CN102437938B - Large-scale network monitoring oriented virtual deployment system and method - Google Patents

Abstract

A virtualization deployment system and method for large-scale network monitoring, the system is provided with a deployment center platform and multiple deployment target machines communicating with it through the network. The deployment center platform is responsible for unified planning and management of the topology of the deployment of measurement probes in the entire computer network, and sends deployment tasks in combination with the existing deployment of measurement probes and network congestion, and manages and version controls the deployment of measurement probes . The deployment target machine receives and executes the commands of the deployment center platform, automatically executes corresponding operations, and makes it an available measurement probe in the network; and uses the software dependency detection mechanism and process matching principle to realize the platform independence of deployment and verify the success of deployment sex. The present invention solves many problems in the prior art, such as dependency detection of target machine deployment environment, upgrade and maintenance of management system version, concurrent deployment of many target machines, automatic installation and detection, and dynamic scheduling, etc., and has good promotion Application prospect.

Description

Virtualization deployment system and method for large-scale network monitoring

technical field

The invention relates to a virtualized deployment system and method for large-scale IP network monitoring, and belongs to the technical field of large-scale computer networks and distributed data collection systems and collection methods.

Background technique

With the rapid development of various network technologies and the popularization of products, as well as the continuous increase of users and business types, the network becomes more and more complex; the continuous increase of users and applications leads to a heavy burden on the network, resulting in network performance degradation, which It is necessary to extract and analyze the performance indicators of the network. Network performance data is of great reference value for analyzing network behavior, understanding network faults, and planning new services.

Because today's computer network is complex and huge, it is very difficult to monitor it in real time, so it is of positive significance to develop a monitoring system and method for large-scale network measurement. How to accurately and flexibly deploy monitoring tasks according to specific monitoring objectives and reduce manual intervention is critical to the performance monitoring of computer networks. The collection of monitoring data can be done by deploying measurement probes in large-scale network measurement, but the deployment of these measurement probes requires macro-control by a central platform. The central platform usually runs on a large service resource pool that can be maintained and managed by itself, and these resource pools are composed of a single server or server clusters. The monitoring probes controlled by the central platform can be dynamically deployed at the location to be monitored according to the computer network conditions. This large-scale network monitoring virtualization deployment system that dynamically deploys monitoring probes according to the monitoring requirements of the monitored computer network can meet the performance monitoring requirements of a huge computer network and dynamically carry the deployment requirements of monitoring network topology.

First introduce the meaning or definition of the related technical term used in the present invention below:

Virtualization deployment: A deployment method that uses a virtualized system platform or a virtual machine as the deployment target point. It is characterized by shielding the internal differences of the system, unifying the deployment interface, improving scalability, and reducing management difficulty.

Virtualized deployment system: a distributed system oriented to large-scale network measurement, with monitoring topology deployment and management as its main function. It is used to solve the needs of monitoring computer networks and dynamically deploying the monitoring environment, and can provide users with the required monitoring environment.

Central platform: A device composed of a single server or a server cluster for performing deployment scheduling control functions.

Target machine: The terminal as the target of virtualization deployment, which can run on a PC physical machine or a virtual machine. After successful deployment, the target machine will become the measurement detection point in the large-scale network measurement system.

RPM (RedHat Package Manager): A package management tool or an open package management system for the mainstream commercial LINUX operating system RedHat. Although the name of the file format indicates that it is related to REDHAT, its original design concept is open , and now Linux distributions including OpenLinux, S.u.S.E., and Turbo Linux are designed and controlled using it, and it has been recognized as an industry standard. It works on Red Hat Linux as well as other Linux and UNIX systems and can be used by anyone.

Extensible Markup Language XML (Extensible Markup Language): Its markup is an information symbol that computers can understand, through which computers can process documents containing various information, which needs to be interpreted by another language to Make it achieve the desired effect or be accepted by the computer.

