TWI755415B - Method, device and system for application cluster resource allocation in cloud environment - Google Patents

Abstract

Embodiments of the present invention provide a method, device and system for allocating application cluster resources in a cloud environment. The method includes: acquiring the resource utilization rate of each application cluster in a specified transaction link in the cloud environment; Determine the resource load level of each application cluster; the resource load level includes a high occupation level and a low occupation level; apply for resources for the high occupation level application cluster; if the application cluster for the high occupation level fails to apply for resources, then Release the resources of the low-occupancy level application cluster according to the resource load level; and allocate the released resources to the high-occupancy level application cluster. The embodiments of the present invention can level the resource load of the application cluster in the entire cloud environment, make full use of existing resources, improve the resource utilization efficiency of the application cluster, and enable the application cluster to support higher access traffic.

Description

Method, device and system for application cluster resource allocation in cloud environment

The invention relates to the technical field of data processing, in particular to a method for allocating application cluster resources in a cloud environment, a device for allocating application cluster resources in a cloud environment, and a system for allocating application cluster resources in a cloud environment.

Cloud computing is a computing model that provides dynamically scalable virtual resources as a service through the Internet. In this way, shared software and hardware resources and information can be provided to computers and other devices on demand. The basic environment of cloud computing is virtualization, and there is resource sharing in the virtualized environment. A virtual machine (VM, Virtual Machine) is often used in a cloud computing environment (hereinafter referred to as a cloud environment), and these virtual machines are mainly associated with system resources and the like.

These resources in the cloud environment are shared by virtual machines. When a virtual machine needs to apply for resources, it will apply for idle resources from the cloud environment. However, in the case of limited resources, if the idle resources are exhausted, the virtual machine fails to apply for resources, and the virtual machine cannot work normally due to the excessive resource load, which greatly affects the work efficiency.

In view of the above problems, embodiments of the present invention are proposed to provide a method for allocating application cluster resources in a cloud environment, an apparatus for allocating application cluster resources in a cloud environment, and a cloud environment that overcome the above problems or at least partially solve the above problems The system under which cluster resource allocation is applied.

In order to solve the above problem, an embodiment of the present invention discloses a system for allocating application cluster resources in a cloud environment. The system includes: one or more processors; a memory; and one or more modules, the one or more modules A module is stored in the memory and configured to be executed by the one or more processors, and the one or more modules have the following functions: obtain each application in a specified transaction link in a cloud environment The resource usage rate of the cluster; the resource load level of each application cluster is determined according to the resource usage rate; the resource load level includes a high occupation level and a low occupation level; apply for resources for the application cluster with the high occupation level; If the application cluster of the high occupation level fails to apply for resources, the resources of the application cluster of the low occupation level are released according to the resource load level; the released resources are allocated to the application cluster of the high occupation level.

The embodiment of the present invention also discloses a method for allocating application cluster resources in a cloud environment, including: acquiring the resource utilization rate of each application cluster in a designated transaction link in the cloud environment; determining the resource utilization rate of each application cluster according to the resource utilization rate Load level; the resource load level includes a high occupancy level and a low occupancy level; apply for resources for the application cluster with the high occupancy level; if the application cluster for the high occupancy level fails to apply for resources, release according to the resource load level resources of an application cluster with a low occupancy level; allocating the released resources to the application cluster with a high occupancy level.

Preferably, the step of acquiring the resource utilization ratio of each application cluster in the designated transaction link in the cloud environment includes: when the access traffic increases, acquiring the resource utilization ratio of each application cluster in the designated transaction link in the cloud environment.

Preferably, the cloud environment includes a resource pool, and the step of applying for resources for an application cluster with a high occupancy level includes: applying for resources from the resource pool; when the resources in the resource pool are sufficient, the An application cluster with a high occupancy level allocates resources in the resource pool; when the resources in the resource pool are exhausted, it fails to apply for resources for the application cluster with a high occupancy level.

Preferably, the application clusters have a corresponding priority order, and the step of allocating resources in the resource pool to the application clusters with high occupancy levels includes: when the number of application clusters with high occupancy levels is two or more At the time, the resources in the resource pool are allocated to the application clusters with high occupancy levels according to the priority order.

Preferably, the resource load level has a corresponding release ratio, and the step of releasing the resources of the low-occupancy level application cluster according to the resource load level includes: determining the release ratio corresponding to the resource load level of the low-occupancy level application cluster. ; release the resources of the low-occupancy level application cluster according to the release ratio; save the released resources into the resource pool.

Preferably, the step of releasing the resources of the low-occupancy level application cluster according to the release ratio includes: acquiring the current resources of the low-occupancy level application cluster; acquiring current access to the low-occupancy level application cluster. traffic; use the current access traffic and the preset stand-alone limit value to calculate demand resources; use the current resources and the demand resources to determine the idle resources of the low-occupancy level application cluster; release according to the release ratio the free resource.

Preferably, the step of allocating the released resources to the high occupancy level application cluster includes: applying for resources from the resource pool storing the released resources; allocating the high occupancy level application cluster to the resource pool. The resources released in the resource pool.

The embodiment of the present invention also discloses a device for allocating application cluster resources in a cloud environment, including: a resource utilization rate acquisition module, which is used to acquire the resource utilization rate of each application cluster in a designated transaction link in the cloud environment; the resource load level is determined a module for determining the resource load level of each application cluster according to the resource utilization rate; the resource load level includes a high occupation level and a low occupation level; a resource application acquisition module is used for the application of the high occupation level The cluster applies for resources; the resource release module is used to release the resources of the low-occupancy level application cluster according to the resource load level if the application for resources for the high-occupancy level application cluster fails; the resource allocation module is used to The released resources are allocated to the application clusters with high occupancy levels.

