KR100612254B1 - Apparatus and method of manage performance data in network management system using snmp - Google Patents

Abstract

The present invention relates to a method and apparatus for managing statistical data of a network management system using SNMP, which minimizes the time spent by the network management system to collect / manage statistical data of each NE, and normalizes the statistical data of each NE. By sending the collection message as a success message or a failure message to the management server, the management server can easily grasp the statistical data collection status of the entire NE, and can also process a large amount of statistical data.

Description

Apparatus and method of manage performance data in network management system using SNMP}

1 is a block diagram illustrating a network management system according to a preferred embodiment of the present invention.

2 is an internal block diagram illustrating an agent according to a preferred embodiment of the present invention.

3 is a flowchart for explaining the flow of a method for managing a network according to a preferred embodiment of the present invention;

<Explanation of symbols for the main parts of the drawings>

10: NE (Network Element) 20: EMS (Element Management System)

30: Agent 31: Index Table

32: MIB (Management Information Base) table

33: data processor 34: timer processor

40: database 50: server

60: management terminal

The present invention relates to a method and apparatus for managing statistical data of a network management system using SNMP. More particularly, the present invention relates to a method for managing statistical data for each NE in real time and to manage a large amount of statistical data in a network management system. The present invention relates to a method and apparatus for managing statistical data of a network management system using SNMP.

Modern communication networks have required various kinds of services, including data transmission and reception, with the development of communication technology to meet the increasing number of high-speed network users and their various needs.

In general, a network is composed of a set of several network elements (NEs), and there may be various types of NEs constituting a network, such as a router and a switch, and a base station, a base station controller, and a base station management system, which constitute a mobile communication system. , Exchange office systems, home location registration systems, etc. can also be viewed as NE.

In addition, the NE may be configured as one, but in most cases, many NEs are gathered together to form a group. In order to effectively manage a network composed of several NEs, the NE statistical data is managed. The importance of Network Management System (NMS) is emerging.

The NMS includes an Element Management System (EMS) that collects / manages statistical data of each NE, a management agent that reads statistical data of each NE managed by the EMS, and each of those read by the management agent. It consists of a management server (manager) which manages statistical data of the NE.

In this case, the management agent may provide statistical data of each NE to the management server while interworking according to the Simple Network Management Protocol (SNMP).

SNMP is for the purpose of network configuration management, performance management, equipment management and security management, and has a management server-agent structure of a server-client structure concept.

In addition, SNMP is operated through a request message for requesting data from a management server and an agent's response message, and briefly used are as follows.

1) GET message: A message requesting specific data from a server to an agent

2) SET message: A message for setting a specific value to an agent in the server.

3) TRAP message: When the data that the agent needs to notify the server occurs (for example, network resource usage data that exceeds the threshold), a message to inform the server that the data has occurred.

At this time, the NMS can be divided into two methods of processing the statistical data of each NE using SNMP.

First, the agent transmits data to the management server using a TRAP message. Second, after the agent stores the data in a management information base (MIB), the management server uses the GET message to transmit the data. How to take.

In the first method, the agent transmits a TRAP message to the server. The agent manages the statistics data for each NE and the failure data for each NE simultaneously. Thus, the statistics data for each NE and the failure data are mixed. There is a possibility that it will not be used as rich.

In the second method, the agent sets time information as index information to store each statistical data, and the management server collects the statistical data through a GET message.

However, there is a disadvantage that the index information stored by the agent increases exponentially with time, and even if the agent automatically deletes the index information according to a certain condition, the management server generally manages data in table units. It is not suitable for collecting statistical data because it is preferred to collect.

That is, in general, as an alternative to the agent using the index information as time information, after the statistical data are stored as MIB tables, the management server simply collects data in table units according to the statistical data cycle. If the statistics data is not collected due to a failure, or if the EMS or agent itself fails, the server cannot identify the failure and cannot trust the collected statistics data.

In addition, when the capacity of the statistical data for each NE increases, there is a case that the statistical data collection is not completed within the period of the server collecting the statistical data.

Accordingly, the present invention was devised to solve the above problems, using SNMP that can clearly check the failure status of each NE or EMS while managing statistical data of a large capacity of each NE in the NMS. An object thereof is to provide a method and apparatus for managing statistical data in a network management system.

