WO2017031988A1

WO2017031988A1 - Resource monitoring method, apparatus and system

Info

Publication number: WO2017031988A1
Application number: PCT/CN2016/078121
Authority: WO
Inventors: 童遥
Original assignee: 中兴通讯股份有限公司
Priority date: 2015-08-24
Filing date: 2016-03-31
Publication date: 2017-03-02
Also published as: CN106487601B; CN106487601A

Abstract

Disclosed are a resource monitoring method, an apparatus and a system. The resource monitoring method comprises: a cloud node determines whether a server has pulled the resource status value of the cloud node during the current pulling period; if the server does not pull the resource status value of the cloud node during the current pulling period, the cloud node obtains the current resource status value and pushes the obtained resource status value to the server. The technical solution of the present invention can reduce the communication consumption for resource monitoring in a cloud computing data center.

Description

Resource monitoring method, device and system

Technical field

This paper relates to the field of resource monitoring technology, and in particular, to a resource monitoring method, device and system.

Background technique

With the development of Internet technology, the computer industry is constantly emerging with new technologies, from distributed computing and parallel computing to grid computing. With the development of resource virtualization technology, cloud computing has emerged. Cloud computing is an emerging business computing model that distributes computing tasks across resource pools of large numbers of computer resources, enabling users to acquire computing power, storage space, and various software services on demand. The cloud computing data center combines huge infrastructure resources, data storage resources, platform resources, and various software service resources to share a pool of collaborative resources, and abstracts some hierarchical services, such as the infrastructure layer, from the resource pool ( Different levels of services such as IaaS), platform layer (PaaS), and software layer (SaaS).

The resource monitoring of the cloud computing data center is to improve the operation and service quality of the cloud computing data center, and to monitor the usage and operation of various resources in the cloud computing data center. In the cloud computing data center, the cloud node periodically pushes its own resource state value to the server, and the server also periodically pulls the resource state value of the cloud node, and the pushing and pulling of the resource state value causes a large amount of Communication consumption. In the related art, the communication consumption of the cloud computing data center resource monitoring is large.

Summary of the invention

The main objective of the present invention is to provide a resource monitoring method, device and system, which aim to reduce communication consumption of cloud computing data center resource monitoring.

In order to achieve the above objectives, the following technical solutions are adopted:

A resource monitoring method, the resource monitoring method includes:

The cloud node determines whether the server pulls the resource state value of the cloud node in the current push period;

When the server does not pull the resource status value of the cloud node in the current push period, the cloud node acquires its current resource status value, and pushes the obtained resource status value to the server.

Optionally, when the server does not pull the resource state value of the cloud node in the current push period, the cloud node acquires its current resource state value, and obtains the obtained resource state value. The steps of pushing to the server include:

When the server does not pull the resource state value of the cloud node in the current push period, the cloud node obtains its current resource state value;

Determining, by the cloud node, whether the acquired resource state value meets a preset pushing condition;

When the resource state value satisfies the preset push condition, the cloud node pushes the resource state value to the server.

Optionally, the step of determining, by the cloud node, whether the acquired resource status value meets a preset push condition comprises:

The cloud node pushes the resource status value into a resource monitoring window of the cloud node, and calculates a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and based on the difference And calculating, by the first preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

The cloud node calculates a resource state change value based on the resource state value and the recorded resource state value acquired by the server at a time;

When the resource state change value is greater than the resource monitoring threshold, the cloud node determines that the resource state value satisfies the preset push condition.

A resource monitoring method, the resource monitoring method includes:

The server determines whether the resource status value pushed by the cloud node is received in the current pull period;

When the resource state value pushed by the cloud node is not received in the current pull period, the server pulls the current resource state value of the cloud node.

Optionally, after the step of extracting, by the server, the current resource state value of the cloud node when the resource state value of the cloud node is not received in the current pull period, the method further includes include:

The server pushes the resource status value into a resource monitoring window of the server, and calculates a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the difference and the Calculating, by the preset calculation parameter, a resource monitoring threshold corresponding to the resource status value;

The server calculates a resource state change value based on the resource state value and the recorded resource state value obtained by the server, and determines whether the resource state change value is smaller than the resource monitoring threshold;

When the resource state change value is greater than or equal to the resource monitoring threshold, the server shortens the pull cycle according to the resource state change value;

When the resource state change value is smaller than the resource monitoring threshold, the server extends the pull cycle according to the resource state change value.

A resource monitoring device includes a first determining module and a pushing module, wherein

The first determining module is configured to: determine whether the server has pulled the resource status value of the cloud node in the current push period;

The pushing module is configured to: when the server does not pull the resource state value of the cloud node in the current pushing period, acquire the current resource state value of the cloud node, and obtain the obtained resource state value. Push to the server.

Optionally, the pushing module includes an obtaining submodule, a determining submodule, and a pushing submodule, wherein

The acquiring sub-module is configured to: when the server does not pull the resource state value of the cloud node in the current push period, obtain the current resource state value of the cloud node;

The determining sub-module is configured to: determine whether the acquired resource state value meets a preset pushing condition;

The push sub-module is configured to: push the resource status value to the server when the resource status value satisfies the preset push condition.

Optionally, the determining submodule includes a calculating unit and a determining unit, where

The calculating unit is configured to: push the resource state value into a resource monitoring window of the cloud node, and calculate a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and Calculating, by the difference value and the first preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

The calculating unit is further configured to: calculate a resource state change value based on the resource state value and the recorded resource state value acquired by the server last time;

The determining unit is configured to: when the resource state change value is greater than the resource monitoring threshold, determine that the resource state value satisfies the preset push condition.

A resource monitoring device includes a second determining module and a pulling module, wherein

The second determining module is configured to: determine whether the server where the server is located receives the resource status value pushed by the cloud node during the current pull period;

The pull module is configured to: when the server does not receive the resource state value pushed by the cloud node in the current pull period, pull the current resource state value of the cloud node.

Optionally, the resource monitoring device further includes a computing module and an adjustment module, where

The calculating module is configured to: push the resource status value into a resource monitoring window of the server, and calculate a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the Calculating, by the difference and the second preset calculation parameter, a resource monitoring threshold corresponding to the resource status value;

The calculating module is further configured to: calculate a resource state change value based on the resource state value, and the recorded resource state value acquired by the server, and determine whether the resource state change value is smaller than the resource monitoring threshold ;

The adjusting module is configured to: when the resource state change value is greater than or equal to the resource monitoring threshold, shorten the pull period according to the resource state change value; and when the resource state change value is smaller than the resource When the threshold is monitored, the pull cycle is extended according to the resource state change value.