The embodiment of the present invention also discloses a method for resource allocation of application clusters in a cloud environment, including: acquiring the resource load status of each application cluster in the cloud environment; determining the resource load level of the application cluster according to the resource load status; The resource load level includes the resource full load level and other levels; apply for resources for the application cluster of the resource full load level; if the application cluster of the resource full load level fails to apply for resources, then release other levels of application clusters according to the resource load level resources; allocating the released resources to the application clusters of the resource full load level.

Preferably, the step of acquiring the resource load status of each application cluster in the cloud environment includes: when the access traffic increases, acquiring the resource load status of each application cluster in the cloud environment.

Preferably, the cloud environment includes a resource pool, and the step of applying for resources for an application cluster with a full resource load level includes: applying for resources from the resource pool; when the resources in the resource pool are sufficient, the The application cluster at the full resource level allocates resources in the resource pool; when the resources in the resource pool are exhausted, the application cluster at the full resource level fails to apply for resources.

Preferably, the application clusters have a corresponding priority order, and the step of allocating resources in the resource pool to the application clusters at the full resource load level includes: when the number of application clusters at the full resource load level is two or more At the time, the resource in the resource pool is allocated to the application cluster of the resource full load level according to the priority order.

Preferably, the resource load level has a corresponding release ratio, and the step of releasing the resources of the application clusters of other levels according to the resource load level includes: determining the release ratio corresponding to the resource load levels of the application clusters of other levels; The resources of the application clusters at other levels are released according to the release ratio; and the released resources are saved in the resource pool.

Preferably, the step of releasing the resources of the other application clusters according to the release ratio includes: acquiring the current resources of the application clusters of the other levels; acquiring the current access traffic of the application clusters of the other levels; The current access traffic and the preset stand-alone limit value are used to calculate the demanded resources; the current resources and the demanded resources are used to determine the idle resources of the application clusters of other levels; the idle resources are released according to the release ratio.

Preferably, the step of allocating the released resources to the application cluster at the resource full level includes: applying for resources from the resource pool storing the released resources; allocating the resource to the application cluster at the resource full level. The resources released in the resource pool.

The embodiment of the present invention also discloses an apparatus for allocating application cluster resources in a cloud environment, including: a resource load state acquisition module, used to acquire the resource load state of each application cluster in the cloud environment; a resource load level determination module, used for The resource load level of the application cluster is determined according to the resource load state; the resource load level includes a resource full load level and other levels; a resource application module is used to apply for resources for the application cluster of the resource full load level; resource release The module is used to release the resources of the application clusters of other levels according to the resource load level when the application cluster fails to apply for the resource at the full load level of the resources; the release resource allocation module is used to allocate the released resources to The application cluster of the resource full load level.

The embodiment of the present invention includes the following advantages: for the application cluster in the cloud environment, the resource load level of each application cluster is determined according to the resource load status of each application cluster, and if there is an application cluster with a full resource load level, it is the The application cluster applies for resources. If the application for resources fails at this time, it requests other levels of application clusters to release resources, and finally allocates the released resources to application clusters at the full resource level. It should be noted that, in the embodiment of the present invention, resources are released according to the resource full load level of the application cluster. Therefore, if the above steps are repeatedly performed, the resource load of the application cluster in the entire cloud environment will eventually be leveled, and existing resources can be fully utilized to improve application performance. The resource utilization efficiency of the cluster enables the application cluster to support higher access traffic.

301‧‧‧Resource load status acquisition module

302‧‧‧Resource Load Level Determination Module

303‧‧‧Resource Application Module

304‧‧‧Resource Release Module

305‧‧‧Release resource allocation module

400‧‧‧Server

422‧‧‧CPU

426‧‧‧Power

430‧‧‧Storage media

432‧‧‧Memory

441‧‧‧Operating System

442‧‧‧Apps

444‧‧‧Data

450‧‧‧Wired or wireless network interface

456‧‧‧Keyboard

458‧‧‧Input and output interface

FIG. 1 is a flow chart of steps of Embodiment 1 of a method for allocating application cluster resources in a cloud environment of the present invention; FIG. 2 is a schematic diagram of resource allocation of an application cluster in a cloud environment of the present invention; FIG. 3 is a cloud environment of the present invention. Fig. 4 is a structural block diagram of an apparatus embodiment of an application cluster resource allocation in a cloud environment according to the present invention; Fig. 5 is a structure of a server embodiment of the present invention Schematic.

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

In the cloud environment, it includes one or more application clusters, which can independently complete specific services. The interaction between application clusters forms a transaction link, which can complete a complete transaction. For example, in an application cluster of a supply chain, there are four application systems including commodities, logistics, authentication, and marketing, and a series of tasks such as commodities, logistics, authentication, and marketing are completed in the application cluster of this supply chain. The transaction link stress test can be used to test the processing efficiency of the transaction link of the application cluster.

The transaction link stress test uses automated testing tools to simulate a variety of normal, peak and abnormal load conditions to stress test various performance indicators of the application cluster, detect the performance bottleneck of the application cluster, and improve the processing capacity of the application cluster.

However, the current transaction link stress test only tests the performance bottleneck of the application cluster, and when the access traffic reaches the target value, such as 15W of access traffic, the processing capacity of the application cluster is no longer detected upwards, so the resources cannot be fully utilized. stand up.

