CN102497292A - Method and system for monitoring computer cluster - Google Patents

Abstract

The embodiment of the present invention proposes a computer cluster monitoring method, including the following steps: the monitored node collects operation information, and sends the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module; the main monitoring module receives the current load status of the node and the monitored content information, and stores the load status of the node and the monitored content information into the database; the parameter adjustment module according to the node The current load status is analyzed, and when the load status reaches the preset threshold, the monitoring strategy of the monitored node is adjusted, and the updated monitoring strategy is notified to the monitored node. The method proposed by the present invention rationally customizes the monitoring content according to the load status of the system, can effectively control the resources occupied by the monitoring program when the system is running under high load, and can conveniently and quickly obtain cluster monitoring status and alarm information.

Description

Method and system for computer cluster monitoring

technical field

The present invention relates to the field of computer communication, in particular, the present invention relates to a computer cluster monitoring method and system.

Background technique

A computer cluster, referred to as a cluster for short, is a computer system that is connected through a group of loosely integrated computer software and/or hardware, and highly closely cooperates to complete computing work. In a sense, they can be thought of as a computer. The individual computers in a cluster system are usually called nodes and are usually connected by a local area network, but there are other possible connections. Cluster computers are often used to improve the computing speed and/or reliability of individual computers. In general, cluster computers are much more cost-effective than individual computers, such as workstations or supercomputers.

Cluster applications are very important to the increasing computing needs of modern times, which can effectively reduce computing time and make full use of server hardware resources. System administrators need to keep abreast of the current running status and resource usage of the cluster, so they need to monitor the cluster in real time.

There are some mature products in the existing WEB cluster monitoring, but there are mainly the following problems: first, the monitoring content is fixed and cannot be customized; second, there is a gap between the timeliness, integrity and computing performance of monitoring. contradiction.

Therefore, it is necessary to propose an effective technical solution to solve the problem of computer cluster monitoring in the existing WEB mode.

Contents of the invention

The purpose of the present invention is to at least solve one of the above-mentioned technical defects, especially to optimize the monitoring performance of the system by adjusting the monitoring strategy of the monitored nodes.

The embodiment of the present invention proposes a method for computer cluster monitoring, comprising the following steps:

The monitored node collects operation information, and sends the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module respectively;

The main monitoring module receives the current load status of the node and the monitored content information, and stores the load status of the node and the monitored content information into a database;

The parameter adjustment module analyzes the current load status of the node, adjusts the monitoring strategy of the monitored node when the load status reaches a preset threshold, and notifies the monitored node of the updated monitoring strategy.

The above solution proposed by the present invention rationally customizes the monitoring content according to the load state of the system, can effectively control the resources occupied by the monitoring program when the system is running under high load, and can conveniently and quickly obtain cluster monitoring status and alarm information. In addition, the above solution proposed by the present invention has little modification to the existing system, does not affect the compatibility of the system, and is simple and efficient to implement.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

Fig. 1 is the method flowchart of computer cluster monitoring of the embodiment of the present invention;

Fig. 2 is a system structure diagram of computer cluster monitoring according to an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

In order to achieve the purpose of the present invention, the embodiment of the present invention proposes a method for computer cluster monitoring, comprising the following steps:

The monitored node collects operation information, and sends the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module respectively;

The main monitor receives the current load status of the node and the monitored content information, and stores the load status of the node and the monitored content information into a database;

The parameter adjustment module analyzes the current load status of the node, adjusts the monitoring strategy of the monitored node when the load status reaches a preset threshold, and notifies the monitored node of the updated monitoring strategy.

As shown in Figure 1, it is a flow chart of a method for computer cluster monitoring in an embodiment of the present invention, including the following steps:

S110: The monitored node collects operation information, and sends the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module respectively.

In step S110, each monitored node in the cluster first runs information collection, and sends the current load status of the node and the monitored content to the parameter adjustment module and the main monitoring module of the cluster management node respectively.

S120: The main monitoring module receives the current load status of the node and the monitored content information, and stores the node load status and the monitored content information into the database.

The main monitoring module analyzes and processes the obtained information, and stores the status and monitoring content of the nodes into the database. In addition, the main monitoring module can also provide a WEB service, and the monitoring results can be viewed through the web page.

S130: The parameter adjustment module analyzes the current load status of the node, adjusts the monitoring strategy when the load status reaches a preset threshold, and notifies the monitored node of the updated monitoring strategy.

