspect, and will not be described in detail here.

In a fifth aspect, a computer readable storage medium is provided, the computer readable storage medium comprising computer program instructions which, when executed by a cluster of computing devices, perform the method of any of claims 1-7.

Drawings

Fig. 1 is a schematic structural diagram of an existing quota management system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a relationship between quota managing nodes according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a multi-level quota management system according to an embodiment of the application;

FIG. 4 is a quota management method provided by an embodiment of the application;

FIG. 5 is a leaf node query method according to an embodiment of the present application;

FIG. 6 is a diagram of a resource reservation method according to an embodiment of the present application;

FIG. 7 is a diagram illustrating a second method for reserving resources according to an embodiment of the present application;

FIG. 8 is a method for using resources according to an embodiment of the present application;

FIG. 9 is a diagram of a resource query method according to an embodiment of the present application;

FIG. 10 is a diagram illustrating a second resource query method according to an embodiment of the present application;

FIG. 11 is a diagram illustrating a method for applying resources according to an embodiment of the present application;

FIG. 12 is a schematic diagram of a computing device according to an embodiment of the present application;

FIG. 13 is a schematic diagram of a computing device cluster according to an embodiment of the present application;

FIG. 14 is a schematic diagram of a computing device cluster according to an embodiment of the present application;

Detailed Description

The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application.

Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the specification. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in one possible implementation," "in other embodiments," and the like in various places throughout this specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless specifically noted otherwise.

In the description of the present specification, "/" means or means, for example, a/B may mean a or B, and "and/or" herein is merely an association relationship describing an association object means that three relationships may exist, for example, a and/or B may mean that a exists alone, a and B exist together, and B exists alone, unless otherwise specified. In addition, in the description of the embodiments of the present specification, "a plurality" means two or more than two.

In the description of this specification, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

The resource is not limited in kind, can be provided for users in a service mode, and can be perceived by the users and bring value to the users. For example, in the cloud service, computing, storage, and network can be regarded as resources, and the cloud service can provide computing, storage, and network services with different levels and different amounts of computing power, storage space, bandwidth, and the like to users according to the needs of clients, and these types of services can be called resources.

Quota-in some services or quota-management nodes that provide resources, the theoretical upper usage limit of a resource allocated by an upper level quota-management node is called the quota of this user. The quota managing node quota amount may not be a real number of resources. For example, as shown in fig. 3, if the quota managing node 102 manages 500 units of quota in the resource pool. The quota sum of the quota managing node 103 and the quota managing node 106 is 600 units, and the quota sum exceeds the quota (500 units) of the parent node, namely the quota managing node 102. If the quota managing node 106 applies to the quota managing node 102 for using 300 units of the new resource, the quota managing node 102 remains 200 units of the quota. However, at this time, the quota of the quota managing node 103 is still 300 units (because it is not applied for use), so the quota of the quota managing node 103 exceeds the real number of resources in the resource pool. Quota amounts may also be referred to simply as quotas.

Over-allocation, in the process of allocating resources, since the user does not use the quota fully under normal conditions, in order to improve the use efficiency of the resources, the resource provider often provides the user with resource quota exceeding the actual owned number. For example, when the resource type is a storage resource, if the resource party actually owns 200 units of the storage resource, it provides 100 units of quota to user a and 100 units of quota to user B. Under the quota mode, although the quota received by each user can be ensured to be a real quota, in the actual use process, the resource utilization rate of the user on the quota is only 50%, so that 200 units of storage resources actually provided by a resource party are only used for 100 units, and 100 units of storage resources are idle, thereby causing resource waste. At this time, the resource provider may provide 100 units of quota to the user C, and in the case of a typical 50% resource usage, the actual resource usage is only 150 units, and although the sum of the quota amounts of the users A, B, C is 300 units, the sum exceeds the storage resources actually owned by the resource provider, and the user will not generally use the resources in full, so the service can still be provided normally. This way of quota beyond the true quota limit actually owned by the resource provider is called overdropping.

Supermatch error refers to a system confusion or error caused by supermatch behavior. For example, when the resource type is a storage resource, if the resource party actually owns 200 units of the storage resource, it provides 100 units of quota to user a and 100 units of quota to user B. Under the quota mode, although the quota received by each user can be ensured to be a real quota, in the actual use process, the resource utilization rate of the user on the quota is often only 50%, so that 200 units of storage resources actually provided by a resource party are only used for 100 units, and 100 units of storage resources are idle, thereby causing resource waste. At this time, the resource providing convenience provides 100 units of quota to the user C, and under the condition of 50% of the normal resource usage rate, the actual total resource usage rate is only 150 units, and although the sum of the quota amounts of the users A, B, C is 300 units, which exceeds the actual quota amount 200 units actually owned by the resource provider, the service can still be normally provided. However, when the resource usage rate of the user a and the user B reaches 70%, that is, the user a and the user B apply for using 70 units of resources, at this time, the quota of the resource provider only remains 60 units, at this time, the user C still has 100 units of quota, but when the user C applies for using 70 units of resource amount, the request of the user C is either refused, or after obtaining consent, the related resources cannot be actually called, or the resources managed by other quota management nodes are forcibly called, so that the system management logic is confused, and the system error that cannot be actually called or forcibly called is called as a overdry error.

The resource provider refers to a main body for providing resources to users for renting and selling in a service mode, for example, cloud service manufacturers, and the resource providing mode can be a network mode or other modes, and is not limited herein, and the cloud manufacturers can be taken as the most common resource provider for example.

And the user points to a main body of the resource provider to rent and purchase the resource, the purchased and rented resource can be provided for the main body in a service mode, and the purchased and rented resource can be provided with the service remotely in a network mode under the cloud service scene.

Cloud-a collection of hardware and software resources that utilizes internet technology to aggregate huge and scalable IT capabilities (i.e., computing, storage, networking, etc. resources) as a model for cloud services to users. Typically, a cloud is provided with a plurality of areas in each country/region, each area including at least one data center, and each data center is provided with hardware resources and software resources. Different cloud service providers build different clouds, such as Hua Chen cloud of Hua Chen corporation, azure cloud of Microsoft corporation, etc. Different clouds offer resources (including computing, storage, networking, applications, etc.) leases to users in the form of cloud services. The cloud is dynamically scalable and can meet the needs of growing application and user sizes by aggregating more resources. Cloud computing supports a user to acquire cloud services at various positions by using various terminals, and hardware resources and software resources supporting the cloud services come from the cloud. Common cloud services include three types, infrastructure AS A SERVICE (IaaS), platform AS A SERVICE (PaaS), software services (software AS A SERVICE, saaS). The infrastructure services include virtual machine computing services, data storage services and the like, resources such as hardware devices and the like are mainly provided for users in the form of virtual machines/containers as services, and the users can load and run own applications through the virtual machines or storage resources provided by the cloud services. Platform services are primarily enabling developers to build and deploy their own applications on the cloud-provided infrastructure and operating environment and allow these applications to utilize resources in the cloud infrastructure. A software service is an application running in the cloud that is provided as a service to a user on demand, such as a voice transcription service on the cloud, a face recognition service on the cloud, etc.

Quota management-resource providers, such as cloud vendors, typically need to perform a restrictive management of resources used by users to purchase, lease, services, for reasons of resource limitation, among other reasons, referred to as quota management. Such restrictive management matters are called quota management matters, and generally include matters such as whether the actual usage amount of resources exceeds a quota amount, whether the inventory resources of the resource provider are insufficient, whether the user is arrears, whether the user is illegal, and the like. The quota management behavior may specifically be processing a resource usage application (consent or refusal), sending a corresponding instruction to a lower node, allocating a new resource to the cloud service, adjusting a quota of the cloud service, directly calling the corresponding new resource, and the like.

The area refers to the division of the resource pool inside the resource provider. The division mode can be according to different regions (such as south China area and north China area) or by other modes. The purpose of the division may be based on the purpose of facilitating internal management, or may be based on other purposes (for example, to ensure the quality of service, different areas are generally divided based on regions, and resources in different areas are allocated according to users in different regions), and geographic locations between different areas of each cloud service manufacturer are generally far apart (for example, a northern china area, a southern china area, a western united states area, etc. of a certain cloud service provider).