In order to solve the above problems, the embodiment of the present invention, for the application cluster in the cloud environment, during the stress test process on the designated platform, evens out the resource load of the overall application cluster of the transaction link, and makes full use of the existing resources and balances the existing resources when the existing resources are limited. The resource load of each application cluster improves the processing capability of the application cluster and enables the application cluster to support higher access traffic.

Referring to FIG. 1, it shows a flow chart of steps of a method embodiment of resource allocation of application clusters in a cloud environment according to the present invention, which may specifically include the following steps: Step 101: Obtain the resource load status of each application cluster in the cloud environment; In the embodiment of the present invention, the application cluster of a certain transaction link is sorted out first, and the application cluster is connected to a designated platform to form a cloud environment. In this cloud environment, the application cluster can be planned as a whole.

As shown in Figure 2, a transaction link includes application A cluster, application B cluster, and application C cluster. These application clusters are connected to a specified platform, and the application clusters on the platform can share resources.

The resource load status of the application cluster refers to the resource utilization rate of the application cluster. Specifically, the resources can include capacity, CPU, memory, disk I/O, network bandwidth, and so on. In some cases, the resource utilization of the application cluster can reach 100%.

In an embodiment of the present invention, the step 101 may include the following sub-steps: sub-step S11, when the access traffic increases, the resource load status of each application cluster in the cloud environment is acquired.

The embodiment of the present invention detects the performance bottleneck of the current application cluster by performing pressure measurement on the transaction link. The so-called performance bottleneck is when the resource usage reaches a dangerous threshold, at which point the application cluster will not work properly.

In the process of stress testing the transaction link, the user's access traffic to the application cluster can be continuously increased. In one case, the embodiment of the present invention will acquire the resource load status of each application cluster in the cloud environment when detecting that the access traffic increases.

Step 102: Determine the resource load level of the application cluster according to the resource load state; the resource load level includes the resource full load level and other levels; in this embodiment of the present invention, for each application cluster, the resource load level is state to determine the resource load level it is in, and the resource load level can be used to represent the resource usage of the application cluster.

The resource load level can include the resource full load level and other levels. The resource full load level means that the resource usage of the application cluster has reached a dangerous threshold, for example, the CPU usage may reach 90%.

As shown in Figure 2, if the CPU usage of application A cluster is greater than 70% (CPU>70%), or if the relationship between the average load in 1 minute and the number of CPU cores is load1>core_num*1.5, it can be considered that the application Cluster A is at the full resource level, while application B cluster and application C cluster are at other levels.

It should be noted that, in the embodiment of the present invention, the number of levels of resource load levels may be divided according to specific requirements, which is not limited in the embodiment of the present invention. In addition, the division method of the resource load level in this embodiment of the present invention is not limited to the CPU usage rate, but can also be divided in combination with other resource parameters, such as the usage rate of memory, disk I/O, network bandwidth, etc., or It may be divided in other manners, which is also not limited in this embodiment of the present invention.

Step 103: Apply for resources for the application cluster with the full resource load level; in the cloud environment, a resource pool may be included, and idle resources are placed in the resource pool. When the application cluster needs to apply for resources, it will first apply for resources from the resource pool .

In an embodiment of the present invention, the step 103 may include the following sub-steps: sub-step S21, applying for resources from the resource pool; sub-step S22, when the resources in the resource pool are sufficient, the resources are The application cluster at the full load level allocates resources in the resource pool; in sub-step S23, when the resources in the resource pool are exhausted, the application cluster at the full resource load level fails to apply for resources.

It is understandable that if the resources in the resource pool are sufficient, the resources in the resource pool can be directly allocated to the application cluster at the full resource level. However, if the resources in the resource pool are exhausted, the resource cannot be applied for the application cluster at the full resource level. , the resource request fails.

In an embodiment of the present invention, the application clusters may have a corresponding priority order, and the sub-step S22 may include the following sub-steps: sub-step S221, when the number of application clusters at the full resource load level is two or two In the above case, the resources in the resource pool are allocated to the application clusters at the full resource load level according to the priority order.

When there are resources that can be allocated in the resource pool, the resources can be allocated to high-priority application clusters. Specifically, if multiple application clusters have performance bottlenecks at the same time, that is, when multiple application clusters with full resource load levels appear at the same time, resources will be allocated to each application cluster in order of priority, and application clusters with high priority will be given priority. Apply for resources, and then apply for resources for low-priority application clusters. Through this allocation principle, the host resources of the application clusters can be fully utilized.

Conversely, if resources are not allocated to each application cluster in order of priority, it may lead to the exhaustion of resources in the resource pool after applying for resources for application clusters of low priority, and finally the application clusters of high priority will not be able to be allocated. resource allocation.

It should be noted that the priority order of application clusters in this embodiment of the present invention can be set by relevant personnel. Of course, the priority order of application clusters can also be set according to the importance of the application clusters or the processing order. The embodiments of the invention do not limit this.

Step 104: If the application for resources for the application cluster with the full resource load level fails, the resources of the application clusters of other levels are released according to the resource load level; in the embodiment of the present invention, when the application for resources from the resource pool fails, other application clusters will be released resources to make up for the current shortage of resources.

In an embodiment of the present invention, the resource load level may have a corresponding release ratio, and the step 104 may include the following sub-steps: sub-step S31, determine the release ratio corresponding to the resource load level of the application cluster of other levels; In sub-step S32, the resources of the application clusters of other levels are released according to the release ratio; and in sub-step S33, the released resources are saved in the resource pool.