Statistical data management apparatus of the network management system according to an aspect of the present invention for achieving the above object, a plurality of network elements (NE) for transmitting statistical data at a predetermined period, a database (DB) for storing the statistical data and EMS including a management terminal for managing the network, and when receiving the statistical data independently from each NE, EMS to provide a completion message containing the information of the corresponding NE is completed after receiving the statistical data after storing each statistical data in the DB Success message or failure including information of each NE after accessing and storing the statistical data stored in the DB according to the (Element Management System) and the information of the NE included in the completion message provided from the EMS within a predetermined time, and storing the data in a table form. Based on the agent providing the message and the NE contained in the success message provided by the agent Access to statistical data stored in table form, and through the failure message includes a management server that manages the fault status of each NE and agents.

In the statistical data management apparatus of the network management system according to the present invention, the agent is driven according to a predetermined timer value, and when the timer expires, a timer processing unit for providing an expiration message, and each NE in a recording space corresponding to each NE. Corresponding to the index table for storing the success or failure information received from the data, the unique information of each NE, the data table for storing the statistical data received from the NE, and the information of the NE included in the completion message received from the EMS. When the success information is recorded in the recording space of the index table, and an expiration message is received from the timer processing unit, the statistics data received from the corresponding NE in which the success information is recorded in the index table is accessed from the DB and stored in the data table. Data processing sending a success message containing information from NE to management server It includes.

On the other hand, the statistical data management method of the network management system according to another aspect of the present invention, after the EMS stores each statistical data received independently from each NE in the DB includes the information of the corresponding NE is completed receiving the statistical data Transmitting a completion message to the agent, recording the success information in a recording space corresponding to the information of the NE included in the completion message received within a predetermined time, and when the agent expires, the success information is Storing the statistical data corresponding to the information of the NE corresponding to the recorded recording space from the DB and storing the data as a data table, and the agent stores the statistical data in the data table, and displays a success message including the information of each NE. The data table according to the steps of the transfer to the management server and the management server's information contained in the success message. Standing by accessing the statistical data comprises the step of providing to the management node after the storage.

The method for managing statistical data of the network management system according to the present invention includes the steps of: when the agent expires a predetermined time, recording the failure information in a recording space in which success information is not recorded; and in the recording space in which the agent records the failure information. Transmitting a failure message including information of the NE corresponding to the management server, the agent transmitting the failure message or the success message, initializing a recording space and a predetermined time, and the management server included in the failure message. If the information of the NE is repeatedly received more than a predetermined number of times, transmitting a failure message including the information of the NE and the failure notification information to the management terminal, the management server receives a failure message or success message for a predetermined period of time from the agent If not, further comprising the step of sending an agent failure message to the management server.

In addition, in the method for managing statistical data of the network management system according to the present invention, if all of the success information is recorded in the recording space within a predetermined time, the agent transmits a success message including the information of each NE to the management server; The method further includes initializing after sending the success message.

Hereinafter, a method and apparatus for managing statistical data in a network management system using SNMP according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a network management system according to a preferred embodiment of the present invention.

Referring to FIG. 1, a network management system includes a plurality of network elements (NEs) 10, an element management system (EMS) 20 connected to each NE 10 by wire or wirelessly, and an agent (Agent) ( 30, a manager server 50, a database for storing statistical data of each NE 10 (Data Base: DB 40), and a plurality of managers (managers) for managing the network ( 60).

The management terminal 60 refers to a terminal used by a network administrator who manages the state of the network through statistical data of each NE 10, and the management terminal 60 according to the statistical data provided from the management server 50. Display information or fault occurrence information.

Each NE 10 means each component constituting the network, and each NE 10 periodically transmits statistical data to the EMS 20 according to the set time information. In addition, when each NE 10 does not generate statistical data or a failure occurs due to an ID or password mismatch in the process of transferring the statistical data to the EMS 20 through a method such as FTP (File Transfer Protocol). The failure message is transmitted to the agent 30.

At this time, the EMS 20 bypasses the failure message so that the agent 30 is transmitted to the agent 30 so that the agent 30 surely fails the statistical processing task at the specific NE 10.

The EMS 20 collects statistical data of each NE 10 and stores the statistics data in the DB 40. In addition, the EMS 20 transmits a completion message indicating that the collection of the statistical data for the corresponding NE 10 in which the statistical data has been collected to the agent 30 through IPC (Inter-Process Communication). .

At this time, the EMS 20 transmits a completion message to the agent 30 according to the index information allocated to the NE 10 on which the statistical data collection is completed.

