JP7400565B2 - Management devices, network systems, management methods, and programs - Google Patents

Description

The present invention relates to a management device and the like that manage network resources.

For controlling network routes in multipath routing, techniques such as distance vector routing protocols, link state routing protocols, and path vector routing protocols are used, for example. Since these protocols determine routes between terminals, routes tend to overlap and traffic tends to concentrate at specific locations. In addition, these protocols tend to change routes for fault tolerance, and although they provide sufficient performance, they lack the ability to detect signs that throughput may decline if left as is.

For example, in the L2 layer, a route can be selected flexibly using a technology such as OpenFlow. However, the OpenFlow method can only be applied to models that support OpenFlow, and cannot be applied to general switches.

Patent Document 1 discloses that before a failure occurs, a route is selected in consideration of the performance differences of each device, and a management information base (MIB) and network flow are incorporated as standards. A method of using the function is disclosed.

Patent Document 2 discloses a control method for centrally managing communication routes based on performance data collected from each of server devices and network devices forming a group of servers that share a virtual address. In the method disclosed in Patent Document 2, an optimal route for reaching the optimal server device selected based on the performance data of the server device is determined based on the performance data of the network device.

Patent Document 3 discloses a network control method using a controller that controls flows of a plurality of nodes. In the method of Patent Document 3, when a flow passing through a certain node is moved to another alternative route, the source and destination of the route corresponding to the flow to be moved match, and the flow to be moved passes through a certain node. Find multiple alternative route candidates that do not pass through. In the method of Patent Document 3, an alternative route is determined from among a plurality of alternative route candidates according to predetermined criteria, and flows of each node on the alternative route are aggregated.

Japanese Patent Application Publication No. 2018-148382 International Publication No. 2012/101890 International Publication No. 2011/043312

According to the method disclosed in Patent Document 1, route optimization for each device is realized. However, the method of Patent Document 1 has a local optimization scope, and is difficult to apply to distributing the load of the entire network.

According to the method disclosed in Patent Document 2, communication routes can be centrally managed based on performance data of server devices and network devices. However, since the method of Patent Document 2 determines the optimal route based on relatively different performance data depending on the device, it is difficult to apply the method to controlling the route of the entire network.

According to the method disclosed in Patent Document 3, the number of flows passing through nodes can be reduced, and the number of flow information entries registered in each node can be reduced. However, although the method of Patent Document 3 can be used to determine an alternative route, it is difficult to apply to controlling the route of the entire network.

An object of the present invention is to provide a management device and the like that can optimize network routes without relying on the resources of network devices that make up the network.

A management device according to one embodiment of the present invention includes a collection unit that collects statistical data of a plurality of network devices that constitute a network, and a group that is configured by a plurality of network devices that are related among the plurality of network devices. a generation unit, a calculation unit that calculates the load factor of the network device and the group load factor of the group using statistical data, and when the load factor exceeds a threshold, determines the load status of the group based on the group load factor. and an updating unit that updates the network route based on the load status of the group.

In a management method according to one aspect of the present invention, a computer collects statistical data of a plurality of network devices that constitute a network, and selects a group formed by a plurality of related network devices among the plurality of network devices. The load factor of the network device and the group load factor of the group are calculated using the statistical data, and if the load factor exceeds the threshold, the load status of the group is determined based on the group load factor, and the load factor of the group is calculated based on the group load factor. Update network routes based on load conditions.

A program according to one embodiment of the present invention includes a process for collecting statistical data of a plurality of network devices that constitute a network, and a process for generating a group made up of a plurality of network devices that are related among the plurality of network devices. and a process of calculating the load factor of the network device and the group load factor of the group using statistical data, and a process of determining the load status of the group based on the group load factor when the load factor exceeds a threshold value. , and updating the network route based on the load status of the group.

According to the present invention, it is possible to provide a management device and the like that can optimize network routes without relying on the resources of network devices that constitute a network.