A cloud node includes any of the above resource monitoring devices.

A server includes any of the above resource monitoring devices.

A resource monitoring system includes: any of the above cloud nodes and any of the foregoing servers.

Optionally, the resource monitoring system further includes a query server,

The query server is configured to forward the query instruction to the server when receiving a query instruction sent by the user terminal;

The server further includes: an obtaining module, configured to: when receiving the query instruction, acquire a resource state value of the cloud node, and send the obtained resource state value to the query server;

The query server is further configured to: when the resource status value is received, send the resource status value to the user terminal for display.

According to the technical solution of the present invention, the cloud node performs the resource status value when the resource status value of the cloud node is not pulled by the server during the current push period by combining the push and pull status of the resource status values. The push operation reduces unnecessary network transmission and reduces the communication consumption of resource monitoring.

BRIEF abstract

1 is a schematic flowchart of a first embodiment of a resource monitoring method according to the present invention;

2 is a diagram showing, when the server does not pull the resource status value of the cloud node in the current push period, the cloud node acquires a current resource status value, and pushes the obtained resource status value. Schematic diagram of the refinement process to the server;

3 is a schematic diagram of a refinement process of determining whether the resource status value obtained by the cloud node in FIG. 2 meets a preset push condition;

4 is a diagram showing an example of a hierarchical structure of a cloud computing data center monitoring system using the resource monitoring method of the present invention;

5 is a schematic flowchart of a fourth embodiment of a resource monitoring method according to the present invention;

6 is a schematic flowchart of a fifth embodiment of a resource monitoring method according to the present invention;

7 is a schematic diagram of functional modules of a first embodiment of a resource monitoring apparatus according to the present invention;

Figure 8 is a schematic view showing the refinement structure of the push module of Figure 7;

9 is a schematic diagram showing the refinement structure of the judging sub-module in FIG. 8;

10 is a schematic diagram of functional modules of a fourth embodiment of a resource monitoring apparatus according to the present invention;

11 is a schematic diagram of functional modules of a fifth embodiment of a resource monitoring apparatus according to the present invention;

12 is a schematic diagram of a topology structure of a first embodiment of a resource monitoring system according to the present invention;

FIG. 13 is a schematic diagram showing the logic level of the first embodiment of the resource monitoring system of the present invention.

The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments.

Preferred embodiment of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The embodiment of the present invention provides a resource monitoring method. Referring to FIG. 1 , in the first embodiment of the resource monitoring method of the present invention, the resource monitoring method includes:

S10. The cloud node determines whether the server pulls the resource state value of the cloud node in the current push period.

The resource monitoring method provided in this embodiment may be applied to resource monitoring in a cloud computing data center. For example, for a cloud node, the cloud node pushes a resource state value according to a preset pushing period, if the current pushing period is in the current pushing period. The server has already pulled the resource status value of the cloud node. The cloud node does not perform the current resource state value push operation, that is, does not perform the current resource state value acquisition and push; if the server does not pull the resource state value of the cloud node in the current cycle, the cloud node executes the current time. Resource status value push operations to reduce communication consumption for cloud computing data center resource monitoring.

It should be noted that monitoring the cloud computing data center often does not require the specific use of the specific resources of the entire system, but only needs to know whether the current usage status of each resource is within a predetermined acceptable range, that is, only state monitoring is required. . In this embodiment, the cloud node determines whether the resource state value of the cloud node is pulled by the server during the current push period to determine whether the current resource state value push operation needs to be performed. The cloud node includes at least one of a virtual machine and a physical machine.

S20. When the server does not pull the resource state value of the cloud node in the current push period, the cloud node acquires a current resource state value, and pushes the resource state value to the server.

It is easy to understand that if the server does not pull the resource status value of the cloud node in the current push period, that is, the resource status value corresponding to the cloud node on the server is not updated, the cloud node needs to be performed. The secondary resource state value push operation updates the resource state value corresponding to the cloud node on the server. In this embodiment, when the server does not pull the resource state value of the cloud node in the current push period, the cloud node acquires a current resource state value, and pushes the resource state value to the a server for the server to perform an update operation based on the resource status value.

It should be noted that cloud computing is a technology that integrates resources to provide services in an instant manner. It mainly reflects technologies and services at three levels:

(1) The infrastructure layer allows hardware resources to provide services in an immediate manner;

(2) The platform layer allows the application platform to provide services in an instant manner;

(3) The application layer allows the application to provide services in an instant manner;

Among them, the way is like hydropower, from the beginning to the end of the use of measurement, login application portal can use the application directly, even without installing the application locally, just like opening the tap can use water, and then pay, it is essentially a Push service.

In this embodiment, when the cloud node acquires the current resource state value, it is preferred to collect each The runtime resource state values of the hierarchy (including the infrastructure layer, platform layer, and application layer), that is, the latest resource state values for each level.

Optionally, in this embodiment, after the step S10, the method further includes:

When the server does not pull the resource status value of the cloud node in the current push period, the cloud node does not perform the current resource status value push operation;

It is easy to understand that if the server has pulled the resource status value of the cloud node during the current push period, that is, the resource status value corresponding to the cloud node on the server has been updated, the cloud node is no longer updated. Need to perform the current resource status value push operation.

In the resource monitoring method of the present embodiment, the cloud node does not pull the resource state value of the cloud node in the current push period by using the method of acquiring the two resource state values in combination with the push and pull. Pushing resource status values reduces unnecessary network transmission and reduces communication consumption of cloud computing data center resource monitoring.

Optionally, based on the first embodiment, a second embodiment of the resource monitoring method of the present invention is proposed. Referring to FIG. 2, in the embodiment, the step S20 includes:

S21: When the server does not pull the resource state value of the cloud node in the current push period, the cloud node acquires a current resource state value.

It should be noted that the difference between this embodiment and the first embodiment is that, in this embodiment, when the cloud node performs the pushing of the resource status value, the preset push condition is added to limit, so as to filter out the uselessness in the monitoring process. Resource status value.

Specifically, when the cloud node determines that the server does not pull the resource state value of the cloud node in the current push period, the cloud node acquires a current resource state value, where the cloud node acquires The resource status values include the latest resource status values of the infrastructure layer, the platform layer, and the application layer.