Adapter mode Adapter: As a very practical design mode, it is used to convert the original interface into another desired interface to solve the problem of not being able to work together due to the incompatibility of the interfaces.

SSH (Secure Shell Hibernate): As one of the most widely used protocols in Linux systems, it is an encrypted data transmission protocol that can prevent DNS and IP spoofing, and encrypt the transmitted data to speed up transmission speed.

Message Digest Algorithm Fifth Edition MD5 (Message Digest Algorithm5): A verification method that performs a hash operation on received transmission data to check the correctness of the data. The specific method is to combine the calculated hash value with the data transmitted along with the If the two values are the same, it means that the transmitted data is complete and correct, and has not been tampered with, so you can use it with confidence. This verification method is used in many fields, such as: the verification of confidential information and downloaded files, etc.

In the prior art, the measurement probes for network measurement are mostly deployed manually. The specific method is that the staff arrive at the geographical location where the measurement probe needs to be deployed, and manually install the measurement device on the physical machine. When monitoring a large-scale computer network, it is necessary to deploy a very large number of measurement probes, and their distribution locations are far apart. If it is installed manually, it will consume a lot of manpower and material resources, which will increase the cost of monitoring. Most importantly, manually deployed measurement probes are extremely inconvenient in terms of management maintenance and version control. If the measurement probes need to be improved or upgraded, it seems unreasonable that these numerous measurement probes still rely on manual upgrade and maintenance. Therefore, the existing deployment methods are difficult to meet or meet the needs of monitoring large-scale computer networks. Therefore, how to improve it has become the focus of scientific and technological personnel in the industry.

Contents of the invention

In view of this, the purpose of the present invention is to provide a virtualization deployment system and method for large-scale network monitoring, which is used to solve the problems of the dependency detection of the target deployment environment and the concurrent deployment of many target machines in the current prior art. .

In order to achieve the purpose of the above invention, the present invention provides a virtualized deployment system oriented to large-scale network monitoring, which is characterized in that: the system includes: a deployment center platform and multiple deployment target machines communicating with it through the network, wherein,

The deployment center platform is responsible for collecting the topology of the deployment of measurement probes in the entire computer network for unified planning and management, and sends deployment tasks in combination with the existing deployment of measurement probes and network congestion: determine the measurement probes to be deployed The topological position of the needle, and deploy the measurement probe at this position; before the deployment of the measurement probe, detect the system environment of the deployment target machine to ensure the compatibility of the deployment; after the deployment of the measurement probe, collect and store the deployment results , update the current deployment topology; and manage and version control the deployed measurement probes; the platform is equipped with: a measurement probe deployment module and a deployment environment detection module connected to it, a system version management module, a deployment topology management module and communication module;

The deployment target machine is responsible for receiving and executing commands from the deployment center platform, including automatically installing the dependent software and measurement probe software sent by the deployment center platform, and automatically performing corresponding operations after sending the deployment results back to the deployment center platform , making it an available measurement probe in the network; the deployment target machine uses the software dependency detection mechanism to realize the platform independence of the deployment, and uses the process matching principle to verify the success of the deployment; it is equipped with: a dependency detection execution module, a deployment execution module and communication module.

In order to achieve the purpose of the above invention, the present invention also provides a virtualized deployment method for a virtualized deployment system oriented to large-scale network monitoring, which is characterized in that: the deployment center platform first analyzes the existing network status according to the deployment topology management module, and proposes Hosts in the network topology that urgently need to deploy measurement probes, and transmit the information of hosts that urgently need to deploy measurement probes in the network topology to the deployment environment detection module; the deployment environment monitoring module communicates with the deployment target machine through the communication module, and detects the deployment environment, and then the deployment target machine sends the detection results back to the deployment center platform; the deployment center platform performs the following operations according to the detection results of the deployment environment: sending the deployment environment system version to the deployment target machine and waiting for the installation result of the deployment environment system version, Or pass the result of the deployment environment detection to the measurement probe deployment module; the measurement probe deployment module connects to the deployment target machine, and the deployment execution module of the deployment target machine automatically installs the measurement probe system version, and returns the result to the Deploy the central platform, and the deployment center platform records the deployment results; if the version of the dependent system or the system version of the measurement probe is updated, the system version management module will issue a notification, prompting whether to start the upgrade of the version of the dependent system or the measurement probe, and pass After that, the upgrade operation is performed automatically.