When the completion message is received from the EMS 20, the agent 30 records that the collection of statistical data for the NE 10 for the index table 31 was successful. At this time, the index table 31 has the same number of recording spaces as the total number of each NE 10 in the network, and each recording space is arranged to correspond to the index information allocated to each NE 10.

2 is an internal block diagram illustrating an agent according to a preferred embodiment of the present invention.

Referring to FIG. 2, the agent 30 includes an index table 31, a MIB table 32, a timer processor 34, and a data processor 33.

The index table 31 has as many recording spaces as the number of each NE 10 in the network, and statistical data is received from each NE 10, or information of reception failure is stored.

The MIB table 32 has a recording space equal to the number of each NE 10 in the network, and stores the statistical data collected by the EMS 20 from each NE 10 and stored in the DB 40. 34 is driven according to a preset timer time value, and when the timer time value expires, the timer 34 provides a timer expiration signal to the data processor 33.

At this time, the timer time value set in the timer processing unit 34 is a time (a) taken to collect / process statistical data from each NE 10 managed by the EMS 20, and the EMS 20 is configured as an agent ( 30) the time it takes to send the TRAP message and receive the response message (b), and the time required for the agent 30 to access and store the statistical data stored in the DB 40 in the MIB table 32. The time c is set to the sum of time d and the time d taken by the agent 30 to transmit the TRAP message according to the success and failure recorded in the index table 31 to the management server 50.

When the completion message is received from the EMS 20, the data processing unit 33 records success information in the recording space of the index table 31 corresponding to the NE 10, and the timer processing unit 34 from the timer processing unit 34. If no completion message is received by the time when the expiration signal is received, failure information indicating that reception of the statistical data has failed is recorded in the recording space where success is not recorded in the index table 31.

Then, the data processing unit 33 reads the statistical data of each NE 10 whose success is recorded in the index table 31 from the DB 40 and stores it in the MIB table 32. At this time, the data processing unit 33 preferably reads only statistical data for the corresponding NE 10 whose success is recorded in the index table 31, and the identification information for each NE 10 is stored in the MIB table 32. That is, it stores the unique information assigned to each NE (10) or IP address information used in the network, and corresponding statistical data.

That is, the data processor 33 reads each of the statistical data corresponding to the unique information or the IP address information of the NE 10 that is recorded in the index table 31 as success, and the unique information of the corresponding NE 10 or the like. It is stored in the recording space of the MIB table 32 corresponding to the IP address information.

Then, after the data processing unit 33 transmits a TRAP message according to the success or failure information recorded in the index table 31 to the management server 50, the management server 50 collects statistical data from the MIB table 32. Upon reading, the timer processor 34, the index table 31, and the MIB table 32 are initialized.

That is, when the data processing unit 33 records all successes in the recording space of the index table 31, the data processing unit 33 transmits a success TRAP message to the management server 50, and when all records in the recording spaces of the index table 31 as failures. , Transmits the failed TRAP message to the management server (50).

If failure or success are mixed in the recording space of the index table 31, the data processing unit 33 includes a success TRAP message including IP address information of the NE 10 corresponding to the recording space in which the success is recorded; A failure TRAP message including IP address information of the NE 10 corresponding to the recording space recorded as failure is transmitted to the management server 50, respectively.

In other words, when success is recorded in the recording space of the index table 31 before the timer processing unit 34 expires, the agent 30 accesses the statistical data stored in the DB 40 by the EMS 20 to the MIB. After storing in the table 32, a success TRAP message is transmitted to the management server 50.

At this time, the agent 30 preferably checks the index table 31 and accesses only the statistical data for the NE 10 in which success has been recorded. That is, since the agent 30 does not store the statistical data for the NE 10 for which the completion message is not received from the EMS 20 in the DB 40, the agent 30 accesses the statistical data for the NE 10. Is unnecessary. Therefore, since the agent 30 accesses only the statistical data for the corresponding NE 10 in which the success is recorded in the index table 31, the agent 30 reduces the time required for processing the statistical data, and the agent 30 Onboard processor load can be minimized.

And, if the success of the index table 31 is not recorded in the state in which the timer processing unit 34 has expired, the agent 30 waits to receive a completion message for each NE 10 from the EMS 20. Stop, and transmits a failure TRAP message to the management server (50).

On the other hand, the agent 30, if the success or failure in the index table 31 mixed with the timer processing unit 34 expired, the statistical data for the corresponding NE (10) recorded as success in the index table 31 Is accessed from the DB 40 and stored in the MIB table 32.