For example, the resource status values obtained to the infrastructure layer include CPU occupancy, memory occupancy, and storage space occupancy.

S22. The cloud node determines whether the acquired resource state value meets a preset pushing condition.

After obtaining the current resource state value, the cloud node first determines the acquired resource shape. Whether the state value satisfies the preset push condition to determine whether the obtained resource state value is required by the user. The preset push condition may include:

The acquired resource status value is located in a first interval of preset trigger information push.

For example, the cloud node obtains the current CPU occupancy rate of 80%, and is located in a first interval [75%, 100%] of the preset trigger information push, and the cloud node determines that the acquired CPU occupancy rate satisfies the preset push. condition.

The preset push condition may further include:

The difference between the obtained resource state value and the last acquired resource state value of the server is located in a second interval of the preset trigger information push, where the last acquired resource state value of the server is the cloud node. The value of the resource state that was last pushed, or the value of the resource state that was last pulled by the server from the cloud node.

For example, the cloud node obtains the current CPU occupancy rate of 80%, and the latest resource status value corresponding to the cloud node on the server is the CPU occupancy rate of the last push of the cloud node, which is 50%. The difference 30% is located in the second interval [20%, +∞) of the preset trigger information push, and the cloud node determines that the acquired CPU occupancy rate satisfies the preset push condition.

S23. When the resource state value satisfies the preset push condition, the cloud node pushes the resource state value to the server.

In this embodiment, when the cloud node obtains the current resource state value, and determines that the acquired resource state value meets the preset pushing condition, the cloud state value is pushed to the server for the The server update corresponds to the resource status value of the cloud node as the latest resource status value.

In this embodiment, by filtering the resource state values collected by the cloud node during the monitoring process, communication consumption can be further reduced.

Optionally, based on the second embodiment, a third embodiment of the resource monitoring method of the present invention is proposed. Referring to FIG. 3, in the embodiment, the foregoing step S22 includes:

S221: The cloud node pushes the resource status value into a resource monitoring window of the cloud node, and calculates a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and based on the Calculating a resource monitoring threshold corresponding to the resource state value by using a difference value and a first preset calculation parameter value;

It should be noted that the difference between this embodiment and the second embodiment is that, in this embodiment, the cloud node determines whether a resource state value push operation needs to be performed by the degree of change of the resource state. The image said that the resource monitoring window is a container, which can be described by a two-group (L, V), which is used to indicate the length of the resource monitoring window, that is, the data capacity that can be accommodated, and V is used to indicate resource monitoring. The size of the window, which is the difference between the maximum resource state value and the minimum resource state value in the resource monitoring window. After collecting the current resource state value, the cloud node pushes the resource state value into the resource monitoring window, then updates the V value of the resource monitoring window, and substitutes the updated V value into the formula (1) to calculate the resource. The resource monitoring threshold corresponding to the status value:

CRMT=CRMW.V*CRMI...Formula (1);

Wherein, CRMT represents a resource monitoring threshold, CRMW.V represents an updated V value, and CRMI represents a monitoring relative error rate, that is, the first preset calculation parameter.

It should be noted that the value range of the CRMI is preferably [0, 1], which is used to indicate the sensitivity and timeliness of the user to the resource status value. The smaller the CRMI, the more accurate the resource status value the user needs, and the opposite is the user. Can tolerate relatively large resource state value inconsistencies. The CRMI can be set by default by the cloud node, and can also be manually set by the user. For example, the CRMI is set to 15% by default by the cloud node.

Optionally, before the cloud node pushes the acquired resource state value into the resource monitoring window, first determining whether the capacity of the resource monitoring window is full, and if so, moving the earliest resource state value out of the window, and then The obtained resource status value is pushed into the resource monitoring window; otherwise, the newly acquired resource status value is directly pushed into the resource monitoring window.

S222, the cloud node calculates a resource state change value based on the resource state value and the recorded resource state value acquired by the server at a time;

It should be noted that, when the cloud node pushes the resource state value, the resource state value of the push is recorded, and when the server detects that the resource state value of the cloud node is pulled, the server pulls the server. Resource status value.

In this embodiment, the cloud node calculates a resource state change value based on the currently obtained resource state value and the recorded resource state value acquired by the server. Specifically, the cloud node is based on the recorded resource state value that is last pushed by the cloud node, and the recorded service The server last pulls the recording time point corresponding to the resource state value of the cloud node, and determines the latest resource state value corresponding to the cloud node on the server, that is, the resource state value obtained by the server last time, and Substituting the obtained resource state value and the latest resource state value into the formula (2) to calculate the resource state change value:

RCV=|CRN-RO|...Formula (2);

The RCV represents a resource state change value, the CRN represents a newly acquired resource state value, and the RO represents a latest resource state value corresponding to the cloud node on the server.

S223. When the resource state change value is greater than the resource monitoring threshold, the cloud node determines that the resource state value meets the preset push condition.

Preferably, in this embodiment, when the RCV is greater than the CRMT, the cloud node determines that the acquired resource state value satisfies the preset push condition.

In this embodiment, the value of the resource state change of the cloud node and the server is not greater than the current resource monitoring threshold of the resource, that is, the degree of change of the resource state does not exceed the range that the user can tolerate, and further, based on maintaining data consistency between the server and the cloud node, further Reduce the communication consumption of resource monitoring.

The following describes the cloud computing data center monitoring system using the resource monitoring method of the embodiment of the present invention:

Referring to FIG. 4, the data center monitoring system includes three levels: a data presentation layer (Presentation Layer), a logical processing core layer (Logic Layer), and a data persistence access layer (Data Access Layer).

Among them, the data display layer uses FLEX to write the display interface, providing rich data graphic display effects, such as histograms, graphs, reports, etc., which can dynamically display the real-time resource status values of each node in the cloud computing data center.

The logical processing core layer is responsible for processing the monitoring data and providing an access interface, including five modules, which are Agent Manager (Agent Management) deployed on each node of the cloud computing data center (including physical machines and virtual machines). Modules, and Event Manager modules deployed on the server, CEP Engine (Complex Event Processing Engine) modules, Strategy Manager (Policy Management) modules, and API Interface (Application Programming Interface) modules. Among them, the Agent Manager module is responsible for collecting all levels on the host node (infrastructure layer, platform layer and application layer) Runtime information, which filters out useless status information in the monitoring process, and then assembles into a data packet to be sent to the server according to the policy provided by the resource monitoring method in the embodiment of the present invention; the Event manager module is responsible for pre-eventing events in the Agent Manager. Processing, and then sent to the CEP Engine (complex event processing engine) module processing; CEP Engine module is responsible for event rule matching, pattern matching events sent by the Event manager module, and then put the event into the event processing queue waiting for the Strategy Manager module to process The Strategy Manager module takes the event from the event processing queue and processes it according to the pre-defined strategy; the API Interface module is responsible for interacting with the Web front end and providing an access interface.