The present invention has the following beneficial effects: the method can combine the current large-scale network status and the deployment of existing large-scale network monitoring nodes to select the topological position of the network monitoring nodes that need to be deployed separately, and realize the on-demand adjustment Transport resource reconstruction. The present invention supports the automatic completion of the deployment target machine environment detection, and, for the deployment target machines of different systems, sends the existing description documents of different working environment detection, realizes the separation of the operating system and monitoring software of the deployment target machine, and deploys monitoring The software does not need to redeploy the operating system, which effectively improves the work efficiency of the virtualized monitoring deployment environment. The system of the present invention uses the system version management module in the deployment center platform to manage the upgrade and maintenance of the system version. Once the system version library is updated, it automatically prompts to upgrade the measurement probe software, and can accurately and effectively control the software version of the measurement probe. It has good scalability, supports the deployment target function to dynamically join the virtualized deployment system, and can be deployed at any set position in the network; it also supports concurrent deployment of multiple machines, and has good deployment efficiency. In a word, the present invention has a very good prospect of popularization and application.

Description of drawings

FIG. 1 is a schematic diagram of the structural composition and data flow of the virtualized deployment system oriented to large-scale network monitoring in the present invention.

FIG. 2 is a flow chart of the monitoring method of the virtualized deployment system oriented to large-scale network monitoring in the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

The technical solution adopted by the present invention to solve the technical problem is: the virtualized deployment system oriented to large-scale network measurement is composed of two kinds of equipment: a deployment center platform and a deployment target machine respectively.

Referring to Fig. 1, the structural composition of the large-scale network monitoring-oriented virtualization deployment system of the present invention is introduced: the system is provided with a deployment center platform and multiple deployment target machines communicating with it through the network. in,

The deployment center platform, as the control center of the system, is responsible for collecting the topology of the measurement probe deployment in the entire computer network for unified planning and management; and combining the existing measurement probe deployment and network congestion to send deployment tasks: Determine the topological position of the measurement probe to be deployed, and execute the deployment of the measurement probe at this position; before the deployment of the measurement probe, detect the system environment of the deployment target machine to ensure the compatibility of the deployment; after the deployment of the measurement probe, Collect and store deployment results, update the current deployment topology; and manage and version control the deployed measurement probes. The deployment center platform is set on the virtualization monitoring platform of the server, and has the following functional modules: a measurement probe deployment module and a deployment environment detection module respectively connected thereto, a system version management module, a deployment topology management module and a communication module. in,

Deployment topology management module, as the presentation interface of the deployment center platform topology management, is used to collect the measurement probe information of the existing deployment, and store it in the database and use WEB presentation; the database saves the users of the host that can deploy the measurement probe Name and password information to facilitate subsequent deployment operations.

The system version management module is responsible for the management and control of the system version. The system version includes the measurement probe installation package and the dependent environment installation package that needs to be installed before deployment; because the virtualization deployment system saves a record for each target machine, the record The related data of configuration, deployment and deployment results of the target machine are stored in the target machine to facilitate the management and upgrade of the target machine after deployment; the system version is uniformly stored in the system version library of this module, and the system version is checked by the management Add, delete, upgrade, get operations.

Deployment environment detection module, because there are many versions of the environment of the target machine, and the difference in the operating environment is large, this module is responsible for detecting the operating environment of the deployment target machine and collecting the dependency detection results before deploying the measurement probe, and then according to the dependency detection results, from The corresponding dependent software installation package is extracted from the system version library of the system version management module, and the dependent deployment is performed on the deployment target machine; then, according to the result of the dependent deployment, it is determined whether to start the deployment operation of measuring the probe system version.

