JPH10327148A - Communication network management method, communication network management system realizing the method and recording medium recording program of the method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To avoid traffic due to polling from being increased by changing a polling interval dynamically and adaptively depending on a state of a network at that time. SOLUTION: The management system is configured such that a management device 1 sends a diagnostic request to a device 2 being a management object at a prescribed interval and detects a state change in the device 2 being the management object by awaiting a reply from the device 2 being the management object. Then the interval of the diagnostic request sent from the management device 1 is varied with a status of a traffic of a network 3 or with traffic information of the network obtained from a channel monitor.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for managing communication devices and servers connected to a network, and more particularly, to a communication network management method using a network management protocol requiring polling such as SNMP.
The present invention also relates to a communication network management system in which the method is realized, and a recording medium on which the method is programmed and recorded.

[0002]

2. Description of the Related Art In general, a network management system is configured so that a function and a device or software for realizing the function are separately provided for a managing side and a managed side. here,
The managing side is called a management system or a manager, and the managed side is called a managed object or an agent. Normally, a communication protocol for exchanging management information is defined between the management system and the management target. For example, in the Internet, as a standard management protocol,
SNMP (Simple Network Management Prot) specified by IETF (Internet Engineering Task Force)
ocol).

In SNMP, a means called polling is used as a means for the management system to grasp the state of a management target. In polling, the management system sends a request message to the management target at regular intervals to inquire about its status, and the management target responds to the request message. The system that the management system detects inside is taken.

[0004]

[0007] The polling time interval used in the above-mentioned conventional network management system has been mainly fixedly set in advance. When the number of managed objects increases, even messages for polling alone cannot be ignored as a cause of increasing traffic on the network, and particularly in the case of congestion, sending and receiving messages for realizing the main communication function is fatally affected. There was a problem.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and avoids an increase in traffic due to polling by dynamically adjusting and changing a polling interval according to the state of the network at that time. It is an object to provide a network management system, a management method in the system, and a memory medium on which the method is programmed and recorded.

[0006]

A network management method in a communication network system according to the present invention issues a diagnosis request at regular intervals from a management device to a device to be managed, and waits for a response from the device to be managed. In a network management system for detecting a change in the state of a device to be managed, an interval between diagnosis requests issued by the management device is variable according to the status of network traffic.

Further, the communication network management system according to the present invention comprises: (1) a network management device and a device to be managed are connected via a network line, and the network management device periodically issues a diagnosis request; Generating means for receiving a diagnostic response issued from the device to be managed in response to the diagnostic request, measuring a time difference between the request and the response, and reflecting the time difference in the diagnostic request generating cycle in the diagnostic request generating cycle determining means. The interval between diagnosis requests issued by the network management device is made variable in accordance with the degree of network congestion. (2) A network management device and a device to be managed are connected via a network line, and a line monitor is connected to the network line, and the network management device periodically generates a diagnosis request. And traffic information collecting means for collecting traffic information measured by the line monitor, and a diagnosis request generation cycle determining means for reflecting the traffic information in the diagnosis request generation cycle, the interval between the diagnosis requests issued by the network management device, It is characterized by being variable according to the status of network traffic.

Further, the memory medium on which the communication network management method of the present invention is programmed and recorded has the following features: (1) The interval of the generation of a diagnostic request is made variable according to the status of network traffic. A diagnostic request is periodically issued to a target device, a diagnostic response is received from a device to be managed, and a time difference between the request and the response is measured and reflected in a generation cycle of a diagnostic request issued from the management device. It is characterized by being programmed.
(2) The interval at which a diagnosis request is generated is made variable in accordance with network traffic information obtained from a line monitor, and the management device periodically issues a diagnosis request to a device to be managed and the device to be managed. The diagnostic program is programmed to collect a traffic response measured by a line monitor, and to reflect the traffic information in a diagnosis request generation cycle.

As a result, the polling interval can be dynamically adapted and changed in accordance with the state of the network at that time, and a network management system which avoids an increase in traffic due to polling, and a management method and method in the system Can be provided and constructed in which a memory medium is programmed and recorded.