The main function of the quota management node is to decide how to process a new resource use application of the service according to the quota of the service. For example, whether the resource usage of the service itself has reached the quota upper limit is checked, and if the quota upper limit has been reached, it is indicated that the number of resource usage of the service has exceeded the maximum number of resource usage allocated to the service, so that negative measures such as refusing or not responding to the resource call application are performed.

Due to the limited nature of resources, quota has become a common means for various types of resource providers to manage resource allocation, with the following benefits:

Protecting the resource provider. Without quota management, each user can use the resources provided by the resource provider without any upper limit, and once some malicious users are encountered, a few users can maliciously occupy a large amount of resources, so that the service integrity provided by the resource provider is affected, and other normal users can not use the service normally.

Protecting the user. In some charged services, the charging mode of the charging service is usually charged according to the use quantity of the resources, and the upper limit of the user on the use of the resources is also usually provided with a certain financial budget. If the quota function is not available, the user cannot accurately know the use condition of the resources of the user, or the user uses excessive quota due to negligence of the user, so that the user suffers a great amount of economic loss.

For the above reasons, quota management on resources has become a common type of management means for various resource providers, especially in the cloud service field, and various cloud service manufacturers use various same or different quota management means.

And the cloud service is a collection of resources and related programs provided by a cloud manufacturer in a service mode under a certain user account. The related program is a program that enables a user to actually control and use cloud service resources.

When the cloud service sends a new resource application to the quota management node, the cloud service means that a certain user applies for a new resource to the quota management system.

The existing main stream technical scheme is as follows:

As shown in fig. 1, a plurality of services are simultaneously connected to the same quota management node, and one quota management node uniformly manages the resource usage of all the services.

The above prior art has the following disadvantages:

The single-point fault easily occurs, because the configuration mode only has one quota management node, if the quota management node fails, all quota management services are invalid, and in the prior art, the quota management node needs to bear all quota management requests at the same time, so that the load pressure is high, and the maintenance of the services is not facilitated.

In order to solve the problems, the invention provides a multi-level quota management system, wherein each quota management node in the quota management system is arranged in a layered manner. For convenience in describing the hierarchical relationship, the quota management nodes are divided into root nodes, father nodes, child nodes and leaf nodes according to different hierarchical levels of the system.

The multi-level quota management system is characterized in that each quota management node is not configured into independent parallel architectures, and instead, each quota management node is distributed at different upper and lower levels in the same system according to different management levels, and the number of quota management nodes with higher levels is generally smaller, and the number of quota management nodes with lower levels is generally larger.

Root node-root node refers to the quota management node at a higher or highest level in the multi-level quota management system.

Leaf node-leaf node refers to a quota management node at a lower or lowest level in a multi-level quota management system, which may be directly connected to a service.

The parent node is a parent node of a quota management node, which is the previous quota management node for managing a certain quota management node, and the leaf node directly connected with the service can also be regarded as the parent node of the service. For example, according to fig. 2, the quota managing node B is directly connected to the quota managing node C and is located at an upper layer of the quota managing node C, so the quota managing node B may be regarded as a parent node of the quota managing node C, and similarly, the quota managing node a is directly connected to the quota managing node B and is located at an upper layer of the quota managing node B, so the quota managing node a may be regarded as a parent node of the quota managing node B. Since the quota management node C is at the lowest layer of this quota management system and is directly connected to the cloud service D, the quota management node C may be referred to as a leaf node, or may be referred to as a parent node of the cloud service D.

The sub-node is a sub-node of the next layer of quota management node managed by a certain quota management node, which is called as the sub-node of the quota management node, and the service can be regarded as the quota management sub-node of the leaf node directly connected with the sub-node, and the quota management sub-node can also be called as the sub-node for short. For example, according to fig. 2, the quota managing node B is directly connected to the quota managing node a and is located at a lower layer of the quota managing node a, so the quota managing node B may be regarded as a child node of the quota managing node a, and similarly, the quota managing node C is directly connected to the quota managing node B and is located at a lower layer of the quota managing node B, so the quota managing node C may be regarded as a child node of the quota managing node B. Since the quota management node C is at the lowest layer of the quota management system and directly connected to the cloud service D, the quota management node C may be referred to as a leaf node, and the cloud service D may be referred to as a child node of the quota management node C.

An architecture diagram of a multi-level quota management system is provided below, see fig. 3.

The quota management system consists of 4 layers of quota management nodes, including a root node 101, a quota management node 102, a quota management node 103, a quota management node 106, a leaf node 104 and a leaf node 107. The above nodes are respectively located at different levels of the quota management system and are connected in sequence, wherein the root node 101 is located at the uppermost layer of the quota management system, and the leaf node 104 is located at the lowermost layer of the quota management system and is directly connected with the cloud service 105.

The real resource quantity is that the real available residual resource quantity in the area corresponding to the resource pool is the real resource quantity. Typically, the highest level quota management node can know the actual number of resources of the resource pool. The actual amount of resources may also be referred to as an actual quota.

In the case where the superdistribution behavior is generally applicable, in order to increase the resource usage, each parent node may superdistribute resources to each child node, which makes the resources inside each quota management node may not be truly allocatable resources. For example, taking fig. 3 as an example, if the quota management node 102 has a resource quota of 500 units, it allocates 300 units of resources to each child node, and the quota management node 103 and the quota management node 106 are both allocated to 300 units of resources, while the quota management node 103 allocates 200 units of resources to each child node, that is, the leaf node 104 and the leaf node 107 are both allocated to 200 units of resources. If the cloud service 105 sends a resource usage application to the leaf node 104 to apply for using 200 units of resources, and if the leaf node 107 has applied for using 200 units of resources to the quota management node 103 at this time, the quota remaining in the leaf node 104 actually remains only 100 units, and cannot be satisfied with 200 units of resource usage application. Meanwhile, if at this time, the quota management node 106 has applied for 300 units of resources to the quota management node 102, and the leaf node 107 applies for 200 units of resources to the quota management node 102 through the quota management node 103, the remaining resources of the quota management node 102 actually only remain 0 units, and cannot satisfy the 200 units of resource usage application of the leaf node 104. Thus, after the cloud service 105 applies for resource usage to the leaf node 104, the leaf node 104 needs to forward the request to the higher node step by step until after obtaining the consent instruction of each parent node, the party can allocate the resource requested by the cloud service 105. The following describes a method for examining a new resource application of cloud service:

In this embodiment, when the cloud service 105 needs to expand capacity, it may send a resource usage application to the leaf node 104, and after the leaf node 104 examines the quota management item in charge of it, if the service passes, the service application continues to be sent to its parent node, that is, the quota management node 103, and after the quota management item in charge of it is examined by the quota management node 103, if the service application passes, the service application continues to be sent to its parent node, that is, the quota management node 102, and this step is repeated until the root node 101. After the root node 101 receives the application sent by the quota management node 102, it examines the quota management item according to the responsible quota management item, if the service is passed, a consent instruction is sent to its child node, namely the quota management node 102, otherwise, a refusal instruction is sent or a reply is not sent. When quota management node 102 receives the grant instruction from root node 101, it continues to send the grant instruction to its child node, quota management node 103. This process is repeated until the cloud service 105, and each child node will issue an approval command to its child node at a further level only after receiving the approval command from the parent node. In this way, the cloud service 105 only needs to connect with the leaf node 104, and can only issue a new resource usage application to obtain the consent of all the associated quota management nodes (root node 101, quota management node 102, quota management node 103, leaf node 104).

In such an embodiment, each quota managing node is configured to send a grant instruction to its child node only after receiving the grant instruction from the parent node. The method effectively avoids the defects caused by the super-allocation behavior in the layered quota management system. Because if the leaf node 104 wants to send the grant instruction to the cloud service 105, the grant instruction of the quota management node 103 needs to be obtained first, but if the quota management node 103 wants to send the grant instruction to the leaf node 104, the grant instruction of the quota management node 102 needs to be obtained first, and similarly, if the resource is to be allocated to the cloud service 105, the grant of each upper quota management node needs to be obtained first, and before each quota management upper node sends the grant instruction to its child node, whether the quota managed by itself satisfies the new resource use request is checked, and only after the upper quota management node of each cloud service passes the internal check, the cloud service 105 can receive the grant instruction of the leaf node 104. The situation that the child node has enough resource amount, but the father node does not have enough amount and agrees to the resource use application of the cloud service is avoided.