The measurement probe deployment module, as the core of the deployment center platform, is responsible for implementing the deployment operation of the measurement probe on the target machine: first obtain the result that the deployment environment can meet the conditions from the deployment environment detection module, and after confirming that it is correct, report to the system version management The module applies for the version of the measurement probe system that needs to be deployed, and then automatically deploys it on the deployment target machine; and the deployment process uses the adapter mode to implement concurrent deployment operations. After the deployment is completed, the deployment results are handed over to the deployment topology management module for processing;

The communication module is a bridge connecting the deployment center platform and the deployment target machine, and is also a channel for the deployment center platform to issue commands and transmit system versions to the deployment target machine.

The deployment target machine is a small embedded system, an ordinary PC or a virtual machine selected as a measurement probe in a computer network. As the execution terminal of the system, it is responsible for receiving and executing commands from the deployment center platform, including automatically installing the dependent software and measurement probe software sent by the deployment center platform, and automatically performing corresponding operations after sending the deployment results back to the deployment center platform , making it a measurement probe available in the network. The deployment target machine uses the software dependency detection mechanism to realize the platform independence of the deployment, and uses the process matching principle to verify the success of the deployment. There are the following functional modules: dependency detection execution module, deployment execution module and communication module; the functions of each module are as follows:

The dependency detection execution module is responsible for the analysis of the dependency detection command and the automatic installation of the dependent system version: this module extracts the dependency detection command from the dependency detection command file issued by the deployment center platform and executes it, and feeds back the dependency detection result to the deployment center platform , in order to obtain the corresponding deployment system version, then automatically install the dependent system version, and return the installation result to the deployment center platform; the dependency detection command description file is written in Extensible Markup Language XML (Extensible Markup Language), which has excellent scalability. scalability.

The deployment execution module is responsible for the analysis of the deployment command description file and the automatic installation of the deployment system version, and completes the self-starting of the measurement probe after installation: obtain the system version of the measurement probe from the deployment center platform and automatically install it, and the installation result Feedback After the central platform is deployed, the measurement probe is automatically started to make the measurement probe an available measurement node in the network.

The communication module has two channels: the channel for receiving the system version and the channel for replying to the deployment status. The former uses the common SSH protocol for communication under various operating system platforms, and its data transmission rate is high, which can complete the receiving of the system version in a short time ; The data transmission rate of the latter is not high, and the network socket communication is used.

The system of the present invention supports the collection and analysis of measurement data in multiple formats, supports data transmission in multiple communication modes, and implements a load balancing strategy during the transmission process, and supports the dynamic joining and exit of measurement probes, so that multiple measurement probes The needle can dynamically generate the optimal topology to balance the load and flow of the network.

The present invention also provides a virtualized deployment method for a virtualized deployment system oriented to large-scale network monitoring. The method is as follows: the deployment center platform first analyzes the existing network status according to the deployment topology management module, and proposes that deployment measurement is urgently needed in the network topology The host of the probe, and transmit the information to the deployment environment detection module. The deployment environment monitoring module communicates with the deployment target machine through the communication module, and detects the deployment environment, and then the deployment target machine sends the detection result back to the deployment center platform. The deployment center platform performs the following operations according to the detection results of the deployment environment: sending the deployment environment system version to the deployment target machine and waiting for the installation result of the deployment environment system version, or transferring the result of the deployment environment detection to the measurement probe deployment module; The probe deployment module is connected to the deployment target machine, and the deployment execution module of the deployment target machine automatically installs the measurement probe system version, and after installation, returns the result to the deployment center platform, and the deployment center platform records the deployment result. If the version of the dependent system or the version of the measuring probe is updated, the system version management module will issue a notification, prompting whether to start the upgrade of the dependent system version or the version of the measuring probe, and automatically perform the upgrade operation after passing.