In the embodiment of the present invention, when the resources of the application cluster are released, the resources are released according to the resource load level of the application cluster. That is, if the application cluster is heavily loaded, less resources are released. Small application clusters release more resources.

In an embodiment of the present invention, the sub-step S32, that is, the step of releasing the resources of the other application clusters according to the release ratio includes: sub-step S321, acquiring the current resources of the application clusters of the other levels ; Sub-step S322, obtaining the current access traffic of the application clusters of the other levels; Sub-step S323, using the current access traffic and the preset stand-alone limit value to calculate the required resources; Sub-step S324, using the current resources and all The demand resources are used to determine the idle resources of the application clusters of other levels; sub-step S325, the idle resources are released according to the release ratio.

In the embodiment of the present invention, the resources required by the current application cluster may be: the total access traffic per second of the current application cluster/the extreme value of the daily stress test single machine, wherein the daily stress test single machine extreme value refers to a single virtual machine in the application cluster the traffic that can be tolerated. According to the current resources and required resources of the application cluster, idle resources that can be released by the application cluster can be determined.

In another embodiment, the resources required by the current application cluster may also be: the total access flow per second of the current application cluster/the extreme value of daily pressure measurement of a single machine/safe water level. The safe water level is a buffer, and the size can be set or not used according to actual needs, which is not limited in the embodiment of the present invention.

As shown in Figure 2, the resource load level of application B cluster is high, the resource usage rate is high, and the CPU usage rate is greater than 50% and less than 60%, then application B cluster only needs to release 5% of the resources, and application C cluster resources If the load level is low, the resource usage rate is low, and the CPU usage rate is greater than 10% and less than 20%, the application C cluster needs to release 15% of the resources. The released resources can be saved in the T4 resource pool. Through the gradient release of resources, the resources of each application cluster can be maintained at a certain level, rather than one size fits all.

Step 105: Allocate the released resources to the application clusters at the resource full load level.

In an embodiment of the present invention, the step 105 may include the following sub-steps: sub-step S41, applying for resources from the resource pool storing the released resources; sub-step S42, for the application cluster of the resource full load level Allocate the resources released in the resource pool.

After the released resources are stored in the resource pool, the application cluster at the full resource level can continuously apply for resources from the resource pool until it meets its own needs, and then it can stop the application clusters at other levels from releasing idle resources.

Specifically, the resource demand of an application cluster with a full resource load level can be determined by determining the number of predicted instances (virtual machine resources) according to the access traffic, and then subtracting the number of predicted instances from the current number of instances (current virtual machine resources), That is, the resource demand of the application cluster.

Of course, the resource requirement of the application cluster at the resource-full level may be to reduce the application cluster from the resource-full level to other levels.

After the embodiment of the present invention completes allocating resources to the application clusters at the full resource load level, the user's access traffic can be continuously increased, and steps 101 to 105 are performed again. If the process is repeated, the resource load of the application clusters of the entire transaction link will eventually be leveled. state, so that the resources of each application cluster can be fully utilized.

In this embodiment of the present invention, for an application cluster in a cloud environment, the resource load level of each application cluster is determined according to the resource load status of each application cluster. If there is an application cluster with a full resource load level, resources are applied for the application cluster. If the application for resources fails, the application clusters at other levels are requested to release the resources, and finally the released resources are allocated to the application clusters at the resource-full level. It should be noted that, in the embodiment of the present invention, resources are released according to the resource full load level of the application cluster. Therefore, if the above steps are repeatedly performed, the resource load of the application cluster in the entire cloud environment will eventually be leveled, and existing resources can be fully utilized to improve application performance. The resource utilization efficiency of the cluster enables the application cluster to support higher access traffic.

Referring to FIG. 3 , a flow chart of steps of Embodiment 2 of a method for allocating application cluster resources in a cloud environment of the present invention is shown, which may specifically include the following steps: Step 201 : Obtain each application cluster in a designated transaction link in a cloud environment In step 202, the resource load level of each application cluster is determined according to the resource utilization rate; Applying for resources; Step 204, if the application for resources for the application cluster with the high occupancy level fails, release the resources of the application cluster with the low occupancy level according to the resource load level; Step 205, allocate the released resources to the high occupancy level. Occupancy level application cluster.

In a preferred embodiment of the present invention, the step 201 may include: when the access traffic increases, acquiring the resource utilization rate of each application cluster in the designated transaction link in the cloud environment.

In a preferred embodiment of the present invention, the cloud environment includes a resource pool, and the step 203 may include: applying for resources from the resource pool; when the resources in the resource pool are sufficient, the high The application cluster of the occupancy level allocates resources in the resource pool; when the resources of the resource pool are exhausted, the application cluster of the high occupancy level fails to apply for resources.

In a preferred embodiment of the present invention, the application clusters have corresponding priorities, and the step of allocating resources in the resource pool to the application clusters with a high occupation level includes: when the application clusters with a high occupation level When there are two or more than two, the resources in the resource pool are allocated to the application clusters with high occupancy levels according to the priority order.

In a preferred embodiment of the present invention, the resource load level has a corresponding release ratio, and the step 204 may include: determining a release ratio corresponding to the resource load level of the application cluster with a low occupancy level; according to the release ratio to release the resources of the low-occupancy level application cluster; and save the released resources into the resource pool.