Meanwhile, in the embodiment, the cloud service 105 only needs to be connected with the leaf node 104, and does not need to be connected with the root node 101, the quota management node 102 and the quota management node 103, so that the development times of the docking API are greatly reduced, quota management is not needed to be carried out in the cloud service 105, the internal code of the cloud service 105 is simplified, the internal maintenance of the cloud service 105 is facilitated, and the development efficiency is improved. Meanwhile, due to the adoption of the multi-level quota management system, even if one leaf node fails, normal service operation of other leaf nodes or other quota management nodes is not affected, the probability of single-point failure is greatly reduced, and the robustness of the system is improved.

In one possible implementation manner, the cloud service 105 may also directly send a new resource usage application to the root node 101, and after the root node 101 receives the new resource usage application of the cloud service 105, according to the result of the internal examination, send an instruction to its child node, namely, the quota management node 102. After receiving the instruction sent by the parent node, i.e., the root node 101, the quota management node 102 sends an instruction to the child node, i.e., the quota management node 103, according to the internal examination result, and the steps are repeated to the cloud service 105.

In the embodiment, the process of gradually upwards transmitting the new resource application can be omitted, the pressure of system operation is relieved, and the overall operation efficiency of the quota management system is improved.

Quota management information, which is information related to quota management matters, according to which a quota management node can judge what action (agreement or rejection) is made by the new resource use application sent by the cloud service or a subordinate node. The quota management information comprises quota amounts of all quota management nodes. For example, the quota management information of the quota management node includes the remaining quota of the area in which the quota management node is responsible, and the quota management node can determine whether the quota management node can meet the number of new resource usage applications according to the number of remaining quota (the determination mode may be that if the quota is greater than the new quota of the application, the quota is agreed, otherwise, the quota is refused), and make a corresponding response. If the child node has quota management information of the parent node, the child node can replace the parent node to make a judgment to determine whether to agree to a new resource application. And after the child node makes corresponding judgment according to the quota management information of the parent node, the child node is regarded as receiving a corresponding instruction of the parent node. For example, when the child node examines the new resource usage application according to the quota management information of the parent node, the child node agrees, and receives the agreement instruction of the parent node.

And checking, namely judging and responding the new resource use application by the quota management node according to the quota management information.

And (3) internal examination, namely judging and responding actions of each quota management node on the new resource use application according to the quota management information of each quota management node.

The child node comprises instructions, wherein the instructions can execute or enable the child node to execute the corresponding quota management behaviors and information. For example, if the conclusion of the parent node inspection is to agree to the application for using the new resource of the child node, a corresponding approval instruction is sent to the child node. If the child node is also a quota management node, the child node can send an agreement instruction to the child node at the next level after receiving the agreement instruction, and if the child node is cloud service, the cloud service can directly call the new resource applied by the child node. If the refusal action of a certain quota management node is set as that the new resource application is not responded. The act of not responding to the new resource application may also be considered to return instructions to the child node. For example, if the parent node is a quota management node and the conclusion of the parent node inspection is to reject the application for new resources of the child node, the child node may send corresponding reject information or not respond to the application for new resources of the child node, and the child node may not send an instruction for consent or execute a consent related quota management action to the child node at a further level before receiving no consent information. In this way, the same effect can be achieved when the information of rejecting the child node and the new resource use application of the child node are not responded, so that the child node can be regarded as issuing related instructions.

In one possible implementation, each quota management node and cloud service may also have a corresponding caching layer in which quota management information for its parent node may be cached.

In this implementation manner, after receiving a new resource usage application of a cloud service or a child node, each quota management node examines the quota management items responsible for itself, if the quota management item passes, the quota management information of the parent node can be queried in the cache layer of the quota management node, if the quota management item has the quota management information, the parent node is replaced to make a corresponding examination, if the quota management item passes, an agreement instruction is sent to a lower child node or the cloud service, and the specific steps are shown in fig. 4.

S101, the cloud service 105 transmits a resource usage application to the leaf node 104.

S102, after receiving the application of the resource use of the child node, and after the leaf node 104 passes the internal examination, inquiring whether the cache layer has quota management information of the parent node.

S103, if the leaf node 104 caches quota management information of the parent node, namely the quota management node 103, corresponding examination is made according to the quota management information of the parent node, and an instruction (approval or rejection) is sent to a subordinate child node, namely the cloud service 105 according to an examination result.

S104, if the leaf node 104 has no parent node quota management information in the cache layer, continuing to normally transmit a resource use application to the parent node, namely the quota management node 103.

S105, after receiving the application of the resource use of the child node, and after the internal examination of the quota management node 103 passes, inquiring whether the quota management information of the parent node exists in the cache layer.

S106, if the quota management node 103 caches the quota management information of the parent node, namely the quota management node 102, then corresponding examination is made according to the quota management information of the parent node, and an instruction (approval or rejection) is sent to the subordinate child node, namely the leaf node 104 according to the examination result.

S107, if the quota management node 103 does not have the parent node quota management information in the cache layer, the quota management node continues to normally transmit the application for using the resource to the parent node, namely the quota management node 102.

S108, after receiving the resource use application of the child node, and after the internal examination of the quota management node 102 passes, inquiring whether the quota management information of the parent node exists in the cache layer.

S109, if the quota management node 102 caches quota management information of the parent node, namely the root node 101, then corresponding examination is made according to the quota management information of the parent node, and an instruction (approval or rejection) is sent to the subordinate child node, namely the quota management node 103 according to the examination result.

S110, if the quota management node 102 does not have the parent node quota management information in the cache layer, the application for using the resource is continuously and normally transferred to the parent node, namely the root node 101.

S111, after receiving the resource usage application transmitted by the quota management node 102, the root node 101 makes an internal examination according to the quota management item for which it is responsible.

S112, after the internal examination is completed, the root node 101 returns an instruction to its child node, that is, the quota management node 102, according to the result of the examination.

S113, after receiving the instruction returned from the parent node, i.e., the root node 101, the quota managing node 102 returns an instruction to the child node, i.e., the quota managing node 103.

S114, after receiving the instruction returned from the parent node, that is, the quota management node 102, the quota management node 103 returns an instruction to the child node, that is, the leaf node 104.

S115, after receiving the instruction returned from the parent node, that is, the quota managing node 103, the leaf node 104 returns an instruction to the child node, that is, the cloud service 105.

It should be noted that, after receiving the instruction returned by the parent node or regarding that the instruction returned by the parent node is received, the child node may directly execute the corresponding quota management behavior. For example, after the quota management node 103 receives the instruction returned by the quota management node 102 in S113 and the relevant quota management information of the parent node, that is, the quota management node 102, is queried from the self-cache layer in S106 and makes a relevant determination instead of the parent node, step S114 may be executed, and a relevant instruction is returned to its child node, that is, the leaf node 104.

In the embodiment, by setting the corresponding cache layer for the quota management node, the cloud service does not need to obtain approval of all the level nodes each time in the process of transmitting the new resource use application upwards, and the operation pressure of the whole quota management system is reduced to a certain extent.

In such an embodiment, a superordinate error may result because the child node may only need to obtain consent of the parent node (censored by quota management information of the parent node in the cache layer instead of the parent node), and not need to obtain consent of all upper nodes. For example, if the quota of the leaf node 104 is 300 units, the quota of the quota managing node 103 cached in the leaf node 104 cache layer is 200 units, and the quota of the quota managing node 102 is 100 units. At this time, if the cloud service 105 sends a resource usage application of 200 units to the leaf node 104, the leaf node 104 passes the examination according to 300 units of the quota, and if the examination according to 200 units of the quota of 200 units of the quota management node 103, which is a parent node cached in the cache layer, passes, an approval instruction is returned to the cloud service 105, but at this time, the actual quota of 100units of the quota management node is only available, so that the cloud service 105 still cannot call when actually calling resources, or the management logic of the whole quota management system is disturbed due to forced calling.