Referring to Fig. 2, introduce the concrete operation step of the inventive method:

Step 1, the topology management module in the deployment center platform obtains and extracts the existing network topology, combines the current network status and the deployment of existing measurement probes, gives the hosts in the network that need to deploy measurement probes urgently, and assigns the host users The name and password information is sent to the communication module.

In step 2, the deployment center platform connects one or more deployment target machines by using the adapter mode through the communication module to perform multi-machine concurrent deployment, so as to improve deployment efficiency. In this step, the multi-machine concurrent deployment operation adopted by the management center platform adapts the process class to a thread class, so that the process has thread asynchrony and can obtain construction parameters at the same time; in order to solve the problem that the management center platform uses network socket communication It is easy to block the problem of the first connection, and it makes good use of the default asynchronous mechanism provided by threads.

Step 3, after the connection is successful, the deployment environment of the deployment target machine is detected by the dependency detection execution module of the deployment target machine in the deployment center platform, and the dependency detection result is recovered; if the dependency detection passes, the follow-up operation step 4 is performed; otherwise, the deployment The deployment environment detection module of the central platform sends the dependent software lacking in the deployment target machine to the deployment target machine through the communication module, and the dependency detection execution module of the deployment target machine installs it, and returns the installation result to the deployment center platform for storage. This step includes the following operations:

(31) Send the working environment detection package: when the deployment center platform is successfully connected to the deployment target machine, the deployment environment detection module sends the deployment environment detection package including the deployment description document and the deployment environment detection script to the deployment target machine; the deployment description document Written in XML language, it describes the deployment environment of deployment target machines of different system types: including five parts: operating system type, deployment required authority, deployment command, existence of dependent software, and deployment result, and provides corresponding detection basis; the deployment The environment detection script is written in Python language. The Python language has a dedicated library for XML file parsing. By calling the XML file parsing method, the deployment description document can be parsed and the deployment environment can be detected;

(32) Automatically detect the deployment environment: the dependency detection execution module of the deployment target machine uses the detection script to extract the deployment description document information, automatically detects the deployment environment of the deployment target machine, and automatically saves the environment detection results in the result file;

(33) Recycle the deployment environment detection result: the deployment center platform receives the detection result file fed back by the communication module of the deployment target machine, extracts relevant information from it, and automatically performs the following processing operations: If the deployment environment meets the deployment requirements, the measurement probe Deploy the measurement probe for the deployment target machine for the deployment module, that is, perform the next step (4); otherwise, if the deployment target machine is found to lack dependent software, the dependency detection execution module of the deployment target machine will automatically perform the dependent system version check. Install and send the results back to the deployment center platform so that the deployment center platform can reprocess the deployment environment detection results.

Step 4, the measurement probe deployment module of the deployment center platform deploys the measurement probes to the deployment target machine, and the deployment target machine uses the deployment execution module to automatically execute the deployment operation, and sends the deployment result to the deployment center platform for processing. This step (4) includes the following operations:

(41) Send measurement probe installation package: After the deployment environment detection module notifies the measurement probe deployment module to start automatic deployment, the measurement probe deployment module sends the measurement probe installation package to the deployment target machine via the communication module. Incomplete or incorrect receipt of the system version document, the deployment target machine needs to check the received system version document with the fifth version of MD5 (MessageDigest Algorithm5). ); otherwise, the deployment center platform is required to resend the measurement probe installation software package;

(42) Automatic deployment of measurement probes: the deployment execution module of the deployment target machine parses the deployment description document information and executes the automatic deployment operation of the measurement probes; after the deployment is completed, the measurement probes are automatically started to make them available in the network Measurement node; because the deployment of the measurement probe is automatically completed by the program, the deployment may fail or be abnormal due to the difference in the operating environment during the deployment process. Therefore, after the deployment is completed, the correctness verification must be performed: the deployment execution module has a corresponding deployment correctness Verification and fault-tolerant processing strategy, under the Linux environment, use process matching to verify correctness: if the verification result is correct, then perform the subsequent step (43); once an exception occurs, immediately send an alarm to the deployment center platform so that corresponding measures can be taken;