In a preferred embodiment of the present invention, the step of releasing the resources of the low-occupancy level application cluster according to the release ratio includes: acquiring the current resources of the low-occupancy level application cluster; acquiring the low-occupancy level application cluster The current access traffic of the application cluster of the level; use the current access traffic and the preset single-machine limit value to calculate the demanded resources; use the current resources and the demanded resources to determine the idle resources of the low-occupancy level of the application cluster; The idle resources are released according to the release ratio.

In a preferred embodiment of the present invention, the step 205 includes: applying for resources from the resource pool storing the released resources; allocating the resources released in the resource pool to the application clusters with high occupation level.

In a specific implementation, a cloud environment can be divided into multiple transaction links to process various transactions in a targeted manner. In a cloud environment, the embodiments of the present invention can be applied to make overall planning for the resources of the application cluster.

In the embodiment of the present invention, for an application cluster in a cloud environment, the resource load status of the application cluster can be known according to the resource utilization rate of each application cluster, so that the resource load level of the application cluster can be determined. If there is an application cluster with a high occupancy level, apply for resources for the application cluster. If the resource application fails at this time, request the application cluster with a low occupancy level to release resources, and finally allocate the released resources to an application cluster with a high occupancy level. It should be noted that, in the embodiment of the present invention, resources are released according to the resource full load level of the application cluster. Therefore, if the above steps are repeatedly performed, the resource load of the application cluster in the entire cloud environment will eventually be leveled, and existing resources can be fully utilized to improve application performance. The resource utilization efficiency of the cluster enables the application cluster to support higher access traffic.

To enable those skilled in the art to better understand the embodiments of the present invention, the following describes the stress measurement process of the transaction link in the embodiments of the present invention through specific examples, as shown in FIG. 2 .

1. Sort out the application clusters of the entire transaction link and connect to the platform; under this platform, there are application A cluster, application B cluster, and application C cluster, and these application clusters form a transaction link.

2. Perform stress testing on the transaction link to detect the performance bottleneck of the application cluster under the transaction link. The resource load level is divided for each application cluster according to the resource load status, and the application cluster with performance bottleneck will be divided into the resource full load level.

3. When a performance bottleneck occurs in an application cluster, such as application A cluster, it will apply for resources from the resource pool; if the application for resources from the resource pool is successful, it will continue to increase the pressure, that is, increase the application cluster resources under the transaction link. User visits.

There is a principle when applying for machines (resources): if multiple application clusters have performance bottlenecks at the same time, resources are allocated according to the priority order of each application cluster. First, apply for resources for the application cluster with high priority, followed by low priority. The application cluster of the application cluster applies for resources to achieve full utilization of the host resources; if this principle is not followed, it may lead to the resource allocation of the application cluster in the low priority order. Sequential application cluster allocation of resources.

4. As the application cluster of the transaction link continues to experience high pressure, and when the pressure is raised, it immediately applies for resources to support the current access traffic of the application cluster. Until a performance bottleneck occurs in an application cluster and the application for resources fails (the resource pool is exhausted). The platform automatically traverses the resource load status of application clusters on other links. Each application cluster releases resources in a gradient to level the resource load level according to its own resource load status.

5. After leveling the water level, resources are released and stored in the resource pool. Application cluster A with a performance bottleneck continues to apply for resources from the resource pool until it obtains its own resource requirements, and stops other low-load application clusters to release idle resources. At this time, continue to touch high pressure, that is, increase the user's access traffic, and iterate again from step 1.

6. Through the iteration of the above 5 steps, it is finally possible to fully utilize the existing resources of the application cluster of the transaction link, so that the transaction link can support the access traffic of larger users.

It should be noted that, for the sake of simple description, the method embodiments are described as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the described action sequences, because According to embodiments of the present invention, certain steps may be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

Referring to FIG. 4, there is shown a structural block diagram of an embodiment of an apparatus for allocating application cluster resources in a cloud environment according to the present invention. Specifically, it may include the following modules: a resource load state acquisition module 301, which is used to acquire various modules in the cloud environment. Resource load status of the application cluster; in an embodiment of the present invention, the resource load status acquisition module 301 may include the following sub-modules: a resource load status acquisition sub-module, used to acquire the cloud when access traffic increases Resource load status of each application cluster in the environment.

The resource load level determination module 302 is used to determine the resource load level of the application cluster according to the resource load state; the resource load level includes the resource full load level and other levels; the resource application module 303 is used for the The application cluster at the resource-full level applies for resources; In an embodiment of the present invention, the cloud environment may include a resource pool, and the resource application module 303 may include the following sub-modules: a first resource pool resource application sub-module group, used to apply for resources from the resource pool; a first resource processing sub-module, used to allocate resources in the resource pool to the application cluster at the full resource level when the resources in the resource pool are sufficient; The second resource processing sub-module is configured to fail to apply for resources for the application cluster with the full resource load level when the resources in the resource pool are exhausted.

In an embodiment of the present invention, the application cluster may have a corresponding priority order, and the first resource processing sub-module includes: a priority order processing unit configured to have two application clusters at the resource full load level Or when there are more than two, the resources in the resource pool are allocated to the application clusters at the full resource load level according to the priority order.

The resource release module 304 is used to release the resources of the application clusters of other levels according to the resource load level when applying for resources for the application cluster of the resource full load level fails; in an embodiment of the present invention, the resource load The level may have a corresponding release ratio, and the resource release module 304 may include the following submodules: a release ratio determination submodule to determine the release ratio corresponding to the resource load levels of application clusters of other levels; a resource ratio release submodule The module is used to release the resources of the application clusters of other levels according to the release ratio; the release resource saving sub-module is used to save the released resources into the resource pool.