Therefore, the invention provides a method which can improve the accuracy of the sub-node cache information and avoid the occurrence of the super-allocation error in a multi-level allocation management system with a cache layer. If the child node does not inquire quota management information of the parent node in the cache layer, a new resource application is transferred to the parent node, and a resource inquiry request can be carried. After the father node examines according to the quota management matters, an instruction is returned to the child node, wherein the instruction can carry the quota management quota of the father node. And after receiving the quota management quota returned by the father node, the child node stores the quota into the cache layer. If the father node returns an instruction to the child node, the father node also caches the quota of the upper node, and the father node returns the quota carried by the instruction to the child node, which may be a quota lower than the quota of the upper node, where the quota of the father node is cached in the father node. Next, as illustrated in fig. 4, it is assumed that the leaf node 104, the quota management node 103, the quota management node 102, and the root node 101 have quota amounts of 40 units, 30 units, 20 units, and 10 units, respectively. If quota management node 102 passes a new resource usage application to root node 101, this application carries a resource query request at the same time. After the root node 101 examines the quota management issue, an instruction is returned to the quota management node 102, which may carry the quota of 10 units for the root node 101. Quota managing node 102 stores this quota amount in the cache layer. When the quota management node 103 transmits a new resource usage application to the quota management node 102, the application carries a resource query request, and after the quota management node examines the quota 20 units of the quota management node and the quota 10 units in the cache layer, the quota management node returns an instruction to the quota management node 103, wherein the instruction can carry the quota of the quota management node 102, and at this time, two quota values are in the quota management node 102, namely, the quota 20 units of the quota management node 102 and the quota 10 units received from the root node 101, buffered in the cache layer. The quota amount returned by the quota managing node 102 to the quota managing node 103 may be a quota amount with a lower amount than the two quota amounts, i.e. a quota amount of 10 units received from the root node 101. Therefore, after the quota management node 103 receives the quota returned by the quota management secondary point 102, the quota thereof is 30 units of quota management number of the quota management node 103 and 10 units of quota queried from the quota management node 102. By the pushing, the last stored resource management number of the child node 104 is that the quota management number of the child node 104 is 40 units, the quota is buffered in the buffer layer and is 10 units, and the quota is received from the quota management node 103. This way, it can be ensured that the quota buffered in the buffer layer of the leaf node 104 is the quota with the lowest number in all the upper nodes. It may be ensured that the result of the examination made from this quota is trusted.

Meanwhile, if the parent node returns the quota to the lower node, the lower node or any node lower than the lower node finally obtains the conclusion of agreeing to the resource opening application, and the returned quota can be regarded as an instruction of agreeing. Similarly, if the amount returned to the lower node eventually results in a rejection judgment or instruction made by any lower node, the parent node is considered as the rejection instruction returned by the parent node.

In one possible implementation, the information returned by the parent node to the child node is. Quota management information of the parent node itself, and all quota management information received from a superior node of the parent node. Next, as illustrated in fig. 4, it is assumed that each of the leaf node 104, quota management node 103, quota management node 102, and root node 101 has a quota of 40 units, 30 units, 20 units, and 10 units, respectively. The quota management node 102 internally manages 20 units of quota management information of itself and 10 units of quota management information of the root node 101 acquired from the root node 101, and if the quota management node 103 requests to return the quota management information to the quota management node 102, the quota management node 102 returns 2 quota management information of 20 units and 10 units to the quota management node, so that the quota management node 103 internally has three quota management information of 30 units, 20 units and 10 units. Similarly, the additional management information last cached by the leaf node 104 includes 40 units, 30 units, 20 units, 10 units.

In one possible implementation, the data in the cache layer may be set to have a certain expiration time, and a resource query request needs to be sent to the parent node again every time a period of time passes, so that real-time performance of the cached data in the cache layer is ensured once. The method of resource query can be seen in the description of fig. 10.

Because the user frequently applies to use new cloud services and frequently closes old cloud services, and needs to decouple the cloud services from the quota management system, finding corresponding management leaf nodes for the newly generated cloud services quickly and efficiently becomes a technical problem to be solved, and a method for quickly finding corresponding leaf nodes for the cloud services in the multi-level quota management system is shown below, which is shown in fig. 5.

S201, the cloud service 105 transmits a leaf node query request to the root node 101.

S202, after receiving the query request of the cloud service 105, the root node 101 determines, according to the information carried in the query request sent by the cloud service 105, the corresponding child node, that is, the quota management node 102, and returns address information of the corresponding child node, that is, the quota management node 102.

S203, the cloud service 105 sends a leaf node query request to the quota management node 102 according to address information of the quota management node 102, which is a corresponding child node returned by the root node 101.

S204, after receiving the query request of the cloud service 105, the quota management node 102 determines a corresponding child node, namely the quota management node 103, according to the information carried in the request sent by the cloud service 105, and returns address information of the corresponding child node, namely the quota management node 103.

S205, the cloud service 105 sends a leaf node query request to the quota managing node 103 according to the address information of the corresponding child node, that is, the quota managing node 103, returned by the quota managing node 102.

S206, after receiving the query request of the cloud service 105, the quota management node 103 determines the corresponding child node, namely the leaf node 104, according to the information carried in the request sent by the cloud service 105, and returns the address information of the corresponding child node, namely the leaf node 104.

S207, the cloud service 105 completes the address query for the leaf node 104.

In such an embodiment, the parent node is aware of the address information of all its child nodes. By way of example, the root node 101 is aware of the address information of the quota managing node 102, but not necessarily of the quota managing node 103, but the quota managing node 102 should be aware of the address information of the quota managing node 103, since the quota managing node 103 is a child of the quota managing node 102.

The address information may be address information of a quota-managing node, such as an IP address, or information generated according to other communication protocols, standards, capable of locating it by finding its location through the network.

In such an embodiment, a trusted root node 101 may be provided whose address information should remain relatively stable to facilitate the newly generated cloud service sending leaf node query requests thereto.

The query request should carry the attribution information of the cloud service 105, where the attribution information can enable the quota management node to resolve the corresponding child node information. For example, when root node 101 receives a leaf node query request of cloud service 105, it may have multiple child nodes, including child nodes corresponding to cloud service 105, namely quota management node 102, and quota management node 10X (not labeled in the figure) not corresponding to cloud service 105. The feature information should enable the root node 101 to identify the child node corresponding to the cloud service 105, that is, the quota management node 102, and return address information of the quota management node 102 to the cloud service 105.

According to the features inherent to the service, for example, the region information can be used as one of the attribute information, if the cloud service address is a Dragon post region, dragon post-Shenzhen-Guangdong-China can be used as the query basis of the features, the resource type can also be used as one of the inherent information, if the cloud service application resource is a container, the computing resource-non-bare metal computing resource-container can be used as the query information of the attribute. Here, the method for setting the example features of the regional information to be carried in the query request is as follows:

And setting attribution information according to the region information of the service, wherein the region information is the region where the newly generated cloud service is located, and the quota management node takes the region information as a distinction. For example, assuming that the leaf node 104, the quota management node 103, the quota management node 102 and the root node 101 are respectively a Shenzhen area, a Guangdong province and a Chinese quota management node, and are responsible for quota management matters of the above areas, when the cloud service 105 from the Longpen area sends a leaf node query request to the root node 101 in the Chinese area, the root node 101 can query the corresponding lower child node, namely the quota management node 102 responsible for the Guangdong province quota management matters, according to feature information (Longpen area) carried by the query request. The query method may be that each quota management node is internally configured with a home information corresponding table, where the feature corresponding table records corresponding rules of various home relation corresponding information and quota management nodes, for example, when the feature information is "Longguang district", the home corresponding table should record a corresponding relation of "Long Gang-Shenzhen-Guangdong-China" so that the quota management node can determine a child node corresponding to the cloud service.