The data persistence access layer is responsible for providing a variety of data persistence access methods, including the Database Provider module and the XML Provider (Extensible Markup Language Support) module, wherein the Database Provider module provides support for database access, and the XML Provider module. Provides support for reading and writing XML files and provides an interface for front-end calls to query.

By applying the resource monitoring method provided by the embodiment of the present invention, the cloud computing data center monitoring system can reduce communication consumption of resource monitoring on the basis of ensuring consistency of server and node resource status values.

The embodiment of the present invention further provides a resource monitoring method. Referring to FIG. 5, a fourth embodiment of the resource monitoring method of the present invention is provided. In this embodiment, the resource monitoring method includes:

S110. The server determines whether a resource status value pushed by the cloud node is received in the current pull period.

The resource monitoring method in this embodiment may be applied to the resource monitoring of the cloud computing data center. For example, for the server, the server pulls the resource state value from the cloud node by using a preset pull period, if the current state is pulled. During the period, the server receives the resource status value pushed by the cloud node, and the server cancels the current resource status value pull operation; if the current pull period does not, the server does not receive the resource pushed by the cloud node. The state value, the server performs the current resource state value pull operation to achieve the purpose of reducing the cloud computing data center resource monitoring communication consumption.

It should be noted that monitoring the cloud computing data center often does not require the specific use of the specific resources of the entire system, but only needs to know whether the current usage status of each resource is in a predetermined state. Within the acceptable range, only state monitoring is required. In this embodiment, the server determines whether the resource status value pushed by the cloud node is received in the current pull period to determine whether the current resource status value pull operation needs to be performed.

S120. The server pulls a current resource state value of the cloud node when the resource state value pushed by the cloud node is not received in the current pull period.

It is easy to understand that if the server does not receive the resource status value pushed by the cloud node during the current pull period, that is, the resource status value corresponding to the cloud node on the server is not updated, the server needs to perform The secondary resource state value pull operation updates the resource state value corresponding to the cloud node on the server. In this embodiment, when the server does not receive the resource state value pushed by the cloud node in the current pull period, the server pulls the current resource state value of the cloud node, and updates the resource state corresponding to the cloud node. value.

In this embodiment, when the server pulls the current resource state value of the cloud node, the server preferably pulls the runtime resource state values of each level of the cloud node (including the infrastructure layer, the platform layer, and the application layer). , that is, the latest resource status values at each level.

Optionally, in this embodiment, after step S110, the method further includes:

When receiving the resource status value pushed by the cloud node in the current pull period, the server does not perform the current resource status value pull operation;

It is easy to understand that if the server receives the resource status value pushed by the cloud node during the current pull period, that is, the resource status value corresponding to the cloud node on the server has been updated, the server no longer needs to be updated. Execute the current resource status value pull operation.

In the resource monitoring method of the present embodiment, the server performs the resource when the resource state value pushed by the cloud node is not received in the current pull period by combining the push and pull state of the resource state values. The pull operation of the status value reduces unnecessary network transmission and can reduce the communication consumption of resource monitoring.

Optionally, based on the fourth embodiment, a fifth embodiment of the resource monitoring method of the present invention is provided. Referring to FIG. 6, in the embodiment, after the step S120, the method further includes:

S130. The server pushes the resource status value into a resource monitoring window of the server, and calculates a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the difference. And calculating, by the second preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

The image said that the resource monitoring window is a container, which can be described by a two-group (L, V), which is used to indicate the length of the resource monitoring window, that is, the data capacity that can be accommodated, and V is used to indicate resource monitoring. The size of the window is the difference between the maximum and minimum values of the resource status values in the current resource monitoring window. After the server pulls the current resource state value of the cloud node, the server pushes the resource state value into the resource monitoring window, and then updates the V value of the resource monitoring window, and substitutes the updated V value into the formula (3). Calculating a resource monitoring threshold corresponding to the resource status value:

SRMT=SRMW.V*SRMI...Formula (3);

The SRMT represents a resource monitoring threshold, the SRMW.V represents an updated V value, and the SRMI represents a monitoring relative error rate, which is a preset value, that is, the second preset calculation parameter.

It should be noted that the value range of the SRMI is preferably [0, 1], which is used to indicate the sensitivity and timeliness of the user to the resource status value. The smaller the SRMI is, the more accurate the resource status value is required by the user. Can tolerate relatively large resource state value inconsistencies. SRMI can be set by default by the server and can also be set manually by the user. For example, SRMI is set to 15% by default.

Optionally, before pushing the pulled resource status value into the resource monitoring window, the server first determines whether the capacity of the resource monitoring window is full, and if so, moves the earliest resource state value out of the window, and then new The resource status value pulled is pushed into the resource monitoring window; otherwise, the newly pulled resource status value is directly pushed into the resource monitoring window.

S140, the server calculates a resource state change value based on the resource state value and the recorded resource state value obtained by the server, and determines whether the resource state change value is smaller than the resource monitoring threshold; Go to step S150, and then go to step S160;

It is to be noted that, when the server receives the resource status value pushed by the cloud node, the resource status value pushed by the cloud node is recorded; when the server pulls the resource status value of the cloud node, the server records The resource status value pulled.

In this embodiment, the server calculates the resource state change value based on the currently pulled resource state value and the resource state value acquired by the server last time. Specifically, the server determines, according to the recorded resource state value that the cloud node last pushed, and the recorded recording time point corresponding to the resource state value that is last pulled by the server, determining that the cloud node corresponds to the The latest resource status value of the cloud node, that is, the resource status value acquired by the server last time, and the pulled resource status value and the latest resource status value are substituted into the formula (4) to calculate the resource status change value:

RCV=|SRN-RO|...Formula (4);

The RCV represents a resource state change value, the SRN represents a newly pulled resource state value, and the RO represents a latest resource state value corresponding to the cloud node on the server.

S150. The server shortens the pull period according to the resource state change value.