In an embodiment of the present invention, the resource proportional release sub-module may include the following units: a current resource acquisition unit, used to acquire the current resources of the application clusters at other levels; a current access traffic unit, used to acquire all the current access traffic of the application clusters of the other levels; the demanded resource calculation unit is used to calculate the demanded resources by using the current access traffic and the preset single-machine limit value; the idle resource determination unit is used to use the current resources and the The idle resources of the application clusters of other levels are determined by requesting resources; the idle resource releasing unit is used for releasing the idle resources according to the releasing ratio.

The release resource allocation module 305 is used for allocating the released resources to the application clusters of the resource full load level.

In an embodiment of the present invention, the released resource allocation module 205 may include the following sub-modules: a second resource pool resource application sub-module for applying for resources to the resource pool that holds the released resources; The third resource processing sub-module is used for allocating the resources released in the resource pool to the application cluster at the full resource load level.

As for the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant part may refer to the partial description of the method embodiment.

An embodiment of the present invention also provides a system for allocating application cluster resources in a cloud environment, the system includes: one or more processors; a memory; and one or more modules, wherein the one or more modules are Stored in the memory and configured to be executed by the one or more processors, the one or more modules have the following functions: obtain the resource load status of each application cluster in the cloud environment; Resource load status to determine the resource load level of the application cluster; the resource load level includes the resource full load level and other levels; apply for resources for the application cluster with the resource full load level; If the resource fails, the resources of the application clusters of other levels are released according to the resource load level; the released resources are allocated to the application clusters of the resource full load level.

Optionally, the one or more modules may have the following function: when the access traffic increases, obtain the resource load status of each application cluster in the cloud environment.

Optionally, the cloud environment includes a resource pool, and the one or more modules may have the following functions: apply for resources from the resource pool; when the resources in the resource pool are sufficient, the resource full load level Allocating resources in the resource pool to the application cluster of the resource pool; when the resources of the resource pool are exhausted, the application cluster of the resource-full level fails to apply for resources.

Optionally, the application cluster has a corresponding priority order, and the one or more modules may have the following functions: when there are two or more application clusters at the full resource load level, according to the priority order, The application cluster at the resource-full level allocates the resources in the resource pool.

Optionally, the resource load level has a corresponding release ratio, and the one or more modules may have the following functions: determine the release ratio corresponding to the resource load level of the application cluster of other levels; release all the resources according to the release ratio. resources of application clusters at other levels; and save the released resources in the resource pool.

Optionally, the one or more modules may have the following functions: obtain the current resources of the application clusters of the other levels; obtain the current access traffic of the application clusters of the other levels; adopt the current access traffic and preset Calculate the demanded resources by using the single-machine limit value; use the current resources and the demanded resources to determine the idle resources of the application clusters of other levels; release the idle resources according to the release ratio.

Optionally, the one or more modules may have the following functions: apply for resources from the resource pool that holds the released resources; allocate the resources released in the resource pool to the application cluster at the resource-full level.

The embodiment of the present invention also provides an apparatus for allocating application cluster resources in a cloud environment. The apparatus may include the following modules: a resource utilization rate acquisition module, which is used to acquire the resources of each application cluster in a designated transaction link in the cloud environment utilization rate; a resource load level determination module, used to determine the resource load level of each application cluster according to the resource utilization rate; the resource load level includes a high occupancy level and a low occupancy level; the resource application acquisition module is used for The application cluster of the high occupancy level applies for resources; the resource release module is used to release the resources of the application cluster of the low occupancy level according to the resource load level if the resource application for the application cluster of the high occupancy level fails; resource allocation a module for allocating the released resources to the application clusters with high occupation level.

FIG. 5 is a schematic structural diagram of a server according to an embodiment of the present invention. The server 400 may vary greatly due to different configurations or performance, and may include one or more central processing units (CPU) 422 (eg, one or more processors) and memory 432 , one or more storage media 430 (eg, one or more mass storage devices) that store applications 442 or data 444. Among them, the memory 432 and the storage medium 430 may be short-term storage or permanent storage. The program stored in the storage medium 430 may include one or more modules (not shown in the figure), and each module may include a series of command operations on the server. Furthermore, the central processing unit 422 may be configured to communicate with the storage medium 430 to execute a series of command operations in the storage medium 430 on the server 400 .

The server 400 may also include one or more power supplies 426, one or more wired or wireless network interfaces 450, one or more input and output interfaces 458, one or more keyboards 456, and/or, a Or more than one operating system 441, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

The central processing unit 422 may execute the following instructions on the server 400: obtain the resource load status of each application cluster in the cloud environment; determine the resource load level of the application cluster according to the resource load status; the resource load level Including the resource full load level and other levels; apply for resources for the application cluster of the resource full load level; if the application cluster for the resource full load level fails to apply for resources, then according to the resource load level, release the resources of the application clusters of other levels; The released resources are allocated to the application clusters at the resource full load level.

Optionally, the one or more modules may have the following function: when the access traffic increases, the resource load status of each application cluster in the cloud environment is obtained.

Optionally, the cloud environment includes a resource pool, and the one or more modules may have the following functions: apply for resources from the resource pool; when the resources in the resource pool are sufficient, the resource full load level Allocating resources in the resource pool to the application cluster of the resource pool; when the resources of the resource pool are exhausted, the application cluster of the resource-full level fails to apply for resources.

Optionally, the application cluster has a corresponding priority order, and the one or more modules may have the following functions: when there are two or more application clusters at the full resource load level, according to the priority order, The application cluster at the resource-full level allocates the resources in the resource pool.