In addition, the agent 30 includes a success TRAP message including the index information or the IP address information of the NE 10 in which the success is recorded in the index table 31, and the index information or the IP address information of the NE 10 in which the failure is recorded. Transmits a failed TRAP message to the management server 50.

The agent 30 includes IP address information, which is identification information of the corresponding NE 10, in which all successes or failures are recorded in the index table 31 in a variable bind in a TRAP message in an IP: IP: IP format. To generate a success or failure TRAP message. At this time, the total number of NE 10 identification information in the Variable Bind of the success TRAP message or the failure TRAP message must match the total number of NE 10 managed by the agent 30.

On the other hand, when the TRAP message is received from the agent 30, the management server 50 reads the statistical data stored in the MIB table 32 and provides it according to the request of the management terminal.

At this time, when the management server 50 receives the success TRAP message from the agent 30, the management server 50 performs a GET-NEXT command among the SNMP commands to read the statistical data stored in the MIB table 32. The management server 50 grasps index information in table units rather than continuously performing GET-NEXT using one Variable Bind, and executes the Multi Variable Bind GET-NEXT command in table units according to the identified index information. To minimize the time to collect statistical data.

When the TRAP message is not received from the agent 30 for a predetermined time, the management server 50 determines that a failure has occurred in the agent 30, and transmits an agent failure occurrence message to each management terminal 60. That is, the management server 50 may check whether the failure of the agent 30 occurs through the TRAP message received from the agent.

3 is a flowchart illustrating a flow of a method for managing a network according to a preferred embodiment of the present invention.

Referring to FIG. 3, each NE 10, which is a component of the network, independently transmits statistical data to the EMS 20 according to the set time information (S 10).

In this case, when the NE 10 succeeds in transmitting the statistical data to the EMS 20, the NE 10 transmits a success message indicating that the statistical data transmission is successful to the EMS 20, and fails in transmitting the statistical data. The NE 10 transmits a failure message indicating that the statistical data transmission has failed to the agent 30 through the EMS 20.

The EMS 20 transmits the completion message for the corresponding NE 10 to the agent 30 after storing the received statistical data in the DB 40 when the statistical data is received from each NE 10. S 20).

That is, the EMS 20 receives statistical data independently from each NE 10, and uses a method previously promised with the agent 30, for example, IP address information of each NE 10 as identification information. Can be stored in the DB (40).

The EMS 20 normally stores the statistical data in the DB 40 and then transmits a completion message for the corresponding NE 10 to the agent 30 according to the IPC.

When the completion message is received from the EMS 20, the agent 30 records success in the recording space of the index table 31 corresponding to the corresponding NE 10 (S 30).

The agent 30 then checks whether statistical data is received from the entire NE 10 of the network. In other words, the agent 30 checks whether all successes are recorded in the index table 31 (S40).

If the statistical data is not received from the entire NE 10 during the collection time for collecting the statistical data, the agent 30 transmits a failed TRAP message to the management server 50, and the management server 50 receives the failed TRAP message. If received, the MIB table 32 of the agent 30 is not accessed. That is, the management server 50 does not access the agent 30 because there is no statistical data in the agent 30.

The agent 30 checks whether or not the timer processing unit 34 in which the predetermined timer time value is set expires while receiving statistical data from each NE 10 (S50).

In this case, the timer time value is a time (a) taken to collect / process statistical data from each NE 10 managed by the EMS 20, and the EMS 20 transmits a TRAP message to the agent 30. , Time (b) for receiving a response message thereto, time (c) for agent 30 to access and store statistical data stored in DB 40 in MIB table 32, agent ( 30) may be a time value obtained by adding up the time ds required to transmit the TRAP message according to the successes and failures recorded in the index table 31 to the management server 50.

If the timer processing unit 34 expires without receiving statistical data from the entire NE 10, the agent 30 records a failure in the recording space of the index table 31 in which success is not recorded (S 60). .

Then, the agent 30 reads the statistical data of the NE 10 corresponding to the record space of the successive recorded index table 31 from the DB 40 and stores it in the MIB table 32 (S 70). That is, the agent 30 accesses the received statistical data from the DB 40 and stores it in the MIB table 32.

The agent 30 records a success TRAP message in which the IP address information of the NE 10 corresponding to the recording space in which the success is recorded in the index table 31 is included in the variable bin, and the failure is recorded in the index table 31. A failure TRAP message including the IP address information of the NE 10 corresponding to the space is generated and transmitted to the management server 50 (S80).