(43) Recycle the deployment result: After the deployment result is generated by the deployment target machine, it will be directly sent back to the deployment center platform through the communication module; the deployment center platform will perform the following processing on the deployment result: if the deployment is successful, the node information that will become the measurement probe will be sent to the deployment center platform. The topology management module is used to manage and control it; if the deployment fails, an alarm message will be generated and wait for human processing.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (10)

1. A virtualization deployment system for large-scale network monitoring is characterized in that: the system comprises: a deployment center platform and a plurality of deployment target machines in communication therewith via a network, wherein,

the deployment center platform is responsible for collecting topological structures of measurement probe deployment in the whole computer network so as to realize unified planning and management, and issues deployment tasks by combining the existing measurement probe deployment situation and network congestion situation: determining a topological position of a measurement probe to be deployed, and performing the deployment of the measurement probe at the position; before the measurement probe is deployed, detecting the system environment of a deployed target machine to ensure the compatibility of deployment; after the measurement probe is deployed, collecting and storing a deployment result, and updating a current deployment topology; managing and version controlling the deployed measuring probes; the platform is provided with: the system comprises a measurement probe deployment module, and a deployment environment detection module, a system version management module, a deployment topology management module and a communication module which are respectively connected with the measurement probe deployment module;

the deployment target machine is responsible for receiving and executing commands from the deployment center platform, automatically installs the dependency software and the measurement probe software sent by the deployment center platform, and automatically starts corresponding operation after returning a deployment result to the deployment center platform so as to enable the deployment target machine to become an available measurement probe in a network; the deployment target machine realizes platform independence of deployment by using a software dependence detection mechanism, and verifies the success of deployment by using a process matching principle; is provided with: a dependency detection execution module, a deployment execution module, and a communication module.

2. The virtualization deployment system of claim 1 wherein: the functions of the modules of the deployment center platform are as follows:

the deployment topology management module is used as a presentation interface of the deployment center platform topology management, is used for collecting the information of the existing deployed measurement probes, and stores the information into a database and presents the information by using WEB; the database stores user name and password information of a host capable of deploying the measurement probe so as to facilitate subsequent deployment operation;

the system version management module is responsible for managing and controlling system versions, and the system versions comprise a measurement probe installation package and a dependent environment installation package which needs to be installed before deployment; the virtualization deployment system stores a record for each target machine, and the record stores the configuration of the target machine, the deployment and the relevant data of the deployment result, so that the target machine can be managed and upgraded conveniently after deployment; the system versions are uniformly stored in a system version library of the module, and the management module performs addition, deletion, upgrading and acquisition operations on the system versions;

the deployment environment detection module is used for detecting the operation environment for deploying the target machine and collecting a dependence detection result before the deployment of the measurement probe, and then extracting a corresponding dependence software installation package from a system version library of the system version management module according to the dependence detection result to perform dependence deployment on the deployed target machine because the environment versions of the target machine are many and the difference of the operation environment is large; then, determining whether to start a deployment operation of the measurement probe system version according to a result of the dependent deployment;

the measurement probe deployment module, as the core of the deployment center platform, is responsible for executing the measurement probe deployment operation of deploying the target machine: firstly, acquiring a result that a deployment environment can meet conditions from a deployment environment detection module, applying for a system version of a measurement probe to be deployed from a system version management module after the result is confirmed to be correct, and then automatically deploying on a deployment target machine; the concurrent deployment operation realized by the adapter mode is utilized in the deployment process, and after the deployment is finished, the deployment result is uploaded to the deployment topology management module for processing;

and the communication module is connected with a bridge between the deployment center platform and the deployment target machine and is also a channel for the deployment center platform to issue commands to the deployment target machine and transmit the system version.