Optionally, the resource load level has a corresponding release ratio, and the one or more modules may have the following functions: determine the release ratio corresponding to the resource load level of the application cluster of other levels; release all the resources according to the release ratio. resources of application clusters at other levels; and save the released resources in the resource pool.

Optionally, the one or more modules may have the following functions: obtain the current resources of the application clusters of the other levels; obtain the current access traffic of the application clusters of the other levels; adopt the current access traffic and preset Calculate the demanded resources by using the single-machine limit value; use the current resources and the demanded resources to determine the idle resources of the application clusters of other levels; release the idle resources according to the release ratio.

Optionally, the one or more modules may have the following functions: apply for resources from the resource pool storing the released resources; and allocate the resources released in the resource pool to the application clusters at the resource-full level.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other.

It should be understood by those skilled in the art that embodiments of the embodiments of the present invention may be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may employ computer program products implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) having computer-usable program code contained therein form.

In a typical configuration, the computer device includes one or more processors (CPUs), an input/output interface, a network interface, and memory. Memory may include forms of non-persistent memory, random access memory (RAM) and/or non-volatile memory in computer readable media, such as read only memory (ROM) or flash memory ( flash RAM). Memory is an example of a computer-readable medium. Computer-readable media includes both permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology for storage of information. Information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM) , Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash Memory or Other Memory Technologies, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, magnetic cassette tape, magnetic tape disk storage or other magnetic storage device or any other non-transmission medium that may be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

Embodiments of the present invention are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present invention. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal equipment to produce a machine for executing the instructions by the processor of the computer or other programmable data processing terminal equipment Means are created to implement the functions specified in the flow or flows in the flowcharts and/or the block or blocks in the block diagrams.

These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal equipment to operate in a particular manner, such that the instructions stored in the computer readable memory produce the manufacture including the instruction device product, the instruction means implements the functions specified in the flow or flows in the flowchart and/or the block or blocks in the block diagram.

These computer program instructions can also be loaded on a computer or other programmable data processing terminal equipment, so that a series of operational steps are performed on the computer or other programmable terminal equipment to produce computer-implemented processing, so that the computer or other programmable terminal equipment The instructions executing on the device provide steps for implementing the functions specified in one or more of the flowcharts in the flowcharts and/or one or more blocks in the block diagrams.

While the preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts are known. Therefore, the scope of the appended claims is intended to be construed to include the preferred embodiment as well as all changes and modifications that fall within the scope of the embodiments of the present invention.

Finally, it should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or that there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or terminal device comprising a list of elements includes not only those elements, but also a non-exclusive list of elements. other elements, or also include elements inherent to such a process, method, article or terminal equipment. Without further limitation, an element defined by the phrase "comprises a..." does not preclude the presence of additional identical elements in the process, method, article or terminal device comprising said element.

A method for allocating application cluster resources in a cloud environment, an apparatus for allocating application cluster resources in a cloud environment, and a system for allocating application cluster resources in a cloud environment provided by the present invention have been described in detail above. The principles and implementations of the present invention are described with a single example, and the descriptions of the above examples are only used to help understand the method and the core idea of the present invention; There will be changes in the implementation manner and the scope of application. To sum up, the content of this specification should not be construed as a limitation to the present invention.

Claims (15)