The parameter adjustment module analyzes the current load situation. The calculation of the load is a comprehensive consideration of memory, CPU, average length of the running queue, I/O and network transmission volume. If the preset threshold is reached, the monitoring strategy will be adjusted to adjust the monitored structure Point monitoring strategies include:

Determine the monitoring strategy based on changes in network response time, CPU usage, or memory usage.

For example, when the overall load of the monitored nodes increases:

If the network response time increases, extend the running time interval of the monitored node information collection module; if the CPU usage increases, reduce the running priority of the monitored node information collection module; if the memory usage increases, run light on the monitored node Quantitative monitoring engine.

For example, when the overall load of the monitored nodes remains constant or decreases:

If the network response time decreases, reduce the running time interval of the monitored node information collection module to the default value; if the CPU usage drops, increase the running priority of the monitored node information collection module to the default value; Switch back to the default monitoring engine on the monitoring node.

For other unspecified cases, the existing parameters can be kept unchanged.

In addition, if the overall load exceeds the threshold for a long time, an alarm device will be connected to alarm or the monitored node will be restarted remotely.

The method proposed by the present invention can realize cluster monitoring based on WEB, can customize monitoring content, can effectively control the resources occupied by the monitoring program when the system is running under high load, and can obtain cluster monitoring status and alarm information conveniently and quickly.

To achieve the above purpose, as shown in FIG. 2 , an embodiment of the present invention also provides a computer cluster monitoring system, including an information collection module 200 , a main monitoring module 100 and a parameter adjustment module 300 .

The information collection module 200 is used to collect operation information on the monitored node, and send the current load status of the node and the monitored content information to the parameter adjustment module 300 and the main monitoring module 100 respectively.

The main monitoring module 100 is used to receive the current load status of the node and the monitored content information, and store the node load status and the monitored content information in the database.

The main monitoring module 100 provides WEB service for viewing the monitored content information through the webpage.

The parameter adjustment module 300 is used to analyze according to the current load status of the node, and when the load status reaches a preset threshold, adjust the monitoring strategy of the monitored node, and notify the information collection module 200 of the updated monitoring strategy.

The analysis performed by the parameter adjustment module 300 according to the current load state of the node includes:

Analyze one or more of the following parameters of the monitored nodes:

Memory usage, CPU running status, run queue length, disk I/O, process group and network transfer rate.

The parameter adjustment module 300 adjusts the monitoring strategy to include:

Determine the monitoring strategy based on changes in network response time, CPU usage, or memory usage.

For example, when the overall load of the monitored nodes increases:

If the network response time increases, extend the running time interval of the monitored node information collection module; if the CPU usage increases, reduce the running priority of the monitored node information collection module; if the memory usage increases, run light on the monitored node Quantitative monitoring engine.

For example, when the overall load of the monitored nodes remains constant or decreases:

If the network response time decreases, reduce the running time interval of the monitored node information collection module to the default value; if the CPU usage drops, increase the running priority of the monitored node information collection module to the default value; Switch back to the default monitoring engine on the monitoring node.

For other unspecified cases, the existing parameters can be kept unchanged.

In addition, if the overall load exceeds the threshold for a long time, an alarm device will be connected to alarm or the monitored node will be restarted remotely.

It should be understood that FIG. 2 is only for the sake of illustration and describes all the units or templates proposed in the present invention together. Apparently, each unit or template proposed by the present invention can also be implemented in a specific computer network system in the form of separate modules. For example, the information collection module 200 and the parameter adjustment module 300 are placed on the monitored node, the main monitoring module 100 is placed on a certain monitoring host, and so on.

For example, the overall system structure is as follows:

The information collection program in the information collection module 200 runs on the monitored nodes, and is responsible for monitoring the cluster to collect and obtain the running status of the cluster nodes and the information to be monitored. The nodes directly communicate with the main monitoring module 100, and the information collection Multiple policies are set in the module, and the monitoring content can be customized according to the extended interface provided by the main monitoring module 100 . The main monitoring program in the main monitoring module 100 runs on the monitoring host, collects the data of each information collection program and saves it in the database. The parameter adjustment program in the parameter adjustment module 300 adjusts the monitoring strategy of each node according to the operating load of each node; the alarm device sends an email and/or SMS alarm or remotely restarts the monitored node according to the preset failure scheme of the cluster system.

For example, the information collection program is composed of a main module, a communication module and multiple functional modules. The main module receives instructions from the parameter adjustment program and configures each functional module. Functional modules are divided into cluster status and load monitoring module, lightweight monitoring engine and default monitoring engine. The default monitoring engine can customize monitoring objects by configuring user scripts.