FIG. 1 is a block diagram showing an example of the configuration of a network system according to a first embodiment. FIG. 1 is a conceptual diagram showing an example of a network included in the network system according to the first embodiment. This is an example of an analysis table that summarizes the results of the analysis of information included in the management information base of the switches that make up the network in FIG. 2, by the management device of the network system according to the first embodiment. FIG. 2 is a conceptual diagram showing another example of a network included in the network system according to the first embodiment. 4 is an example of an analysis table summarizing the results of the analysis of information included in the management information base of the switches forming the network of FIG. 4, by the management device of the network system according to the first embodiment. This is an example of a table obtained by sorting the analysis table of FIG. 5 using the transmission source and transmission destination as keys, by the management device of the network system according to the first embodiment. This is an example of a table that summarizes groups generated by the network system management device according to the first embodiment based on the table of FIG. 6. FIG. 1 is a conceptual diagram showing an example in which a network system according to a first embodiment includes a plurality of networks. FIG. 2 is a conceptual diagram showing an example of a group generated from a plurality of networks included in the network system according to the first embodiment. 2 is a flowchart for explaining an example of the operation of the management device of the network system according to the first embodiment. 3 is a flowchart for explaining an example of route optimization processing by the management device of the network system according to the first embodiment. FIG. 2 is a conceptual diagram showing an example of the load status of a network configured with 12 nodes. FIG. 2 is a conceptual diagram for explaining an example of changing a network route using a method according to related technology. FIG. 2 is a conceptual diagram for explaining an example of changing a network route using the method according to the first embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of a management device according to a second embodiment. FIG. 2 is a block diagram illustrating an example of a hardware configuration that implements a management device according to each embodiment.

EMBODIMENT OF THE INVENTION Below, the form for implementing this invention is demonstrated using drawing. However, although the embodiments described below include technically preferable limitations for carrying out the present invention, the scope of the invention is not limited to the following. In addition, in all the figures used for the description of the following embodiments, the same reference numerals are given to the same parts unless there is a particular reason. Furthermore, in the following embodiments, repeated explanations of similar configurations and operations may be omitted. Further, the directions of the arrows in the drawings are merely examples, and do not limit the directions of signals between blocks.

(First embodiment)
First, a network system according to a first embodiment of the present invention will be described with reference to the drawings. The network system of this embodiment includes a communication network (also referred to as a network) configured by a plurality of network devices such as switches and routers.

(composition)
FIG. 1 is a block diagram showing an example of the configuration of a network system according to this embodiment. The network system of this embodiment includes a management device 10, a router 11, a plurality of switches 13, and a plurality of terminals 15. The plurality of switches 13 constitute a network 130. Note that the network 130 may include the management device 10 and the router 11. Although only one router 11 is illustrated in FIG. 1, the network system of this embodiment may include a plurality of routers 11. The switches 13 included in the network 130 and the individual terminals 15 connected to the network 130 are also called nodes.

SNMP (Simple Network Management Protocol) is implemented in network devices such as the router 11 and the plurality of switches 13 that constitute the network 130. Further, network devices such as the router 11 and the plurality of switches 13 that make up the network 130 are implemented with protocols such as NetFlow and sFlow that can transmit statistical data of individual devices as flows. There are multiple redundant routes inside the network 130, and a particular route or network device may suddenly become highly loaded.

The management device 10 is placed in a network system that includes a network 130 configured from network devices that meet the above conditions. The management device 10 acquires various information about each of the plurality of switches 13 that make up the network 130. The management device 10 analyzes the acquired information and issues instructions for changing the configuration of the network 130 according to the load status of each switch 13.

Management device 10 is connected to router 11 . The management device 10 periodically acquires information regarding the physical wiring that connects the plurality of switches 13 that constitute the network 130, and monitors the overall load of the network 130 by managing the acquired information.

When priorities are set among the plurality of switches 13 that make up the network 130, the management device 10 manages the priorities set for each of the plurality of switches 13 that make up the network 130. For example, in order to preferentially allocate resources to the switch 13 of the same manufacturer as the server's manufacturer for packets sent from a server connected via the Internet, the priority of the switch 13 of that manufacturer is set to be high. There are times when I am.

The router 11 is connected to the management device 10. Further, the router 11 is connected to at least one of the plurality of switches 13 that constitute the network 130. Further, the router 11 is connected to the Internet or a local network (not shown). Although not shown in FIG. 1, the router 11 may be connected to a switch 13 configuring a network different from the network 130. The router 11 routes signals propagated through the plurality of switches 13 that constitute the network 130. For example, the ports of the router 11 are divided into a local VLAN (Virtual Local Area Network) and a global VLAN. The router 11 performs routing by specifying the exit of the local VLAN as the IP (Internet Protocol) address of the global VLAN.

The switch 13 is connected to one of the plurality of switches 13, the router 11, and one of the plurality of terminals 15 that constitute the network 130. The switch 13 periodically transmits statistical data such as the number of packets that have passed through the switch 13 to the management device 10 as a flow. In the example of FIG. 1, each switch 13 is physically connected to either the router 11 or the upper switch 13, and the lower switch 13 or the terminal 15. The switch 13 transfers the signal propagated via the network path set by the management device 10 to the destination switch 13, router, or terminal 15. Switch 13 has multiple ports. The switch 13 stores which port the transfer destination terminal 15 set by the management device 10 is connected to. Usually, a firewall is set up between the switch 13 connected to the router 11 and the router 11.