In addition, the attribution information content may be attribution information set manually. Here, the manually set cloud service name is exemplified:

And setting identification characteristic information according to the naming space information, namely, when a new cloud service is generated, carrying out programmatic naming on the name of the cloud service, wherein the programmatic naming is required to enable the quota management node to directly judge the corresponding child node according to the naming content. For example, the naming content of the new cloud service may be "root node 101/quota management node 102/quota management node 103/leaf node 104", wherein "/" is used as a division flag of a parent node and a child node, in this way, when the cloud service 105 sends a leaf node query request to the root node 101, the root node 101 can identify that its corresponding child node is the quota management node 102 according to the content of its name, and accordingly, address information of the quota management node 102 is returned to the cloud service 105. The naming space information can be obtained through calculation, for example, if a user applies for using 100 units of storage resources on a Dragon post, the cloud service can calculate and obtain a naming rule of 'China/Guangdong/Shenzhen/Dragon post/storage/500 units storage' according to the information, the naming rule can be set according to specific requirements of each quota management node, and the invention does not limit the naming rule in any way.

In one possible implementation manner, a certain interval is reserved between a new resource usage application issued by the cloud service and the actual use of the resource, so that the invention provides a resource allocation manner of reservation before use. The method can effectively ensure that the resource opening and the quota consumption are the same thing, ensure the logical self-consistency in the quota management system, and simultaneously provide an implementation basis for some other requirements (reservation). The reservation behavior can be specifically classified into a "real-time reservation" method and a "reservation" method, where the real-time reservation method refers to that each quota management node transmits a reservation request received from a child node to a parent node in real time. According to the CAP principle, the availability of a quota management system preset in real time is reduced, corresponding response delay is increased, and the pressure of a server is increased.

The CAP principle, also called CAP theorem, refers to consistency, availability and partition fault tolerance in a distributed system, and at most two points can be realized at the same time, so that the three elements cannot be simultaneously considered. The following describes a specific procedure for real-time reservation in a multi-level quota management system with reference to fig. 6:

S301, the cloud service 105 issues a resource reservation request to the leaf node 104, the reservation request including the kind, the number of reservation resources.

S302, after receiving the resource reservation request of the cloud service 105, the leaf node 104 reserves a corresponding resource inside the node according to the content of the request.

S303, after the leaf node 104 reserves the corresponding resource inside the node, the resource reservation request of the cloud service 105 is sent to its parent node, that is, the quota management node 103.

S304, after receiving the resource reservation request of the cloud service 105, the quota managing node 103 reserves a corresponding resource inside the node.

S305, after the quota managing node 103 reserves the corresponding resource inside the node, the quota managing node sends a resource reservation request of the cloud service 105 to its parent node, that is, the quota managing node 102.

S306, after receiving the resource reservation request of the cloud service 105, the quota management node 102 reserves a corresponding resource inside the node.

S307, after the quota managing node 102 reserves the corresponding resource inside the node, it sends a resource reservation request of the cloud service 105 to its parent node, that is, the root node 101.

S308, after receiving the resource reservation request of the cloud service 105, the root node 101 reserves a corresponding resource inside the node.

S309, after reserving the corresponding resources inside the node, the root node 101 sends a command of reservation success to its child node.

S310, after receiving the preset success instruction returned by the father node, the quota management node 102 sends the preset success instruction to the child node thereof.

S311, after receiving the predetermined success instruction returned by the parent node, the quota management node 103 sends the predetermined success instruction to the child node thereof.

S312, after receiving the predetermined success instruction returned by the parent node, the leaf node 104 sends the predetermined success instruction to its child node, i.e. the cloud service 105.

Each quota managing node may first examine whether related resources can be reserved for cloud service 105 according to the quota managing information inside the node before reserving the corresponding resources inside the node. When the inspection passes, a resource reservation request of the cloud service 105 is sent to the upper parent node. When the audit is not passed, a reject instruction may be sent directly to the child node. When all quota management nodes pass the internal examination and reserve corresponding resources internally, instructions of reservation success are transmitted downwards step by root node 101 until cloud service 105.

The implementation logic of the preset behavior is that after the quota management node receives a preset application of the child node, the quota management node internally examines the quota management item (for example, whether the quota amount is enough) which is responsible for the quota management node, if the preset application of the child node is agreed, the corresponding quota amount is deducted internally, preset information is put into a cache layer, and the child node waits for sending out an actual calling application. And the preset information is quota limit preset by the agreeing child node. For example, if the quota of the leaf node 104, the quota management node 103, the quota management node 102, and the root node 101 is 100 units, 90 units, and 80 units, respectively. When the cloud service 105 sends a reservation request of 50 units of resources to the leaf node 104, the leaf node firstly checks whether the quota is enough or not, deducts the quota of 50 units inside after the checking is passed, and transfers the resource application to the father node, so that the operation is repeated to the root node 101, and after the reservation is completed, the quota of each node is changed into 50 units, 40 units and 30 units. If the cloud service 101 applies for a resource of 40 units again at this time, when a new resource reservation application is transferred to the root node 101, the root node 101 is rejected because it no longer has enough quota, so that reservation fails, and at the same time, the resource of 40 units deducted by each next-level quota management node is returned to each quota management node, and waits for reservation next time.

In one possible embodiment, all of the reservations may be provided with an expiration time. And if the cloud service does not actually call the preset resource or does not receive a preset success instruction returned by the father node after the expiration time is preset, the preset automatically expires, and the preset resource returns to the resource pool. The resource pool is a quota which can be called or preset and is inquired by each quota management node.

In one possible implementation, the expiration time of the parent node's scheduled resource is set to be later than the expiration time of its child node's scheduled resource. In this embodiment, the availability of the system may be ensured, preventing the child node from agreeing to the relevant call when the cloud service applies for actually invoking the resource to the child node, but the parent node has returned the predetermined resource to the resource pool.

The quota management system adopting the real-time reservation mode has higher consistency but lower availability, and in some businesses with low consistency requirements, the reservation mode can be adopted to improve the availability of the quota management system and reduce the running time delay of the system.

In the quota management system for executing the reservation method, each quota management node can actively send a resource reservation request to a father node in advance under the condition that a cloud service resource reservation application is not received, if the father node agrees to the resource reservation application, the relevant resource reservation application subsequent to the father node is in the range of the agreed quantity, the quota management node is regarded as receiving the agreement instruction of the father node, and the agreement instruction can be directly returned to the father node. For example, if the quota of the leaf node 104 is 300 units, it may actively issue a resource reservation application of 200 units to the quota management node 103, if the quota management node 103 agrees, data of 200 units of the quota is put into the cache layer, if the cloud service 105 issues a resource reservation request of 50 units to the leaf node 104 at this time, because 50 units <200 units and 50 units <300 units, the leaf node 103 may directly agree to the resource reservation application of the cloud service 105 without issuing a request again to the quota management leaf node 103, and deduct 50 units from each quota in the cloud service 105, that is, the remaining self quota of 250 units and the parent quota of 150 units.

In this embodiment, the pressure of the quota managing node 103 may be greatly reduced, because the leaf node 104 may reserve a larger number of resources at a time each time, and after reaching the predetermined condition, the leaf node issues a second reservation request, which greatly reduces the frequency with which the quota managing node 103 receives the reservation request. Of course, if the pressure of the quota management node 102 is greater in the actual operation, the quota management node 103 may also use the reserved predetermined logic of the leaf node 104, and even each quota management node in the entire quota management system may perform the method, so as to relieve the operation pressure of the entire system.

The following describes a specific procedure for reservation in a multi-level allocation management system with reference to fig. 7:

s401, the cloud service 105 issues a request for resource reservation to the leaf node 104, the request for resource reservation including a type and an amount of the requested reserved resource.

S402, after receiving the resource reservation request of the cloud service 105, the leaf node 104 internally reserves the relevant resource.

S403, after the internal reservation of the relevant resources is successful, the leaf node 104 returns an instruction of reservation success to the cloud service 105.

S404, after meeting the predetermined condition, the leaf node 104 actively sends a resource reservation request to its parent node, namely, the quota management node 103, where the request may include a type and a number of resources reserved.