S160. The server extends the pull period according to the resource state change value.

In this embodiment, when the server shortens the pull cycle, the pull cycle may be shortened according to a preset reduction amount, for example, the pull cycle is shortened by 5 seconds each time. The server may further determine a reduction amount corresponding to the resource state change value according to a preset relationship between the resource state change value and the reduction amount, and shorten the pull cycle according to the determined reduction amount.

When the server extends the pull cycle, the pull cycle may be extended according to a preset increase amount, for example, the pull cycle is extended by 5 seconds each time. The server may further determine an increase amount corresponding to the resource state change value according to a preset relationship between the resource state change value and the increase amount, and extend the pull cycle according to the determined increase amount.

In this embodiment, by continuously adjusting the duration of the pull cycle, the communication consumption of resource monitoring can be further reduced while maintaining the consistency of the cloud node and server resource state values.

The embodiment of the present invention further provides a resource monitoring apparatus. Referring to FIG. 7, in the first embodiment of the resource monitoring apparatus of the present invention, the resource monitoring apparatus includes:

The first determining module 10 is configured to: determine whether the server pulls the resource status value of the cloud node in the current push period;

The resource monitoring device of the present embodiment can be applied to the resource monitoring of the cloud computing data center. For example, the resource monitoring device is built in the cloud node, and the cloud node pushes the resource state value at intervals according to a preset pushing period. During the current push period, the server has pulled the resource state value of the cloud node, and the cloud node does not perform the current resource state value push operation, that is, does not perform the current resource state value acquisition and push; if in the current cycle The server does not pull the resource state value of the cloud node, and the cloud node performs the current resource state value push operation to reduce the communication consumption of the cloud computing data center resource monitoring.

It should be noted that monitoring the cloud computing data center often does not require the specific use of the specific resources of the entire system, but only needs to know whether the current usage status of each resource is within a predetermined acceptable range, that is, only state monitoring is required. . The resource monitoring device provided in this embodiment is built in the cloud node. Specifically, the first determining module 10 determines whether the server pulls the resource state value of the cloud node in the current push period to determine whether the current need is performed. Resource status value push operation. The cloud node includes at least one of a virtual machine and a physical machine.

The pushing module 20 is configured to: when the server does not pull the resource state value of the cloud node in the current pushing period, acquire the current resource state value of the cloud node, and obtain the obtained resource state value. Push to the server.

It is easy to understand that if the server does not pull the resource status value of the cloud node in the current push period, that is, the resource status value corresponding to the cloud node on the server is not updated, the push module 20 needs to perform The secondary resource state value push operation updates the resource state value corresponding to the cloud node on the server. In this embodiment, when the server does not pull the resource state value of the cloud node in the current push period, the pushing module 20 acquires the current resource state value of the cloud node, and obtains the resource state. A value is pushed to the server for the server to perform an update operation based on the resource status value.

In this embodiment, when the current resource state value of the cloud node is acquired, the push module 20 preferably collects runtime resource state values of each level of the cloud node (including an infrastructure layer, a platform layer, and an application layer). , that is, the latest resource status values at each level.

Optionally, in this embodiment, when the server pulls the resource status value of the cloud node in the current push period, the push module 20 does not perform the resource status value push operation;

It is easy to understand that if the server has pulled the resource status value of the cloud node in the current push period, that is, the resource status value corresponding to the cloud node on the server has been updated, the push module 20 does not It is then necessary to perform a resource state value push operation.

The resource monitoring device provided in this embodiment is built in the cloud node, and combined with the method of obtaining the two resource state values by pushing and pulling, so that the cloud node does not pull the server in the current push period. When the resource state value of the cloud node is used, the resource state value is pushed, the unnecessary network transmission is reduced, and the communication consumption of the cloud computing data center resource monitoring can be reduced.

Optionally, based on the first embodiment, a second embodiment of the resource monitoring apparatus of the present invention is provided. Referring to FIG. 8, in the embodiment, the pushing module 20 includes:

The obtaining sub-module 21 is configured to: when the server does not pull the resource state value of the cloud node in the current push period, obtain the current resource state value of the cloud node;

It should be noted that, in this embodiment, the difference between the embodiment and the first embodiment is that, in the embodiment, the push module 20 adds a preset push condition to limit the push of the resource status value to filter out the monitoring process. Unwanted resource status value.

Specifically, when the first determining module 10 determines that the server does not pull the resource state value of the cloud node in the current pushing period, the acquiring sub-module 21 acquires the current cloud node. The resource status value, wherein the resource status value acquired by the obtaining submodule 21 includes an latest resource status value of an infrastructure layer, a platform layer, and an application layer.

For example, the resource status values obtained by the acquisition sub-module 21 to the infrastructure layer include the CPU occupancy rate, the memory occupancy rate, and the storage space occupancy rate.

The determining sub-module 22 is configured to: determine whether the acquired resource state value satisfies a preset pushing condition;

After the obtaining sub-module 21 acquires the current resource state value of the cloud node, the determining sub-module 22 first determines whether the acquired resource state value satisfies a preset pushing condition, and determines the acquired resource state value. Whether it is needed by the user. The preset push condition may include:

For example, the obtaining sub-module 21 acquires that the current CPU occupancy rate of the cloud node is 80%, and is located in a first interval [75%, 100%] of the preset trigger information push, and the determining sub-module 22 determines The CPU occupancy rate obtained by the acquisition sub-module 21 satisfies a preset push condition.

The preset push condition may further include:

For example, the acquisition sub-module 21 obtains that the current CPU occupancy rate of the cloud node is 80%, and the latest resource status value corresponding to the cloud node on the server is the CPU occupancy rate of the last pull of the server. 50%, the difference between the two is 30% in the second interval [20%, +∞) of the preset trigger information push, and the determining sub-module 22 determines the CPU possession acquired by the obtaining sub-module 21 The rate meets the preset push conditions.

The pushing sub-module 23 is configured to: when the resource state value satisfies the preset pushing condition, push the resource state value to the server.

In this embodiment, when the obtaining sub-module 21 acquires the current resource state value of the cloud node, and the determining sub-module 22 determines that the acquired resource state value satisfies the preset pushing condition, Pushing the resource status value to the server for the server to update the pair The resource status value of the cloud node should be the latest resource status value.