A system for resource allocation of application clusters in a cloud environment, characterized in that the system comprises: one or more processors; a memory; and one or more modules, and the one or more modules are stored in the memory is configured to be executed by the one or more processors, and the one or more modules have the following functions: obtain the resource utilization rate of each application cluster in the specified transaction link in the cloud environment; Determine the resource load level of each application cluster; the resource load level includes high occupancy level and low occupancy level; apply for resources for the application cluster with high occupancy level; level to release the resources of the application cluster with low occupation level; and allocate the released resources to the application cluster with high occupation level; wherein, the obtaining the resource utilization rate of each application cluster in the designated transaction link in the cloud environment includes: when When the access traffic increases, the resource utilization rate of each application cluster in the specified transaction link in the cloud environment is obtained; wherein, the specified transaction link is formed by the interaction between each application cluster and is used to complete the complete transaction work. A method for allocating application cluster resources in a cloud environment, characterized in that the method comprises: acquiring the resource utilization rate of each application cluster in a designated transaction link in the cloud environment; determining the resource load level of each application cluster according to the resource utilization rate ; The resource load level includes high occupancy level and low occupancy level; apply for resources for the application cluster of the high occupancy level; if the application cluster for the high occupancy level fails to apply for resources, the application of the low occupancy level will be released according to the resource load level resources of the cluster; and allocating the released resources to the high-occupancy-level application cluster; wherein, the step of obtaining the resource utilization rate of each application cluster in the designated transaction link in the cloud environment includes: when the access traffic increases, obtaining The resource utilization rate of each application cluster in the specified transaction link in the cloud environment; wherein, the specified transaction link is formed by the interaction between each application cluster and is used to complete the complete transaction work. The method according to item 2 of the scope of application, wherein the cloud environment includes a resource pool, and the step of applying for resources for an application cluster with a high occupancy level includes: applying for resources from the resource pool; when the resources in the resource pool are sufficient When the application cluster with high occupancy level is allocated the resources in the resource pool; and when the resources of the resource pool are exhausted, the application cluster with the high occupancy level will be allocated the resources in the resource pool. The cluster failed to apply for resources. The method according to item 3 of the scope of the application, wherein the application cluster has a corresponding priority order, and the step of allocating resources in the resource pool to the application cluster with a high occupancy level includes: when the application cluster with a high occupancy level is When there are two or more, resources in the resource pool are allocated to application clusters with high occupancy levels according to the priority order. The method according to item 2 or 3 of the scope of the application, wherein the resource load level has a corresponding release ratio, and the step of releasing the resources of the application cluster with a low occupancy level according to the resource load level includes: determining the low occupancy level of the application cluster The release ratio corresponding to the resource load level of the application cluster; the resources of the low-occupancy level application cluster are released according to the release ratio; and the released resources are saved in the resource pool. The method according to item 5 of the scope of the application, wherein the step of releasing the resources of the low-occupancy level application cluster according to the release ratio includes: acquiring the current resources of the low-occupancy level application cluster; acquiring the low-occupancy level application cluster The current access traffic of the application cluster; the current access traffic and the preset single-machine limit value are used to calculate the demand resources. source; use the current resource and the demand resource to determine the idle resource of the application cluster with the low occupation level; and release the idle resource according to the release ratio. The method according to item 2 or 3 of the scope of application, wherein the step of allocating the released resources to the application cluster with high occupation level includes: applying for resources from a resource pool storing the released resources; The application cluster at the occupation level allocates the resources released in the resource pool. A device for resource allocation of application clusters in a cloud environment, characterized in that the device comprises: a resource utilization rate acquisition module for acquiring the resource utilization rate of each application cluster in a designated transaction link in a cloud environment; a resource load level determination module group, which is used to determine the resource load level of each application cluster according to the resource utilization rate; the resource load level includes high occupation level and low occupation level; the resource application acquisition module is used to apply for resources for the application cluster with high occupation level; The resource release module is used to release the resources of the low-occupancy level application cluster according to the resource load level if the resource application for the high-occupancy level application cluster fails; and the resource allocation module is used to allocate the released resources give this high occupancy level application cluster; wherein, the resource utilization rate acquisition module is further used to acquire the resource utilization rate of each application cluster in the specified transaction link in the cloud environment when the access traffic increases; wherein, the specified transaction link is determined by each application cluster The interaction between them is used to complete the complete transaction work. A method for resource allocation of application clusters in a cloud environment, characterized in that the method comprises: acquiring the resource load status of each application cluster in a designated transaction link in the cloud environment; determining the resource load level of the application cluster according to the resource load status ; The resource load level includes the resource full load level and other levels; apply for resources for the application cluster of the resource full load level; if the application cluster fails to apply for the resource full load level, the application cluster of other levels will be released according to the resource load level. resources; and allocating the released resources to the application clusters of the resource full load level; wherein, the step of acquiring the resource load status of each application cluster in the designated transaction link in the cloud environment includes: when the access traffic increases, acquiring the cloud environment The resource load status of each application cluster below; wherein, the designated transaction link is formed by interaction between each application cluster and is used to complete the complete transaction work. The method according to item 9 of the application scope, wherein the cloud environment includes a resource pool, and the step of applying for resources for an application cluster with a full resource load level includes: applying for resources from the resource pool; when the resources in the resource pool are sufficient When the resource is fully loaded, the resource in the resource pool is allocated to the application cluster of the resource-full level; and when the resource of the resource pool is exhausted, the application cluster of the resource-full level fails to apply for resources. The method according to item 10 of the scope of the application, wherein the application cluster has a corresponding priority order, and the step of allocating the resources in the resource pool to the application cluster with the full resource load level includes: when the application cluster with the full resource load level is When there are two or more, the resources in the resource pool are allocated to the application cluster with the full resource load level according to the priority order. The method according to item 9 or 10 of the scope of the application, wherein the resource load level has a corresponding release ratio, and the step of releasing resources of application clusters of other levels according to the resource load level includes: determining application clusters of other levels The release ratio corresponding to the resource load level of the specified resource load level; release the resources of the application clusters of the other levels according to the release ratio; and save the released resources into the resource pool. The method according to item 12 of the scope of the application, wherein the step of releasing the resources of the other application clusters according to the release ratio includes: acquiring the current resources of the application clusters of the other levels; acquiring the current resources of the application clusters of the other levels Access traffic; use the current access traffic and a preset stand-alone limit to calculate demanded resources; use the current resources and the demanded resources to determine the idle resources of the application clusters of other levels; and release the idle resources according to the release ratio. The method according to item 9 or 10 of the scope of the application, wherein the step of allocating the released resource to the application cluster of the resource-full level includes: applying for a resource from a resource pool storing the released resource; and assigning the resource for the resource The application cluster at the full load level allocates the resources released in the resource pool. A device for allocating application cluster resources in a cloud environment, characterized in that the device comprises: a resource load state acquisition module for acquiring the resource load state of each application cluster in a designated transaction link in a cloud environment; a resource load level determination module group, used to determine the resource load level of the application cluster according to the resource load state; the resource load level includes the resource load level The source full load level and other levels; the resource application module is used to apply for resources for the application cluster of the resource full load level; the resource release module is used to apply for resources for the resource full load level application cluster fails, according to the resource load level. Release resources of application clusters at other levels; and release a resource allocation module to allocate the released resources to application clusters of the resource full load level; wherein, the resource load status acquisition module includes: a resource load status acquisition sub-module The group is used to obtain the resource load status of each application cluster in the specified transaction link in the cloud environment when the access traffic increases; wherein, the specified transaction link is formed by the interaction between each application cluster and is used to complete the complete transaction work.

Images