In FIG. 1, solid lines connecting multiple switches 13 indicate physical wiring. The network path formed by the wiring connecting the upper switch 13 and the lower switch 13 can be flexibly changed depending on the load on the network 130. Wiring that is not set as a network route is set so that it cannot communicate depending on the settings of each switch 13. Even wiring that is set to be incapable of communication can be used as a network path by changing the configuration of the network path according to the load on the network 130.

The management device 10 monitors the load on the entire network 130, and if there is a region where a high load is occurring on the network 130, the management device 10 monitors the load on the entire network 130, and if there is a region where a high load is occurring on the network 130, the management device 10 adjusts the network in the network 130 according to the load status of each switch 13 making up the network 130. Optimize routes.

The management device 10 creates candidates for changing the configuration of the network 130 using information obtained from each of the plurality of switches 13 that configure the network 130. The management device 10 sets routes between the plurality of switches 13 via the router 11. For example, the management device 10 sets connection paths between the plurality of terminals 15 and connection paths between a device (not shown) and the plurality of terminals 15. The management device 10 selects a switch 13 with a specific relationship. The management device 10 generates a group including the selected plurality of switches 13. The management device 10 dynamically changes the group according to the load status of the plurality of switches 13 making up the network 130. That is, the management device 10 dynamically generates a group including a plurality of switches 13 having a specific relationship.

For example, the management device 10 generates groups using the following method. First, the management device 10 refers to the management information base (MIB) of each network device such as the router 11 and the plurality of switches 13 and investigates adjacent network devices. The management device 10 records network devices that are adjacent to each other. If a plurality of network devices are adjacent to each other, the management device 10 records all of them. Next, the management device 10 analyzes the conversation including the source and destination of the network flow flowing through each network device, and examines the flows flowing at the same time. The management device 10 determines that network devices that relay flows with matching conversations at close times are highly related. The management device 10 generates a group made up of a plurality of network devices that are determined to be highly related.

FIG. 2 is an example of a network composed of a plurality of switches 13. The network of FIG. 2 is comprised of 13 switches 13 numbered from 1 to 12. Each of the plurality of switches 13 is connected by a solid line or a broken line. The solid line is the signal transmission path (also called a network path) at the time of FIG. For example, at the time of FIG. 2, a network path is set between SW5 and SW2, SW9, and SW10.

FIG. 3 is an example of an analysis state (analysis table 131) in which the management device 10 analyzes the MIB information of the network shown in FIG. The analysis table 131 records information that associates the two switches 13 for which network paths are set at the time point in FIG. 2. For example, based on the information in the analysis table 131, the management device 10 determines that SW2, SW5, SW9, and SW10 within the range surrounded by the dashed line in FIG. Generate a group consisting of:

FIG. 4 is another example of a network composed of a router 11 and a plurality of switches 13. The network of FIG. 4 includes thirteen switches 13, numbered 1 to 13. SW1 is connected to router 11. Each of SW8-13 is connected to each of the plurality of terminals 15 (END-A, END-B, END-C, END-D, END-E, END-F). Router 11 is connected to terminal 15 (END-Z). A network path indicated by a solid line is set between each of the plurality of switches 13. Note that in FIG. 4, network routes that have not been set are omitted. In FIG. 4, arrows indicate flows from each of END-A, END-B, END-C, END-D, END-E, and END-F toward END-Z.

FIG. 5 is an example of an analysis state (analysis table 132) in which the management device 10 analyzes the MIB information of the network shown in FIG. In the analysis table 132, information on the flow of each switch 13 at the time point in FIG. 4 is recorded. For example, the management device 10 analyzes the analysis table 132, sorts by the source and destination included in the information in the analysis table 132, and generates an analysis table 133 as shown in FIG. For example, the management device 10 groups the plurality of switches 13 configuring the network in FIG. 4 based on the analysis table 133 in FIG. 6. FIG. 7 is a table 134 that summarizes groups generated by the management device 10 based on the analysis table 133 of FIG. 6. For example, the management device 10 generates group 1 and group 2 as shown in the table 134 of FIG. Group 1 includes SW2, SW4, SW5, SW8, SW9, and SW10. Group 1 includes SW3, SW6, SW7, SW11, SW12, and SW13.