That is, the agent 30 accesses the DB 40 and accesses the MIB table 32 when success is recorded in the index table 31 for all NEs 10 managed before the set timer time value expires. Update the statistical data, and send a success TRAP message to the management server 50.

On the other hand, if there is a recording space in which the failure is recorded in the index table 31 while the timer time value has expired, the agent 30 checks the NE 10 corresponding to the recording space in which the success is recorded in the index table 31. Only the statistical data is updated in the MIB table 32, and the successful TRAP message is transmitted by including the IP address information of the NE 10 in the format of IP: IP: IP in the Variable Bind.

The agent 30 transmits the failed TRAP message to the management server 50 by including the IP address information of the NE 10 corresponding to the recording space in which the failure is recorded in the index table 31 in the Variable Bind.

The management server 50 reads statistical data of the corresponding NE 10 from the MIB table 32 according to IP address information, which is unique information of each NE 10 included in the success TRAP message (S 90).

On the other hand, the management server 50 may grasp the information of the NE (10) has a failure through the failure TRAP message received from the agent (30). That is, if the IP address information of the NE 10 included in the received failed TRAP message is repeated for a predetermined number of times, the management server 50 determines that a failure that cannot transmit statistical data from the NE 10 occurs and manages it. The failure occurrence message may be transmitted to the terminal 60 (S 100).

If the TRAP message is not received for a predetermined time, the management server 50 determines that an abnormality has occurred in the agent 30, and transmits an agent failure message to the management terminal 60.

After the agent 30 transmits each TRAP message to the management server 50, the agent 30 performs the next number of statistical data collection operations after initializing the index table 31 and the timer processor 34 (S 110).

The management server 50 performs a GET-NEXT command in table units according to the unique information of each NE 10 included in the variable bin of the success TRAP message received from the agent 30, and manages the agent 30. The statistical data of the NE 10 is read from the MIB table 32.

Although the present invention has been described in detail only with respect to the described embodiments, it will be apparent to those skilled in the art that various modifications and changes are possible within the technical spirit of the present invention, and such modifications and modifications belong to the appended claims.

As described above, according to the present invention, the network management system minimizes the time spent collecting / managing statistical data for each NE to determine whether the normal collection of statistical data of each NE is performed as a success message or failure message to the management server. By sending it so that the management server can easily see the status of statistical data collection for the entire NE.

In addition, the network management system processes the statistical data according to the operation of the timer set at a predetermined time value, thereby limiting the limitations generated while managing the statistical data according to SNMP, that is, the restriction of completing the collection of the statistical data within a limited time. The branch management server can minimize the waiting time for receiving the statistical data for the failed NE to secure enough time for periodic statistical data collection, and can also process a large amount of statistical data.

In addition, the management server may check the NE or agent failure situation through the received TRAP message.

Claims (21)