[0010]

FIG. 1 is a block diagram showing an embodiment of the present invention. In the figure, reference numeral 1 denotes a management system, which is assumed to be a PC server. 2 is a management target,
Network devices such as routers and hubs are assumed. The management system 1 and the management target 2 are interconnected by a LAN line 3.

The management system functionally includes a diagnostic signal transmitting unit 11, a diagnostic response receiving unit 12, a management target list 13,
Reaction time measurement timer 14, delay calculation unit 15, polling activation timer 16, cycle determination unit 17, delay cycle correspondence table 18
It consists of. These are all realized by software, take the form of network management software, are stored on a floppy disk or CDROM, and are provided as middleware.

Hereinafter, a mechanism for exchanging a diagnosis request and a diagnosis response for the management system 1 to grasp the state of the management target 2 will be described in detail with reference to a flowchart shown in FIG. First, the polling start timer 16
Activates the diagnostic signal transmission unit 11 when the set cycle expires. The activated diagnostic signal transmitting unit 11 refers to the management target list 13 and knows that the management target in the next order is the management target 2, sends a diagnosis request to the management target 2, and measures the reaction time. The timer 14 is reset.

The management target 2 returns a diagnosis response to the diagnosis request. The diagnostic response receiver 12 receives the diagnostic response and sends a reception notification to the delay calculator 15. The delay calculating unit 15 sets the response time measurement timer 1
By reading the time difference information from No. 4, the time difference between the diagnosis request and the diagnosis response is measured. Delay time calculation unit 1
In 5, a delay time is obtained by performing an appropriate operation such as taking an average value from the data of some past time differences, and is transmitted to the cycle determination unit 17. The cycle determination unit 17 provides a delay cycle correspondence table 18
, A cycle corresponding to the delay time is obtained, and the cycle of the polling start timer is reset.

Here, the management target list 13 is a list including the address information of a plurality of management targets 2 to be managed by the management system 1, and is prepared in advance by a user or automatically generated by the system. May be. FIG. 3 shows the structure of the delay period correspondence table 18. When the minimum delay time value exceeding the calculated delay time is searched from the table, the value of the polling cycle to be set is written in that row. The learning table may be prepared in advance based on experimental data or may be updated repeatedly based on experience.

In the delay time calculating section 15, besides a method of taking a simple average of a predetermined number of continuous time difference information, a method of appropriately weighting each time difference information, and a method of applying the latest time difference information directly to the delay time And the like.

By dynamically adjusting the cycle set in the polling start timer 16 according to the time difference between the diagnosis request and the response as described above, when the network becomes congested,
Since the time difference is increased, the polling period can be increased, and the diagnostic signal can be further prevented from causing network congestion.

FIG. 4 shows another embodiment of the present invention. In the figure, 1 is a management system, 2 is a management target, and LA
They are connected by an N line 3. A LAN monitor 4 for monitoring the state of the LAN line 3 is connected. The management system 1 functionally includes a diagnostic signal transmitting unit 11, a diagnostic response receiving unit 12, a management target list 13, a polling start timer 16, a cycle determining unit 17, a traffic information collecting unit 19, a congestion degree calculating unit 20, The congestion degree-cycle correspondence table 21 is configured.

Hereinafter, the operation of the embodiment shown in FIG. 4 will be described in detail with reference to the flowchart shown in FIG. In the embodiment shown in FIG. 4, the congestion degree of the network is determined from information measured by the LAN monitor 4 directly attached to the network. As in the embodiment shown in FIG. 1, the management system 1 and the management target 2 are LAN-connected as shown. A LAN monitor 3 is attached to the LAN line 3, and the LAN monitor 3 measures traffic information such as a LAN usage rate and a collision occurrence frequency.
The traffic information measured by the LAN monitor 4 is sent to the traffic information collecting unit 19. The traffic information collection unit 19 sends the traffic information collected at regular intervals to the congestion degree calculation unit 20. In the convergence degree calculation unit 20,
Appropriate calculation is performed on values such as the LAN usage rate and the frequency of occurrence of collision to calculate the congestion degree and send it to the cycle determination unit 17.
The cycle determining unit 17 obtains a cycle from the congestion degree-period correspondence table 21 and the congestion degree, and sets the polling start timer 16.

As an arithmetic expression for calculating the degree of congestion from traffic information, for example, the following expression (1) can be considered.