3. The virtualization deployment system of claim 1 wherein: the functions of the modules of the deployment target machine are as follows:

the dependence detection execution module is responsible for analyzing the dependence detection command and automatically installing the dependence system version: the module extracts and executes a dependency detection command from a dependency detection command file issued by the deployment center platform, and feeds back a dependency detection result to the deployment center platform so as to obtain a corresponding deployment system version, then automatically install the dependency system version, and return the installation result to the deployment center platform; the dependency detection command description file is compiled by adopting extensible markup language XML, and has excellent extensibility;

the deployment execution module is responsible for analyzing the deployment command description file and automatically installing the deployment system version, and completes the self-starting of the measurement probe after the installation: acquiring a system version of the measurement probe from a deployment center platform and automatically installing the system version, and after an installation result is fed back to the deployment center platform, automatically starting the measurement probe to enable the measurement probe to become an available measurement node in a network;

the communication module is provided with two channels: the system version receiving channel and the deployment state reply channel are communicated by utilizing the SSH protocol which is commonly used under various operating system platforms, have high data transmission rate and can complete the system version receiving in a short time; the latter has low requirement on data transmission rate and utilizes network socket communication.

4. The virtualization deployment system of claim 1 wherein: the deployment center platform is arranged on a virtualization monitoring platform of the server; the deployment target machine is a small embedded system, a common PC machine or a virtual machine which is selected as a measuring probe in a computer network.

5. The virtualization deployment system of claim 1 wherein: the system supports the acquisition and analysis of measurement data in various formats, supports data transmission in various communication modes, executes a load balancing strategy in the transmission process, and supports the dynamic joining and exiting of the measurement probes, so that the plurality of measurement probes can dynamically generate an optimal topology to balance the load and the flow of a network.

6. A virtualization deployment method using the virtualization deployment system of claim 1, characterized in that: the deployment center platform firstly analyzes the existing network state according to the deployment topology management module, provides a host which is urgently required to deploy the measurement probe in the network topology, and transmits the host information which is urgently required to deploy the measurement probe in the network topology to the deployment environment detection module; the deployment environment detection module is communicated with the deployment target machine through the communication module, detects the deployment environment, and then sends a detection result to the deployment center platform through the deployment target machine; the deployment center platform executes the following operations according to the detection result of the deployment environment: sending installation results of the deployment environment system version and the deployment environment system version waiting to be deployed to a deployment target machine, or transferring the results of the deployment environment detection to a measurement probe deployment module; the measurement probe deployment module is connected with a deployment target machine, a deployment execution module of the deployment target machine automatically installs a measurement probe system version, and after installation, the result is sent back to a deployment center platform, and the deployment center platform records the deployment result; if the system version or the measurement probe system version is updated, the system version management module sends a notice to prompt whether the upgrade depending on the system version or the measurement probe version is started or not, and automatically executes the upgrade operation after the upgrade passes.

7. The virtualization deployment method of claim 6, wherein: the method comprises the following operation steps:

(1) a topology management module in a deployment center platform acquires and extracts the existing network topology, gives a host which needs to deploy a measurement probe urgently in a network by combining the current network condition and the deployment condition of the existing measurement probe, and sends the user name and password information of the host to a communication module;

(2) the deployment center platform is connected with one or more deployment target machines through the communication module in an adapter mode to execute multi-machine concurrent deployment so as to improve the deployment efficiency;

(3) after the connection is successful, detecting the deployment environment of the deployment target machine through a dependence detection execution module of the deployment target machine in the deployment center platform, and recovering a dependence detection result; if the dependency detection is passed, performing a subsequent operation step (4); otherwise, the deployment environment detection module of the deployment center platform sends the dependence software lacking in the deployment target machine to the deployment target machine through the communication module, the dependence detection execution module of the deployment target machine installs the dependence software, and the installation result is sent back to the deployment center platform for storage;

(4) and the measurement probe deployment module of the deployment center platform deploys the measurement probe on the deployment target machine, the deployment target machine automatically executes deployment operation by using the deployment execution module, and the deployment result is sent to the deployment center platform for processing.