Optionally, based on the second embodiment, a third embodiment of the resource monitoring apparatus of the present invention is provided. Referring to FIG. 9, in the embodiment, the determining sub-module 22 includes:

The calculating unit 221 is configured to: push the resource state value into a resource monitoring window of the cloud node, and calculate a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and Calculating, by the difference value and the first preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

It should be noted that the difference between this embodiment and the second embodiment is that, in this embodiment, the determining sub-module 22 determines whether a resource state value pushing operation is required by the degree of change of the resource state. The image said that the resource monitoring window is a container, which can be described by a two-group (L, V), which is used to indicate the length of the resource monitoring window, that is, the data capacity that can be accommodated, and V is used to indicate resource monitoring. The size of the window, which is the difference between the maximum resource state value and the minimum resource state value in the resource monitoring window. After the acquisition sub-module 21 collects the current resource state value, the computing unit 221 pushes the resource state value into the resource monitoring window, and then updates the V value of the resource monitoring window, and updates the V value. Substituting formula (1) to calculate a resource monitoring threshold corresponding to the resource status value:

CRMT=CRMW.V*CRMI...Formula (1);

Optionally, the calculating unit 221 first determines whether the capacity of the resource monitoring window is full before pushing the acquired resource state value into the resource monitoring window, and if so, the earliest one. The resource status value is moved out of the window, and the newly acquired resource status value is pushed into the resource monitoring window; otherwise, the newly acquired resource status value is directly pushed into the resource monitoring window.

The calculating unit 221 is further configured to: calculate a resource state change value based on the resource state value and the recorded resource state value acquired by the server last time;

It should be noted that the resource monitoring apparatus further includes a recording module configured to: when the cloud node (the push module 20) pushes the resource state value, record the resource state value pushed by the cloud node; and pull the cloud node at the server When the resource status value is used, the resource status value pulled by the server is recorded.

In this embodiment, the calculating unit 221 calculates the resource state change value based on the currently obtained resource state value and the resource state value acquired by the server last time recorded by the recording module. Specifically, the calculating unit 221 determines, according to the recorded resource state value that is last pushed by the cloud node, and the recorded recording time point corresponding to the resource state value of the cloud node that is last pulled by the server. The latest resource status value corresponding to the cloud node on the server, that is, the resource status value acquired by the server last time, and the obtained resource status value and the latest resource status value are substituted into the formula (2) computing resource status Change value:

RCV=|CRN-RO|...Formula (2);

The determining unit 222 is configured to: when the resource state change value is greater than the resource monitoring threshold, determine that the resource state value meets the preset push condition.

Preferably, in the embodiment, when the RCV is greater than the CRMT, the determining unit 222 determines that the acquired resource state value satisfies the preset push condition.

The following describes the cloud computing data center monitoring system using the resource monitoring device of the embodiment of the present invention:

Referring to FIG. 4, the data center monitoring system includes three levels: a data presentation layer (Presentation Layer), a logical processing core layer (Logic Layer), and a data persistence access layer (Data Access). Layer).

The logical processing core layer is responsible for processing the monitoring data and providing an access interface, including five modules, which are Agent Manager (Agent Management) deployed on each node of the cloud computing data center (including physical machines and virtual machines). Modules, and Event Manager modules deployed on the server, CEP Engine (Complex Event Processing Engine) modules, Strategy Manager (Policy Management) modules, and API Interface (Application Programming Interface) modules. The Agent Manager module is responsible for collecting runtime information of each level (infrastructure layer, platform layer and application layer) on the host node, which filters out useless status information in the monitoring process, and then assembles into a data packet according to the present invention. The policy provided by the resource monitoring method is sent to the server; the Event manager module is responsible for preprocessing the events in the Agent Manager and then sending them to the CEP Engine (complex event processing engine) module for processing; the CEP Engine module is responsible for matching the event rules, and the Event is The event sent by the manager module performs pattern matching, and then puts the event into the event processing queue for processing by the Strategy Manager module; the Strategy Manager module takes the event from the event processing queue and performs corresponding processing according to the pre-defined strategy; the API Interface module is responsible for Interact with the web front end and provide an access interface.

By applying the resource monitoring apparatus provided by the embodiment of the present invention, the cloud computing data center monitoring system can reduce communication consumption of resource monitoring on the basis of ensuring consistency of server and node resource status values.

The embodiment of the present invention further provides a resource monitoring apparatus. Referring to FIG. 10, a fourth embodiment of the resource monitoring apparatus of the present invention is provided. In this embodiment, the resource monitoring apparatus includes:

The second determining module 110 is configured to: determine whether the server where the server is located receives the resource status value pushed by the cloud node during the current pull period;

The resource monitoring device of the present embodiment can be applied to the resource monitoring of the cloud computing data center. For example, the resource monitoring device is built in the server, and the server pulls the resource state value from the cloud node by using a preset pull period. If the server receives the resource status value pushed by the cloud node during the current pull period, the server cancels the current resource status value pull operation; if the current pull period, the server does not receive the The resource status value pushed by the cloud node, and the server performs the current resource status value pull operation to reduce the consumption of the cloud computing data center resource monitoring communication.

It should be noted that monitoring the cloud computing data center often does not require the specific use of the specific resources of the entire system, but only needs to know whether the current usage status of each resource is within a predetermined acceptable range, that is, only state monitoring is required. . In this embodiment, the second determining module 110 is configured to: determine whether the server where the server is located receives the resource state value pushed by the cloud node during the current pull period, to determine whether the current resource state value pull operation needs to be performed.

The pull module 120 is configured to: when the server does not receive the resource state value pushed by the cloud node in the current pull period, pull the current resource state value of the cloud node.

It is easy to understand that if the server does not receive the resource status value pushed by the cloud node during the current pull period, that is, the resource status value corresponding to the cloud node on the server is not updated, the pull module 120 needs to be updated. The current resource status value pull operation is performed to update the resource status value corresponding to the cloud node on the server. In this embodiment, the pull module 120 pulls the current resource state value of the cloud node when the server does not receive the resource state value pushed by the cloud node in the current pull period. The server update corresponds to a resource status value of the cloud node.

Among them, the way is like hydropower, from the beginning to the end of use to measure, the login should With the portal, you can use the app directly, even without installing the app locally, just like opening the tap to use water, and then paying, it is essentially a push service.

In the embodiment, the pull module 120 preferably pulls the running time of each level (including the infrastructure layer, the platform layer, and the application layer) of the cloud node when the current resource state value of the cloud node is pulled. The resource status value, which is the latest resource status value for each level.