Congestion degree = LAN use rate [%] + collision occurrence frequency [times / second] × 10 (1) FIG. 6 shows the structure of the congestion degree-period correspondence table 21. The cycle to be set at that time is entered in the line having the minimum congestion degree larger than the corresponding congestion degree.

When a plurality of LANs exist between the management system 1 and the management target 2, the plurality of LANs
It is conceivable that the LAN monitor 4 is installed in the AN, and all data collected from the LAN monitor 4 is handled as traffic information. In the case where there is a plurality of LANs that are particularly slow in speed and are in the overall rate stage, it is possible to handle only the traffic information there.

With the above configuration, the polling cycle can be dynamically adjusted according to the degree of LAN congestion.
When the AN is congested, it is possible to prevent the congestion from being further strengthened by the diagnostic signal.

Incidentally, the above-mentioned method is used as program information which can be executed by a computer, for example, on a recording medium such as a magnetic disk (floppy disk, hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), a semiconductor memory, etc. It is of course possible to write and apply to various devices, or transmit to a communication medium and apply to various devices. The computer that realizes this device is
The program information recorded on the recording medium is read, and the above-described processing is executed by controlling the operation according to the program information.

[0024]

As described above, according to the present invention, the polling interval is dynamically adapted and changed according to the state of the network at that time, whereby the network in which the increase in traffic due to polling is avoided. A management system, a management method in the system, and a memory medium on which the method is programmed and recorded can be provided and constructed. In addition, the following effects can be obtained.

(1) Based on the degree of network congestion, for example, by increasing the polling interval, when the network is congested, the network load increases due to the exchange of management diagnosis request / response messages, and the network is originally transferred. Conventional harmful effects such as an increase in delay of transmission / reception of a packet which must be performed and discarding can be prevented.

(2) Even if the route between the management device and the management target is complicated, it is possible to comprehensively evaluate the degree of network congestion during the route.

(3) It is possible to reduce the error of the congestion degree measurement due to a short-time load change of the network based on the past several data.

(4) By adding a LAN monitor, it is possible to more accurately measure the degree of network congestion. Further, there is an effect that the congestion degree can be measured even when a plurality of networks are interposed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a flowchart cited for explaining the operation of the embodiment shown in FIG. 1;

FIG. 3 is a view showing the structure of a delay cycle correspondence table shown in FIG. 1;

FIG. 4 is a block diagram showing another embodiment of the present invention.

FIG. 5 is a flowchart cited for explaining the operation of the embodiment shown in FIG. 4;

FIG. 6 is a diagram showing a structure of a congestion degree cycle correspondence table shown in FIG. 4;

[Explanation of symbols]

1 management system, 2 management target, 3 LAN line, 4
... LAN monitor, 11 ... diagnosis signal transmission unit, 12 ... diagnosis response reception unit, 13 ... management list, 14 ... reaction time measurement timer, 15 ... delay calculation unit, 16 ... polling start timer, 17 ... cycle determination unit, 18 ... Delay period correspondence table, 19 ...
Traffic information collecting unit, 20 ... congestion degree calculating unit, 21 ...
Congestion degree cycle correspondence table

Claims (19)