A network management system comprising a plurality of network elements (NEs) for transmitting statistical data at predetermined intervals, a database (DB) for storing the statistical data, and a management terminal for managing a network, An EMS (Element Management System) for providing a completion message including information on the corresponding NE that has received the statistical data after storing the respective statistical data in the DB when the reception of the statistical data independently from each NE is completed; According to the information of the NE included in the completion message provided from the EMS within a predetermined time to access the statistical data stored in the DB and stored in a table form to provide a success message or failure message containing the information of each NE Agent, A management server for accessing statistical data stored in the form of a table according to the information of the NE included in the success message provided from the agent, and managing a failure state of each NE and the agent through the failure message. Statistical data management device in the management system. The method of claim 1, wherein the EMS, And receiving the statistical data from each of the NEs, and storing the information of each of the NEs and the statistical data received from the NEs during the corresponding period into the DB. The method of claim 1, wherein the agent, A timer processor that is driven according to a predetermined timer value and provides an expiration message when the timer expires; An index table for storing statistical data reception success or failure information from each NE in a recording space corresponding to each NE; A data table for storing the unique information of each NE, and statistical data received from the NE; Success information is recorded in the recording space of the index table corresponding to the information of the NE included in the completion message received from the EMS, and when the expiration message is received from the timer processor, the corresponding NE in which the success information is recorded in the index table. And a data processing unit for transmitting the success message including the information of each NE to the management server after accessing the statistical data received from the DB and storing the data in the data table. The method of claim 3, wherein the data processing unit, If success is recorded in all recording spaces of the index table before the expiration message is received from the timer processor, a success message including the information of each NE is transmitted to the management server. Data management device. The method of claim 3, wherein the data processing unit, When the expiration message is received from the timer processor, a success message including information of NE corresponding to the recording space in which the success information is recorded after the failure information is recorded in the recording space in which the success information is not recorded in the index table; And a failure message including information of the NE corresponding to the recording space in which the failure information is recorded, to the management server. The method of claim 3, wherein the predetermined timer value, Collecting statistics data from each NE managed by EMS and storing them in the DB, EMS sending a completion message to the agent and receiving a response message, Agent accessing the statistical data in the DB Statistical data management of a network management system, characterized in that the time value stored in the table and the time that the agent sends the success message or failure message according to the success and failure information recorded in the index table to the management server Device. The method of claim 1, wherein the management server, And access only a storage area of the data table corresponding to the information of the NE included in the success message received from the agent. The method of claim 1, wherein the management server, And when the failure message is not received for a predetermined time, transmitting the failure occurrence message of the agent to the management terminal. The method of claim 1, wherein the management server, If the information of the NE included in the failure message is repeatedly received more than a predetermined number of times, it is determined that the failure has occurred in the NE, and the statistical data management device of the network management system, characterized in that for transmitting the failure occurrence message to the management terminal. . The method of claim 1, wherein the NE information, And at least one of identification information and IP address information allocated to the NE. The method of claim 1, wherein the EMS, Statistical data management device of the network management system, characterized in that for transmitting the completion message to the agent via IPC (Inter Process Communication). The method of claim 1, wherein the management server, And access statistical data corresponding to the information of the NE included in the success message received from the agent according to the Simple Network Management Protocol (SNMP) from the agent. A plurality of network elements (NEs) for transmitting statistical data at predetermined intervals, a database (DB) for storing the statistical data, a management terminal for managing a network, an element management system (EMS) for managing the statistical data, and In the statistical data management method of a network management system including an agent and a management server, After the EMS stores each statistical data independently received from each NE in the DB, transmitting a completion message including the information of the corresponding NE that has received the statistical data to the agent; Recording, by the agent, success information in a recording space corresponding to the information of the NE included in the completion message received within a predetermined time; When the agent has expired, accessing the statistical data corresponding to information of the NE corresponding to a recording space in which the success information is recorded from the DB and storing the statistical data in a data table; Storing, by the agent, the statistical data in the data table, and transmitting a success message including the information of each NE to the management server; And providing, by the management server, to the management terminal after accessing and storing the statistical data in the data table according to the information of the NE included in the success message. The method of claim 13, Recording, by the agent, failure information in the recording space in which success information is not recorded when the predetermined time expires; Transmitting, by the agent, a failure message including information of NE corresponding to a recording space in which the failure information is recorded to the management server; The agent transmitting the failure message or the success message and initializing the recording space and the predetermined time; Transmitting, by the management server, a failure message including information of the NE and failure occurrence notification information to the management terminal when the information of the NE included in the failure message is repeatedly received a predetermined number of times; And transmitting, by the management server, the agent failure message to the management server when the failure message or the success message is not received from the agent for the predetermined period of time. The method of claim 13, wherein the success message or failure message, And generating and including the information of the NE by the agent according to SNMP. The method of claim 13, wherein the predetermined time, Collecting statistics data from each NE managed by EMS and storing them in the DB, EMS sending a completion message to the agent and receiving a response message, Agent accessing the statistical data in the DB And a time for storing the time stored in the table and the time for transmitting the success message or the failure message according to the success and failure information recorded in the index table to the management server. . The method of claim 13, wherein the DB storage of the statistical data, And storing the information of each NE and statistical data received from the NE during a corresponding period in a DB in a table form. The method of claim 13, The agent, if all the success information is recorded in the recording space within the predetermined time, transmitting the success message including the information of each NE to the management server; The agent further comprises the step of initializing after transmitting the success message. The method of claim 13, wherein the agent, Statistical data management of the network management system, characterized in that to access only the statistical data in accordance with the information of the NE corresponding to the recording space in which the success information is recorded from the DB, the information of the NE and the data in the form of a table Way. The method of claim 13, wherein the NE information, And at least one of identification information or IP address information allocated to the NE. The method of claim 13, wherein the transmission of the completion message, And transmitting the EMS to the agent through IPC (Inter Process Communication).

Images