S405, after receiving the resource reservation request of the child node, the quota managing node 103 internally reserves the related resource.

S406, after the internal reservation of the relevant resource is successful, the quota managing node 103 returns a reservation successful instruction to the leaf node 104.

S407, after the predetermined condition is satisfied, the quota managing node 103 issues a resource reservation request to its parent node, that is, the quota managing node 102, where the request may include a type and a number of resources to be reserved.

S408, after receiving the resource reservation request of the child node, the quota managing node 102 internally reserves the related resource.

S409, after the internal reservation of the relevant resource is successful, the quota managing node 102 returns a reservation success instruction to the quota managing node 103.

S410, after meeting the predetermined condition, the quota managing node 102 issues a resource reservation request to its parent node, i.e. the root node 101, where the request may include a type and amount of the requested reserved resource.

S411, after receiving the resource reservation request of the child node, the root node 101 internally reserves the relevant resource.

S412, after the internal reservation of the relevant resource is successful, the root node 101 returns a reservation successful instruction to the quota management node 102.

The predetermined condition may be that a certain threshold value (for example, 70% of the remaining quota of the node) is reached at certain time intervals (for example, 10 minutes) or by the resource predetermined by the child node.

In this embodiment, when the cloud service subscribes a quota to the leaf node each time, the leaf node does not need to forward a reservation request to the parent node each time, and only if resources reserved to the parent node are stored in the leaf node, a reservation success instruction can be directly returned to the cloud service, and the corresponding quota is deducted from the internal resource pool. The method greatly increases the usability of the system and relieves the operation pressure of the whole system.

Every time a certain period of time or the resources reserved by the child nodes reach a certain threshold value, each quota management node actively reserves a certain amount of resources to the parent node for reservation use of the child nodes in a next period of time. The more the quota managing nodes are actively scheduled to the parent, the less stress the system is, but the real-time may be reduced. It may happen that some child nodes have more resources free, but other child nodes under the same parent node have not been subscribed to new resources for this parent node.

In one possible implementation, if a child node does not apply to actually invoke a predetermined resource for a period of time, the predetermined resource expires, and the resource returns to the resource pool to wait for reallocation.

Besides the preset resource, the cloud service needs to truly allocate the resource to the cloud service side to complete the complete allocation process of the resource, wherein the allocation of the resource is simpler for the quota management system which does not execute the preset method, the cloud service can directly and actually call the applied resource after receiving the agreement instruction of the leaf node, and the actual allocation mode of the resource is complex for the quota management system which executes the preset method. In the quota management system performing the predetermined method, the actual calling method of the resource is described below with reference to fig. 8:

s501, the cloud service 105 issues an application to the leaf node 104 to use a predetermined resource.

S502, after receiving the application of the child node, inquiring whether the preset information corresponding to the child node is recorded in the child node, and after inquiring the preset information corresponding to the child node, the leaf node 104 deducts the corresponding preset resource in the child node and deletes the preset information.

S503, after deducting the corresponding predetermined resources internally, the leaf node 104 issues an instruction to use the resources to the cloud service 105.

S504, after the predetermined condition is satisfied, the leaf node 104 reports the usage information of the predetermined resource to its parent node, that is, the quota management node 103.

S505, after receiving the predetermined resource usage information reported by the leaf node 104, the quota managing node 103 deducts the corresponding predetermined resource internally, and deletes the corresponding predetermined information.

S506, after the preset condition is met, the quota management node 103 reports the use information of the preset resource to the father node, namely the quota management node 102.

S507, after receiving the predetermined resource usage information reported by the quota management node 103, the quota management node 102 deducts the corresponding predetermined resource internally and deletes the corresponding predetermined information.

S508, after the predetermined condition is satisfied, the quota managing node 102 reports the usage information of the predetermined resource to its parent node, i.e., the root node 101.

S509, after receiving the predetermined resource usage information reported by the quota management node 102, the root node 101 deducts the corresponding predetermined resource internally and deletes the corresponding predetermined information.

The predetermined condition may be that, as soon as a usage reservation application of the child node is received, the usage situation is immediately reported to the parent node, so that the parent node deducts the related reserved resource, and prevents the reserved resource from reaching the expiration time and being returned to the resource pool.

The predetermined condition may also be that a predetermined expiration time is reached. In this implementation, once the predetermined expiration time is reached, the quota management node automatically reports the information of the used and unused predetermined resources, returns the unused predetermined resources to the resource pool, and deducts the corresponding predetermined resources from the used resources. For example, if the cloud service 105 sends a 100-unit application to the leaf node 104, the leaf node 104 and the quota management node 103 each internally cache 100 units of resource reservation information. When the predetermined information of the leaf node 104 reaches the expiration time, the cloud service 105 actually applies that 60 units of resources, and 40 units of resources are actually used, the leaf node 104 recovers the quota of 40 units, and reports the information to the quota management node 103. After receiving this information, quota managing node 103 also restores the quota of 40 units, and so on up to root node 101.

In one possible implementation, the expiration time of the parent node is set to be later than the expiration time of the child node, which may ensure that the parent node's resources do not expire on its own without receiving the child node information before the child node reports the information to the parent node. For example, if the expiration time set by the quota management node 103 is 1:00, 2:00, 3:00, the expiration time (information reporting time) of the leaf node may be set to 0:59, 1:29, 2:59. Or the expiration time of the parent node is set to five minutes after receiving the child node information. The information is report information sent to the father node after the preset resource of the child node is out of date.

In the reserved quota management system, the reserved conditions may be that when a certain period of time is set or a resource reserved by a child node reaches a certain threshold value, a parent node is actively reserved with a new resource, and at the same time, information of the reserved resource which is already used is reported to the parent node.

The instruction for using the resources may be that the leaf node 104 allocates the corresponding resources to the cloud service 105, or may instruct the cloud service 105 to actively call the corresponding resources, or open the rights of the corresponding resources to the cloud service 105, or the like.

In one possible implementation, when a child node sends a usage reservation application to a parent node, but a corresponding reservation resource in the parent node has expired, the parent node returns an instruction to the child node that the usage reservation failed, and the child node steps down the formulation of the usage reservation failure until the cloud service. At this time, if the cloud service wants to use the new resource, it needs to initiate a new resource reservation application to the leaf node again.

During actual operation of the quota management system, the cloud service may have a need to query the maximum quota that it can use. The maximum quota line which can be used is the maximum quota line which can be actually called. For example, if the quota management node of the cloud service displays that the cloud service still has 100 units of maximum storage resources available for use, but the parent node of the quota management node, that is, the quota management node displays that the area for which the quota management node is responsible currently only has 80 units of resource quota, for the cloud service, the maximum quota that can be currently used is 80 units. A method of querying a multi-level quota management system for a maximum quota amount that can be used is described herein in connection with fig. 9.

S601, the cloud service 105 issues a resource query request to its parent node, i.e., the leaf node 104.

S602, after receiving the resource query request of the child node, the leaf node 104 sends the resource query request to the parent node, namely the quota management node 103.

S603, after receiving the resource query request of the child node, the quota management node 103 sends the resource query request to the parent node, namely the quota management node 102.

S604, after receiving the resource query request of the child node, the quota managing node 102 sends the resource query request to its parent node, i.e., the root node 101.

S605, after receiving the resource query request, the root node 101 returns its quota to its corresponding child node.

S606, after receiving the quota amount returned by the parent node, namely the root node 101, the quota management node 102 returns the preferred quota amount to the lower child node according to the quota amount of the own node and the quota amount of the parent node.

S607, after receiving the quota amount returned by the parent node, that is, the quota management node 102, the quota management node 103 returns the preferred quota amount to the lower child node according to the quota amount of the own node and the quota amount of the parent node.

S608, after receiving the quota amount returned by the parent node, that is, the quota management node 103, the leaf node 103 returns the preferred quota amount to the lower child node according to the quota amount of the parent node and the quota amount of the parent node.

The quota limit is a quota limit which is adaptive to the resource type managed by the quota management node. For example, the remaining quota of the area resource that the quota management node is responsible for is the quota of the quota management node. When the quota management node is a quota management node, the remaining number of the quota of cloud services which is responsible for the quota management node is the quota of the quota management node.