8. The virtualization deployment method of claim 7, wherein: in the step (2), the concurrent deployment operation adopted by the management center platform adapts the process class to the thread class, so that the process has the asynchronism of the thread and can obtain the construction parameters at the same time; so as to solve the problem that the management center platform is easy to block the first connection by using network socket communication, and a default asynchronous mechanism provided by a thread is well utilized.

9. The virtualization deployment method of claim 7, wherein: the step (3) comprises the following operations:

(31) sending a working environment detection packet: when the deployment center platform is successfully connected with the deployment target machine, the deployment environment detection module sends a deployment environment detection package comprising a deployment description document and a deployment environment detection script to the deployment target machine; the deployment description document is written by adopting an XML language and respectively describes deployment environments of deployment target machines of different system types: the method comprises five parts of operating system type, deployment requirement authority, deployment command, dependent software existence and deployment result, and provides corresponding detection basis; the deployment environment detection script is compiled by adopting a Python language, the Python language is provided with a special library for XML file analysis, and the deployment description document can be analyzed and the deployment environment can be detected by calling an XML file analysis method;

(32) automatically detecting a deployment environment: the dependence detection execution module of the deployment target machine extracts deployment description document information by using the detection script, automatically detects the deployment environment of the deployment target machine, and automatically stores an environment detection result in a result file;

(33) recovering and deploying environment detection results: the deployment center platform receives the detection result file fed back by the communication module of the deployment target machine, and automatically executes the following processing operations after extracting relevant information from the detection result file: if the deployment environment meets the deployment requirement, the measurement probe deployment module deploys the measurement probe to the deployment target machine, namely, the subsequent step (4) is executed; otherwise, if the deployment target machine is found to lack the dependent software, the dependent detection execution module of the deployment target machine automatically installs the dependent system version and sends the result back to the deployment center platform, so that the deployment center platform can reprocess the deployment environment detection result.

10. The virtualization deployment method of claim 7, wherein: the step (4) comprises the following operations:

(41) sending a measurement probe installation package: after the deployment environment detection module informs the measurement probe deployment module to start automatic deployment, the measurement probe deployment module sends a measurement probe installation software package to a deployment target machine through a communication module, in order to avoid incomplete or incorrect reception of a system version document caused by poor network environment, the deployment target machine needs to perform message digest Algorithm fifth version MD5 (MessageDigest Algorithm 5) verification on the received system version document, and only if the reception is correct, a subsequent automatic deployment step (42) is executed; otherwise, the deployment center platform is required to retransmit the measurement probe installation software package;

(42) automatic deployment of the measurement probe: a deployment execution module of the deployment target machine analyzes the deployment description document information and executes the automatic deployment operation of the measurement probe; after the deployment is finished, the measuring probe is automatically started to become an available measuring node in the network; since the deployment of the measurement probe is automatically completed by the program, the deployment may fail or be abnormal due to the difference of the operating environments or other factors during the deployment process, and therefore, after the deployment is completed, correctness verification must be performed: the deployment execution module is provided with corresponding deployment correctness verification and fault-tolerant processing strategies, and the correctness verification is carried out by adopting process matching under the Linux environment: if the verification result is correct, executing the subsequent step (43); once abnormity occurs, immediately alarming to a deployment center platform so as to take corresponding measures;

(43) recovering the deployment result: after the deployment target machine generates a deployment result, directly returning the deployment result to the deployment center platform through the communication module; the deployment center platform executes the following processing aiming at the deployment result: if the deployment is successful, transmitting the node information which becomes the measurement probe to a deployment topology management module so as to manage and control the node information; and if the deployment fails, generating alarm information and waiting for manual processing.

Images