For example, the parameter tuning program includes a policy selector for priority, time interval and supervisory engine switching by load status.

The above-mentioned device proposed by the present invention can realize WEB-based cluster monitoring, can customize monitoring content, can effectively control the resources occupied by the monitoring program when the system is running under high load, and can conveniently and quickly obtain cluster monitoring status and alarm information.

Those of ordinary skill in the art can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. During execution, one or a combination of the steps of the method embodiments is included.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are realized in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like.

The above descriptions are only part of the embodiments of the present invention. It should be pointed out that those skilled in the art can make some improvements and modifications without departing from the principles of the present invention. It should be regarded as the protection scope of the present invention.

Claims (12)

1. a method for computer cluster monitoring, is characterized in that, comprises the following steps: The monitored node collects operation information, and sends the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module respectively; The main monitoring module receives the current load status of the node and the monitored content information, and stores the load status of the node and the monitored content information into a database; The parameter adjustment module analyzes the current load status of the node, adjusts the monitoring strategy of the monitored node when the load status reaches a preset threshold, and notifies the monitored node of the updated monitoring strategy. 2 . The computer cluster monitoring method according to claim 1 , further comprising: the main monitoring module providing WEB services for viewing the monitored content information through a web page. 3 . 3. The method for computer cluster monitoring as claimed in claim 1, wherein said parameter adjustment module analyzes according to the current load state of said node comprises: Analyzing one or more of the following parameters of the monitored node: Memory usage, CPU running status, run queue length, disk I/O, process group and network transfer rate. 4. the method for computer cluster monitoring as claimed in claim 1, is characterized in that, adjusting the monitoring policy of monitored node comprises: Determine the monitoring strategy based on changes in network response time, CPU usage, or memory usage. 5. the method for computer cluster monitoring as claimed in claim 4, is characterized in that, when the overall load of monitored node rises, described monitoring strategy comprises following one or more modes: If the network response time increases, extend the running time interval of the monitored node information collection module; If the CPU usage rate rises, reduce the running priority of the monitored node information collection module; If memory usage increases, run a lightweight monitoring engine on the monitored node. 6. The method for computer cluster monitoring as claimed in claim 4, characterized in that, when the overall load of the monitored node is constant or declines, the monitoring strategy includes one or more of the following methods: If the network response time decreases, reduce the running time interval of the monitored node information collection module to the default value; If the CPU usage drops, increase the running priority of the monitored node information collection module to the default value; If the memory usage increases, switch back to the default monitoring engine on the monitored node. 7. A system for computer cluster monitoring, characterized in that it comprises an information collection module, a main monitoring module and a parameter adjustment module, The information collection module is used to collect operation information on the monitored node, and send the current load status of the node and the monitored content information to the parameter adjustment module and the main monitoring module respectively; The main monitoring module is configured to receive the current load status of the node and the monitored content information, and store the load status of the node and the monitored content information into a database; The parameter adjustment module is configured to analyze according to the current load state of the node, and when the load state reaches a preset threshold, adjust the monitoring strategy of the monitored node, and notify the information collection module of the updated monitoring strategy . 8 . The computer cluster monitoring device according to claim 7 , further comprising: the main monitoring module provides WEB services for viewing monitored content information through web pages. 9. The equipment for computer cluster monitoring as claimed in claim 7, wherein the parameter adjustment module performing analysis according to the current load state of the node comprises: Analyzing one or more of the following parameters of the monitored node: Memory usage, CPU running status, run queue length, disk I/O, process group and network transfer rate. 10. The equipment of computer cluster monitoring as claimed in claim 7, is characterized in that, the monitoring strategy of described parameter adjustment module adjustment monitored node comprises: Determine the monitoring strategy based on changes in network response time, CPU usage, or memory usage. 11. The equipment for computer cluster monitoring as claimed in claim 10, characterized in that, when the overall load of the monitored nodes rises, the monitoring strategy includes one or more of the following methods: If the network response time increases, extend the running time interval of the monitored node information collection module; If the CPU usage rate rises, reduce the running priority of the monitored node information collection module; If memory usage increases, run a lightweight monitoring engine on the monitored node. 12. The equipment for computer cluster monitoring as claimed in claim 10, wherein when the overall load of the monitored node is constant or drops, the monitoring strategy includes one or more of the following methods: If the network response time decreases, reduce the running time interval of the monitored node information collection module to the default value; If the CPU usage drops, increase the running priority of the monitored node information collection module to the default value; If the memory usage increases, switch back to the default monitoring engine on the monitored node.

Images