The preferred quota may be lower quota compared with the quota returned by the parent node, or the quota of the parent node and the quota returned by the parent node may be returned to the lower child node together.

In a possible implementation manner, the cloud service 105 may also directly send a resource query request to the root node 101, where the root node 101 may determine, according to information carried in the query request of the cloud service 105, child node information corresponding to the child node information, and return, to the child node corresponding to the child node, a quota of the root node 101 and information carried in the query request of the cloud service 105. After receiving the quota information returned by the parent node, that is, the root node 101, the quota management node 102 first determines corresponding child node information according to information carried in the query request of the cloud service 105, and returns to the child node information a preferred quota determined according to the quota received from the parent node and the quota of the parent node. The carried information should contain characteristic information of the cloud service 105, which enables the quota management node to distinguish its corresponding child node information. For example, when root node 101 receives a leaf node query request of cloud service 105, it may have multiple child nodes, including child nodes corresponding to cloud service 105, namely quota management node 102, and quota management node 10X (not labeled in the figure) not corresponding to cloud service 105. The feature information should enable the root node 101 to identify the child node to which the cloud service 105 corresponds, namely the quota managing node 102. The specific implementation of the feature information is shown in fig. 5 and described in detail with respect to the method of querying leaf nodes, and will not be described here.

In one possible implementation manner, each quota management node is configured with a corresponding cache layer, and information of a parent node quota can be cached in the cache layer. And after receiving the resource query request sent by the child node, each quota management node preferentially queries whether the quota of the parent node exists in the cache layer, and if the quota of the parent node is queried in the cache layer, the preferred quota is returned to the child node according to the quota of the parent node and the quota of the parent node.

The resource query method of the multi-hierarchy allocation management system when having a cache layer is described below with reference to fig. 10:

s701, the cloud service 105 transmits a resource query request to the leaf node 104.

S702, after receiving the resource query request of the child node, the leaf node 104 queries whether the quota of the parent node exists in the cache layer.

S703, if the leaf node 104 inquires the quota of the parent node, namely the quota management node 103, in the cache layer, the preferred quota is returned to the child node according to the quota of the parent node and the quota of the child node.

S704, if the parent node quota limit is not queried in the leaf node 104 cache layer, a resource query request is transmitted to the parent node.

S705, after receiving the resource query request of the child node, the quota managing node 103 queries whether the quota of the parent node exists in the cache layer.

S706, if the quota management node 103 caches the quota of the parent node, namely the quota management node 102, then the quota management node returns the preferred quota to the child node according to the quota of the parent node and the quota of the child node.

S707, if the parent node quota limit is not queried in the leaf node 104 cache layer, a resource query request is transmitted to the parent node.

S708, after receiving the resource query request of the child node, the quota management node 102 queries whether the quota of the parent node exists in the cache layer.

S709, if the quota management node 102 has queried the quota of the parent node, i.e. the root node 101, in the cache layer, the preferred quota is returned to the child node according to the quota of the parent node and the quota of itself.

S710, if the quota management node 102 does not query the quota of the parent node, a resource query request is transmitted to the parent node.

S711, after receiving the resource query request of the child node, the root node 101 returns the quota to the child node.

S712, after receiving the quota information returned by the father node, the quota management node 102 returns the preferred quota to the child node according to the quota returned by the father node and the self quota.

S713, after receiving the quota information returned by the father node, the quota management node 103 returns the preferred quota to the child node according to the quota returned by the father node and the self quota.

S714, after receiving the quota information returned by the father node, the leaf node 104 returns the preferred quota to the child node according to the quota returned by the father node and the quota itself.

In this embodiment, if the quota management node caches the quota of the parent node, the quota management node can directly return the preferred quota to the child node, so that each query of the quota is prevented from being required to be queried to the root node step by step, the operation pressure of the whole system is greatly reduced, and the operation efficiency of the system is improved.

In a possible implementation manner, the quota of the parent node cached by the quota management node is the preferred quota returned by the parent node and is not the quota of the parent node, so that the accuracy of resource query can be effectively improved. For example, when the quota managing node 102 receives the quota of the root node 101, the quota managing node 103 returns the preferred quota. The preferred quota is assumed to be a lower amount of the parent quota than the child quota. At this time, the quota managing node 103 caches the received preferred quota into the cache layer, and waits for the next query of the child node, i.e., the leaf node 104. When the leaf node 104 issues a resource query request to the quota managing node 103, the quota managing node 103 returns a quota amount (a preferred number) of the quota managing node 103 to the leaf node 104, which is lower in amount than a preferred quota amount accepted from the quota managing node 102. In this way, the parent node quota amount queried by the leaf node 104 is the quota amount with the lowest amount compared with the quota amounts of the three quota management nodes of the quota management node 103, the quota management node 102 and the root node 101. Similarly, the quota returned by the leaf node 104 to the cloud service 105 is the quota with the lowest amount compared with the quota of four quota management nodes, namely, the leaf node 104, the quota management node 103, the quota management node 102 and the root node 101. This way, the available quota queried by the cloud service 105 can be ensured to be the most reliable available.

One specific embodiment is described below in conjunction with fig. 11:

S1, a cloud service sends a resource use request to a cloud service deployment module;

The request should contain the kind of resources and the number of resources that the cloud service wants to use;

s2, inquiring the IP address of the first quota management node;

The query manner may be a query method provided in fig. 5, or may directly query IP address information of a first quota management node from a cache, where the first quota management node may be a leaf node directly corresponding to the cloud service. The upper node of the first quota managing node may be any upper node of the first quota managing node.

S3, sending a resource use application to the first quota management node;

the cloud service deployment module sends a resource application request to the cloud service deployment module according to the obtained IP address of the first quota management node, wherein the request contains the number of the resource types.

S4, the first quota management node sends a resource use application to the upper quota management node;

After the first quota management node receives the resource use application sent by the cloud service deployment module, whether the self resource quota meets the request of the first quota management node can be judged first, and when the self resource quota meets the request, the resource use request is forwarded to the upper quota management node.

S5, the upper quota management node of the first quota management node gives first feedback to the first quota management node;

The first feedback may be an instruction of agreement or rejection after examination, and in a certain case, the first feedback may also be the number of a specific resource quota, where the number of resource quotas may be the number of resource quotas of the upper layer quota management node itself, or the number of resource quotas returned by the upper layer quota management node to the upper layer quota management node. The basis for the upper layer quota management node to make the first feedback can be that the father node of the first quota management node returns quota information or instructions according to the quota information of the father node and the upper layer nodes of the father node. In one possible implementation, to make the first feedback, the parent node may report the resource request layer by layer up to the root node of the system, and then make the first feedback according to the instruction that the root node downloads layer by layer.

S6, giving a second feedback according to the first feedback;

When the first quota management node receives a grant or refusal instruction from the upper quota management node, the instruction is returned to the cloud service deployment module, and when the first quota management node receives a specific quota amount from the upper quota management node, the first quota management node can automatically judge whether the resource application request sent by the cloud service deployment module should be granted or refused according to the amount, and simultaneously give a second feedback.

S7, deploying cloud service according to the second feedback;

and when the second feedback is agreeing, deploying the cloud service according to the content of the feedback.

The present application also provides a computing device 1000. As shown in fig. 12, computing device 1000 includes a bus 1002, a processor 1004, a memory 1006, and a communication interface 1008. Communication between the processor 1004, memory 1006 and communication interface 1008 is via bus 1002. Computing device 1000 may be a server or a terminal device. It should be understood that the present application is not limited to the number of processors, memories in computing device 1000.

Bus 1002 may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, only one line is shown in fig. 12, but not only one bus or one type of bus. Bus 1004 may include a pathway to transfer information between various components of computing device 1000 (e.g., memory 1006, processor 1004, communication interface 1008).

The processor 1004 may include any one or more of a central processing unit (central processing unit, CPU), a graphics processor (graphics processing unit, GPU), a Microprocessor (MP), or a digital signal processor (DIGITAL SIGNAL processor, DSP).