Optionally, in this embodiment, when the server receives the resource status value pushed by the cloud node in the current pull period, the pull module 120 does not perform the current resource status value pull operation;

It is easy to understand that, if the server receives the resource status value pushed by the cloud node in the current pull period, that is, the resource status value corresponding to the cloud node on the server has been updated, the pull module 120 is updated. It is no longer necessary to perform the current resource state value pull operation.

The resource monitoring device provided in this embodiment is built in the server and combined with the method of obtaining the two resource state values by pushing and pulling, so that the server does not receive the cloud node push in the current pull cycle. When the resource status value is used, the pull operation of the resource status value is performed, the unnecessary network transmission is reduced, and the communication consumption of the resource monitoring can be reduced.

Optionally, based on the fourth embodiment, a fifth embodiment of the resource monitoring apparatus of the present invention is provided. Referring to FIG. 11, in the embodiment, the resource monitoring apparatus further includes:

The calculating module 130 is configured to: push the resource status value into a resource monitoring window of the server, and calculate a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the Calculating, by the difference and the second preset calculation parameter, a resource monitoring threshold corresponding to the resource status value;

SRMT=SRMW.V*SRMI...Formula (3);

Optionally, the calculating module 130 first determines whether the capacity of the resource monitoring window is full before pushing the pulled resource state value into the resource monitoring window, and if yes, moves the earliest resource state value out of the window, and then Pushing the newly pulled resource status value into the resource monitoring window; otherwise, directly pushing the newly pulled resource status value into the resource monitoring window.

The calculating module 130 is further configured to: calculate a resource state change value based on the resource state value and the recorded resource state value obtained by the server last time;

It should be noted that the resource monitoring apparatus further includes a recording module, configured to: when the server receives the resource status value pushed by the cloud node, record a resource status value pushed by the cloud node; and pull the server status When the resource state value of the cloud node is taken, the resource state value pulled by the server is recorded.

In this embodiment, the calculation module 130 calculates a resource state change value based on the currently pulled resource state value and the resource state value acquired by the server last time. Specifically, the calculating module 130 determines that the cloud node corresponds to the recorded resource time value that is last pushed by the cloud node and the recorded time point corresponding to the resource state value that is last pulled by the server. The latest resource status value of the cloud node, that is, the resource status value acquired by the server last time, and the pulled resource status value and the latest resource status value are substituted into the formula (4) to calculate the resource status change value:

RCV=|SRN-RO|...Formula (4);

The adjusting module 140 is configured to: when the resource state change value is greater than or equal to the resource monitoring When the threshold is controlled, the pull period is shortened according to the resource state change value; and when the resource state change value is smaller than the resource monitoring threshold, the pull period is extended according to the resource state change value.

In this embodiment, when the adjustment module 140 shortens the pull cycle, the pull cycle may be shortened according to a preset reduction amount. For example, the adjustment module 140 shortens the pull cycle by 5 seconds each time. The adjustment module 140 may further determine the reduction amount corresponding to the resource state change value according to the preset relationship between the resource state change value and the reduction amount, and shorten the pull cycle according to the determined reduction amount.

The adjustment module 140 may extend the pull period according to a preset increase amount when the pull period is extended. For example, the adjustment module 140 extends the pull period by 5 seconds each time. The adjustment module 140 may further determine an increase amount corresponding to the resource state change value according to a preset relationship between the resource state change value and the increase amount, and extend the pull cycle according to the determined increase amount.

The embodiment of the present invention further provides a resource monitoring system. Referring to FIG. 12, in the first embodiment of the resource monitoring system of the present invention, the resource monitoring system includes a cloud node 100 and a server 200, where

The cloud node 100 includes:

The first determining module is configured to: determine whether the server 200 pulls the resource status value of the cloud node 100 during the current push period;

The pushing module is configured to: when the server 200 does not pull the resource state value of the cloud node 100 in the current pushing period, acquire the current resource state value of the cloud node 100, and obtain the obtained resource The status value is pushed to the server 200;

The server 200 includes:

The second determining module is configured to: determine whether the server 200 receives the resource status value pushed by the cloud node 100 during the current pull period;

Pulling the module, configured to: when the server 200 does not receive the current pull period When the resource state value pushed by the cloud node 100 is pulled, the current resource state value of the cloud node 100 is pulled.

In order to reduce the pressure on the server side and from the aspect of easy management, referring to FIG. 12, the resource monitoring system provided in this embodiment adopts a sub-area deployment server 200 manner, and the server 200 in each area only receives and processes each cloud node 100 in the area. For the monitoring data, the process for the server 200 to obtain the monitoring data (including the server 200 pull and the cloud node 100 push) may be specifically implemented by referring to the foregoing embodiment, and details are not described herein again.

In addition, in order to facilitate the user's query, the resource monitoring system provided in this embodiment is further provided with a query server 300 for responding to the query of the user terminal 400.

Specifically, the query server 300 is configured to: when receiving the query instruction sent by the user terminal 400, forward the query instruction to the server 200;

The server 200 further includes an obtaining module, configured to: when receiving the query instruction, acquire a corresponding resource status value, and send the obtained resource status value to the query server 300;

The query server 300 is further configured to: when receiving the resource status value, send the resource status value to the user terminal 400 for display.

Further, referring to FIG. 13, FIG. 13 is a diagram showing an example of a logical hierarchical structure of a resource monitoring system according to an embodiment of the present invention.

The resource monitoring system provided by this embodiment adopts a three-level architecture, and is composed of three layers: a presentation layer, a data processing layer, and a resource layer.

The data display layer is responsible for the unified display of the monitoring data, and is displayed in a complete B/S manner through the portal, and realizes interaction with the user, responding to the user's operation and setting, and the like.

The data processing layer is responsible for monitoring the execution of the policy, and the collected raw data is summarized by the data and written into the database, so that the data display layer calls the monitoring data from the database.

The monitored layer includes all managed objects, that is, resources, which can be acquired by the Agent.

The embodiment of the invention also discloses a computer program, including program instructions, when the program instruction When executed by the cloud node, the cloud node can perform any of the above resource monitoring methods.

The embodiment of the invention also discloses a carrier carrying the computer program.