[Claims] In a network management system, a management device issues a diagnosis request to a device to be managed at regular intervals, waits for a response from the device to be managed, and detects a change in state of the device to be managed. A communication network management method, characterized in that the interval between diagnosis requests issued by the management device is variable according to the status of network traffic. 2. The traffic situation in the network is based on a time required from when a management device issues a diagnosis request to when a response message from a device to be managed arrives as an index. The communication network management method as described. 3. The communication according to claim 2, wherein the status of the traffic in the network is obtained by performing an appropriate operation on a required time required for each response message to arrive in response to several past diagnostic requests as an index. Network management method. 4. The communication network management method according to claim 1, wherein the traffic index in the network is calculated based on a network usage rate or a packet collision rate collected by a line monitor connected to the network. 5. Indices of traffic in a network include a plurality of networks when a plurality of networks exist between a management device and a device to be managed.
5. The communication network management method according to claim 4, wherein the calculation is performed based on a network usage rate or a packet collision rate collected by a line monitor attached to the network having the highest network usage rate and the overall rate-limiting step. . 6. A network management device and a device to be managed are connected via a network line. The network management device periodically generates a diagnosis request, and generates a diagnosis request in response to the diagnosis request. A diagnostic request generation cycle determining means for receiving a diagnostic response issued from the target apparatus, measuring a time difference between the request and the response and reflecting the time difference in the diagnostic request generation cycle, and setting an interval between the diagnostic requests issued by the network management apparatus, A communication network management system characterized by being variable according to the degree of network congestion. 7. The network management device further includes: an arithmetic unit configured to perform an appropriate operation on a required time required to arrive at each diagnostic response to the past several diagnostic requests to generate a delay time;
A delay period correspondence table in which a delay time and a diagnosis request generation period optimal for the delay time are stored as a set; and a diagnosis request generation period determination unit refers to the delay period correspondence table to determine an optimum diagnosis request period. 7. The communication network management system according to claim 6, wherein an occurrence period is obtained. 8. The communication network system according to claim 7, wherein said calculating means calculates the delay time by taking an average value from a plurality of past time difference data. 9. The communication network management system according to claim 7, wherein said calculating means appropriately weights each time difference information and calculates a delay time according to the weighting coefficient. 10. The communication network system according to claim 7, wherein the calculating means calculates the delay time by passing the latest time difference information. 11. The delay cycle correspondence table is characterized in that a table index is obtained from the minimum delay time value exceeding the calculated delay time, and the value of the cycle to be set at that time is written in the entry. The communication network system according to claim 7, wherein 12. A network management device and a device to be managed are connected via a network line, a line monitor is connected to the network line, and the network management device periodically issues a diagnosis request. Generating means, traffic information collecting means for collecting traffic information measured by the line monitor, and diagnostic request generation cycle determining means for reflecting the traffic information in the diagnostic request generation cycle; Communication network management system, wherein the communication network management system is configured to be variable according to the status of network traffic. 13. The network management apparatus further includes an arithmetic unit for performing an appropriate operation on traffic information obtained from the line monitor to calculate a congestion degree, and a set of the congestion degree and a diagnosis request generation cycle optimal for the congestion degree. A congestion degree-period correspondence table stored as
13. The communication network management system according to claim 12, wherein an optimum diagnosis request generation cycle is obtained by referring to the congestion degree-cycle correspondence table. 14. The method according to claim 13, wherein the calculating means obtains information on the line utilization rate and the frequency of occurrence of collision via the traffic information collecting means, and calculates the congestion degree by performing the following calculation *. Communication network system. * Congestion degree = line usage rate (%) + collision frequency (times / second)
* 10 15. The congestion degree cycle correspondence table, wherein a cycle to be set at that time is entered in an entry having a minimum congestion degree larger than the corresponding congestion degree.
3. The communication network system according to 3. 16. An apparatus according to claim 1, wherein the interval between the generation of the diagnosis request is made variable in accordance with the status of the traffic on the network. A recording medium which is programmed and recorded so as to receive a response, measure a time difference between the request and the response, and reflect the time difference in a generation cycle of a diagnosis request issued from the management device. 17. A delay time is generated by performing an appropriate operation on a required time required for each diagnostic response to arrive in the past several diagnostic requests, and a delay time and a diagnostic request generation cycle optimal for the delay time are set. 17. The recording medium according to claim 16, wherein the recording medium is programmed so as to obtain an optimum diagnosis request generation period by indexing a delay period correspondence table stored as follows. 18. An apparatus according to claim 1, wherein an interval between the generation of the diagnosis request is made variable according to the traffic information of the network obtained from the line monitor. A recording medium which receives a diagnostic response from an apparatus, collects traffic information measured by a line monitor, and is programmed and recorded so as to reflect the traffic information in a diagnostic request generation cycle. 19. A congestion degree cycle in which an appropriate operation is performed on traffic information obtained from a line monitor to calculate a congestion degree, and the congestion degree and a diagnosis request generation cycle optimal for the congestion degree are stored as a set. 19. The recording medium according to claim 18, wherein the recording medium is programmed and recorded so as to obtain an optimum diagnosis request generation cycle by indexing a correspondence table.