The memory 1006 may include volatile memory (RAM), such as random access memory (random access memory). The processor 1004 may also include non-volatile memory (non-volatile memory), such as read-only memory (ROM), flash memory, mechanical hard disk (HARD DISK DRIVE, HDD) or solid state disk (SSD STATE DRIVE).

The memory 1006 has stored therein executable program code that is executed by the processor 1004 to implement the functions of the flow analysis device and the simulation modeling device, respectively, to implement the method of any of fig. 4-10. That is, the memory 1006 has instructions stored thereon for performing the method of any of the methods of FIGS. 4-10.

Communication interface 1003 enables communication between computing device 1000 and other devices or communication networks using a transceiver module such as, but not limited to, a network interface card, transceiver, or the like.

The embodiment of the application also provides a computing device cluster. The cluster of computing devices includes at least one computing device. The computing device may be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device may also be a terminal device such as a desktop, notebook, or smart phone.

As shown in fig. 13, the cluster of computing devices includes at least one computing device 1000. The memory 1006 in one or more computing devices 1000 in the cluster of computing devices may have stored therein the same instructions for performing the method of any of figures 4-10.

In some possible implementations, some of the instructions for performing the methods shown in any of FIGS. 4-10 may also be stored in the memory 1006 of one or more computing devices 1000 in the computing device cluster, respectively. In other words, a combination of one or more computing devices 1000 may collectively execute instructions for performing the methods illustrated in any of figures 4-10.

It should be noted that the memory 1006 in different computing devices 1000 in the computing device cluster may store different instructions for performing part of the functions of the simulation system 600. That is, the instructions stored by the memory 1006 in the different computing devices 1000 may implement the functionality of one or more modules in the flow analysis apparatus, simulation modeling apparatus.

In some possible implementations, one or more computing devices in a cluster of computing devices may be connected through a network. Wherein the network may be a wide area network or a local area network, etc. Fig. 14 shows one possible implementation. As shown in fig. 14, two computing devices 1000A and 1000B are connected by a network. Specifically, the connection to the network is made through a communication interface in each computing device. In this type of possible implementation, instructions to perform the functions of the flow analysis device are stored in memory 1006 in computing device 1000A. Meanwhile, instructions of the simulation modeling apparatus are stored in the memory 1006 in the computing device 1000B.

It should be appreciated that the functionality of computing device 1000A shown in fig. 14 may also be performed by multiple computing devices 1000. Likewise, the functionality of computing device 1000B may also be performed by multiple computing devices 1000.

Embodiments of the present application also provide a computer program product comprising instructions. The computer program product may be software or a program product containing instructions capable of running on a computing device or stored in any useful medium. The computer program product, when run on at least one computing device, causes the at least one computing device to perform the method of any of figures 4-10.

The embodiment of the application also provides a computer readable storage medium. The computer readable storage medium may be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk), etc. The computer-readable storage medium includes instructions that direct a computing device to perform the method of any of figures 4-10.

It should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention, and not for limiting the same, and although the present invention has been described in detail with reference to the above-mentioned embodiments, it should be understood by those skilled in the art that the technical solution described in the above-mentioned embodiments may be modified or some technical features may be equivalently replaced, and these modifications or substitutions do not make the essence of the corresponding technical solution deviate from the protection scope of the technical solution of the embodiments of the present invention.

Claims (17)

1. A method of resource management, characterized in that a quota management system manages cloud computing resources for deploying cloud services, the cloud computing resources being provided by a plurality of servers, the quota management system comprising a plurality of levels of quota management nodes, the levels of quota management nodes comprising a plurality of first quota management nodes and a superior quota management node of the first quota management nodes, the method comprising:

The cloud service deployment module receives a request for deploying cloud service;

The cloud service deployment module determines an Internet Protocol (IP) address of the first quota management node;

The cloud service deployment module sends a resource application request to the first quota management node according to the IP address of the first quota management node, wherein the resource application request carries a target resource quota of cloud computing resources required for deploying the cloud service;

The first quota management node applies for cloud computing resources of the target resource quota to an upper level quota management node of the first quota management node;

the first quota management node receives feedback of the upper quota management node;

If the feedback is a consent application, the first quota management node provides cloud computing resources of the target resource quota to the cloud service deployment module;

the cloud service deployment module deploys the cloud service by utilizing cloud computing resources of the target resource quota.

2. The method according to claim 1, characterized in that the method comprises:

And if the feedback is not agreeing to apply, the first quota management node feeds back an application result to the cloud service deployment module.

3. The method according to any of claims 1-2, wherein the quota of at least one quota managing node is smaller than the sum of the quota of all its child nodes.

4. The method of any of claims 1-3, wherein the resource application comprises one or more of a resource usage application, a resource reservation application.

5. The method of any of claims 1-4, wherein the quota managing node is configured with a cache that stores quota managing information for an upper level quota managing node, or wherein the cache is configured with a cache that stores quota managing information for an upper level quota managing node

The cloud service deployment module is configured with a cache, and the cache stores one or more of IP address information of the first quota management node and quota management information of the first quota management node.

6. The method of claim 5, wherein the cache of the cloud service deployment module includes IP address information of the first quota managing node and, correspondingly,

The cloud service deployment module determining an internet protocol, IP, address of the first quota managing node includes:

the cloud service deployment module determines an Internet Protocol (IP) address of the first quota management node according to the cached IP address information of the first quota management node.

7. The method of claim 5, wherein the cache of the first quota managing node contains a quota returned by the superordinate quota managing node.

8. A quota management system for deploying cloud computing resources of a cloud service, the cloud computing resources being provided by a plurality of servers, the quota management system comprising a plurality of levels of quota management nodes, the levels of quota management nodes comprising a plurality of first quota management nodes and a top level quota management node of the first quota management nodes, the quota management system comprising:

The cloud service deployment module is used for receiving a request for deploying the cloud service, determining an Internet Protocol (IP) address of the first quota management node according to the request, sending a resource application request to the first quota management node according to the IP address of the first quota management node, wherein the resource application request carries a target resource quota of cloud computing resources required for deploying the cloud service, and deploying the cloud service by utilizing the cloud computing resources of the target resource quota after receiving the cloud computing resources of the target resource quota provided by the first quota management node;

The first quota management node is configured to apply for the cloud computing resource of the target resource quota to a superior quota management node of the first quota management node, receive feedback from the superior quota management node, and if the feedback is an approval application, provide the cloud computing resource of the target resource quota to the cloud service deployment module by the first quota management node.

9. The system of claim 8, wherein the first quota management node is further configured to:

And if the feedback is not agreeing to apply, the first quota management node feeds back an application result to the cloud service deployment module.

10. The system of any of claims 8-9, wherein at least one quota managing node has a quota amount that is less than a sum of quota amounts of all its child nodes.

11. The system of any of claims 8-10, wherein the resource application comprises one or more of a resource usage application, a resource reservation application.

12. The system of any of claims 8-11, wherein the quota managing node is configured with a cache that stores quota managing information for an upper level quota managing node, or wherein the cache is configured to store quota managing information for an upper level quota managing node

The cloud service deployment module is configured with a cache, and the cache stores one or more of IP address information of the first quota management node and quota management information of the first quota management node.

13. The system of claim 12, wherein the cache of the cloud service deployment module includes IP address information of the first quota managing node and, correspondingly,

The cloud service deployment module determining an internet protocol, IP, address of the first quota managing node includes:

the cloud service deployment module determines an Internet Protocol (IP) address of the first quota management node according to the cached IP address information of the first quota management node.

14. The system of claim 12, wherein the cache layer of the quota managing node includes a quota amount returned by its quota managing superordinate quota managing node.

15. A cluster of computing devices, comprising at least one computing device, each computing device comprising a processor and a memory;

the processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method of any of claims 1-7.

16. A computer program product containing instructions that, when executed by a cluster of computing devices, cause the cluster of computing devices to perform the method of any of claims 1-7.

17. A computer readable storage medium comprising computer program instructions which, when executed by a cluster of computing devices, perform the method of any of claims 1-7.