The embodiment of the invention also discloses a computer program, comprising program instructions, which, when executed by the server, enable the server to perform any of the above resource monitoring methods.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

One of ordinary skill in the art will appreciate that all or a portion of the steps of the above-described embodiments can be implemented using a computer program flow, which can be stored in a computer readable storage medium, such as on a corresponding hardware platform (eg, The system, device, device, device, etc. are executed, and when executed, include one or a combination of the steps of the method embodiments.

Alternatively, all or part of the steps of the above embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve. Thus, the invention is not limited to any specific combination of hardware and software.

The devices/function modules/functional units in the above embodiments may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.

When each device/function module/functional unit in the above embodiment is implemented in the form of a software function module and sold or used as a stand-alone product, it can be stored in a computer readable storage medium. The above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Variations or substitutions are readily conceivable within the scope of the present invention by those skilled in the art and are within the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the claims.

Industrial applicability

According to the technical solution of the present invention, the cloud node performs the resource status value when the resource status value of the cloud node is not pulled by the server during the current push period by combining the push and pull status of the resource status values. The push operation reduces unnecessary network transmission and reduces the communication consumption of resource monitoring. Therefore, the present invention has strong industrial applicability.

Claims

A resource monitoring method, the resource monitoring method includes:

The cloud node determines whether the server pulls the resource state value of the cloud node in the current push period;

When the server does not pull the resource status value of the cloud node in the current push period, the cloud node acquires its current resource status value, and pushes the obtained resource status value to the server.
The resource monitoring method according to claim 1, wherein the cloud node acquires its current resource state value when the server does not pull the resource state value of the cloud node in the current push period. And the step of pushing the obtained resource status value to the server includes:

When the server does not pull the resource state value of the cloud node in the current push period, the cloud node obtains its current resource state value;

Determining, by the cloud node, whether the acquired resource state value meets a preset pushing condition;

When the resource state value satisfies the preset push condition, the cloud node pushes the resource state value to the server.
The resource monitoring method according to claim 2, wherein the step of determining, by the cloud node, whether the acquired resource status value satisfies a preset push condition comprises:

The cloud node pushes the resource status value into a resource monitoring window of the cloud node, and calculates a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and based on the difference And calculating, by the first preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

The cloud node calculates a resource state change value based on the resource state value and the recorded resource state value acquired by the server at a time;

When the resource state change value is greater than the resource monitoring threshold, the cloud node determines that the resource state value satisfies the preset push condition.
A resource monitoring method, the resource monitoring method includes:

The server determines whether the resource status value pushed by the cloud node is received in the current pull period;

The service is not received when the cloud node pushes the resource status value in the current pull period The device pulls the current resource state value of the cloud node.
The resource monitoring method according to claim 4, wherein the server pulls the current resource state value of the cloud node when the resource state value pushed by the cloud node is not received within the current pull period After the step, the method further includes:

The server pushes the resource status value into a resource monitoring window of the server, and calculates a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the difference and the Calculating, by the preset calculation parameter, a resource monitoring threshold corresponding to the resource status value;

The server calculates a resource state change value based on the resource state value and the recorded resource state value obtained by the server, and determines whether the resource state change value is smaller than the resource monitoring threshold;

When the resource state change value is greater than or equal to the resource monitoring threshold, the server shortens the pull cycle according to the resource state change value;

When the resource state change value is smaller than the resource monitoring threshold, the server extends the pull cycle according to the resource state change value.
A resource monitoring device includes a first determining module and a pushing module, wherein

The first determining module is configured to: determine whether the server has pulled the resource status value of the cloud node in the current push period;

The pushing module is configured to: when the server does not pull the resource state value of the cloud node in the current pushing period, acquire the current resource state value of the cloud node, and obtain the obtained resource state value. Push to the server.
The resource monitoring device according to claim 6, wherein the push module comprises an acquisition submodule, a determination submodule, and a push submodule, wherein

The acquiring sub-module is configured to: when the server does not pull the resource state value of the cloud node in the current push period, obtain the current resource state value of the cloud node;

The determining sub-module is configured to: determine whether the acquired resource state value meets a preset pushing condition;

The push sub-module is configured to: push the resource status value to the server when the resource status value satisfies the preset push condition.
The resource monitoring device according to claim 7, wherein said judgment sub-module comprises a calculation unit and a determination unit, wherein

The calculating unit is configured to: push the resource state value into a resource monitoring window of the cloud node, and calculate a difference between a maximum resource state value and a minimum resource state value in the resource monitoring window, and Calculating, by the difference value and the first preset calculation parameter, a resource monitoring threshold corresponding to the resource state value;

The calculating unit is further configured to: calculate a resource state change value based on the resource state value and the recorded resource state value acquired by the server last time;

The determining unit is configured to: when the resource state change value is greater than the resource monitoring threshold, determine that the resource state value satisfies the preset push condition.
A resource monitoring device includes a second determining module and a pulling module, wherein

The second determining module is configured to: determine whether the server where the server is located receives the resource status value pushed by the cloud node during the current pull period;

The pull module is configured to: when the server does not receive the resource state value pushed by the cloud node in the current pull period, pull the current resource state value of the cloud node.
The resource monitoring device according to claim 9, wherein said resource monitoring device further comprises a calculation module and an adjustment module, wherein

The calculating module is configured to: push the resource status value into a resource monitoring window of the server, and calculate a difference between a maximum resource status value and a minimum resource status value in the resource monitoring window, and based on the Calculating, by the difference and the second preset calculation parameter, a resource monitoring threshold corresponding to the resource status value;

The calculating module is further configured to: calculate a resource state change value based on the resource state value, and the recorded resource state value acquired by the server, and determine whether the resource state change value is smaller than the resource monitoring threshold ;

The adjusting module is configured to: when the resource state change value is greater than or equal to the resource monitoring threshold, shorten the pull period according to the resource state change value; and when the resource state change value is smaller than the resource When the threshold is monitored, the pull cycle is extended according to the resource state change value.
A cloud node comprising the resource monitoring device according to any one of claims 6-8.
A server comprising the resource monitoring device of claim 9 or 10.
A resource monitoring system comprising: the cloud node according to claim 11 and the server according to claim 12.
The resource monitoring system according to claim 13, wherein the resource monitoring system further comprises a query server,

The query server is configured to forward the query instruction to the server when receiving a query instruction sent by the user terminal;

The server further includes: an obtaining module, configured to: when receiving the query instruction, acquire a resource state value of the cloud node, and send the obtained resource state value to the query server;

The query server is further configured to: when the resource status value is received, send the resource status value to the user terminal for display.