The management device 10 calculates, for each of the generated groups, a load factor (also referred to as a group load factor) that takes into account the load of the switches 13 that make up the group. For example, the management device 10 calculates the highest load factor within the group or the average value of the load factors within the group as the group load factor.

The management device 10 acquires statistical data from each of the plurality of switches 13 making up the network 130. For example, the management device 10 acquires traffic from each of the plurality of switches 13. For example, the management device 10 regularly acquires statistical data such as the number of packets passing through the plurality of switches 13 as a flow from each of the plurality of switches 13. For example, statistical data is data that is monitored based on MIB standards. The management device 10 receives from each of the plurality of switches 13 data related to interfaces such as transmission/reception traffic and traffic usage rate that are monitored based on MIB standards. Furthermore, the management device 10 may monitor the load status of the switch 13 based on index values other than the transmitted and received traffic. For example, the management device 10 may determine the load status of the switch 13 based on data regarding resources such as CPU (Central Processing Unit) usage rate, memory usage rate, disk usage rate, and data regarding processes.

The management device 10 collects statistical data of each of the plurality of switches 13 from each of the plurality of switches 13 that constitute the network 130 . For example, the management device 10 regularly collects statistical data of each of the plurality of switches 13 from each of the plurality of switches 13 that constitute the network 130 . The management device 10 monitors the load status of each of the plurality of switches 13 based on statistical data obtained from each of the plurality of switches 13 configuring the network 130 .

The management device 10 executes a process for changing a network route within the network 130 (also referred to as a network route change process) when any one of the plurality of switches 13 becomes highly loaded. For example, the management device 10 executes a network route change process when the statistical data of any one of the plurality of switches 13 included in any group exceeds a threshold value. In the following, a group including the high-load switches 13 will be referred to as a first group, and a group other than the first group will be referred to as a second group.

For example, the management device 10 calculates the group load factor using the following method disclosed in Patent Document 1 (Japanese Unexamined Patent Publication No. 2018-148382). For example, the management device 10 calculates the maximum value among the load factors of the plurality of switches 13 included in the group as the group load factor. For example, the management device 10 calculates the average value of the load factors of the plurality of switches 13 included in the group as the group load factor.

The management device 10 acquires the following two parameters regarding the usage rate of the switching capacity of the switch 13 at a certain point in time. The first parameter is the total number of frames F1 processed by the switch 13 at a certain point in time. The second parameter is the total number of frames F2 processed by the switch 13 after a period t has elapsed from a certain point.

The management device 10 calculates the average value A of the CPU usage rate in the period t using the two acquired parameters. For example, the management device 10 uses Equation 1 below to calculate the average value A of the CPU usage rate in the period t.

However, in the above equation 1, T is the number of times the CPU usage rate was measured in the period t, and C _p is the CPU usage rate at each measurement time.

For example, the management device 10 calculates the switching capacity S of the switch 13 using Equation 2 below.

When the traffic of F ₃ is generated due to a change in the network configuration, the management device 10 calculates the load factor γ of the switch 13 after resource distribution using Equation 3 below.

The management device 10 sets the average value or maximum value of the load factors γ of the plurality of switches 13 included in each group as the group load factor for each group. The management device 10 determines whether to regenerate the group of the plurality of switches 13 that constitute the network based on the group load factor for each group. For example, when the group load factor exceeds a threshold, the management device 10 regenerates a group of the plurality of switches 13 that constitute the network. For example, the management device 10 assigns any route between the switches 13 included in the first group to any switch 13 included in the second group in order to reduce the group load factor of the first group. Regenerate the group as follows.

When regenerating a group of a plurality of switches 13 constituting the network 130, the management device 10 generates group candidates (also referred to as group candidates) within a configuration changeable range. The management device 10 calculates a group load factor for each of the generated group candidates. If the group load factor of a group candidate is below the threshold, the management device 10 regenerates the group candidate as a group.

The terminal 15 is connected to any one of the plurality of switches 13 making up the network 130. The terminal 15 is a general communication terminal having a communication interface. For example, the terminal 15 receives input of information according to a user's operation, and executes information processing based on the received information. For example, the terminal 15 displays the results obtained through information processing on the screen. For example, the terminal 15 may transmit data based on information input by the user to another terminal 15 or a server (not shown) designated as a transmission destination via the network 130, or Receive data sent from a server, etc. The terminal 15 may be a general stationary computer or a mobile terminal.

[Configuration example]
FIG. 8 is a conceptual diagram showing an example in which the network system of this embodiment includes a plurality of networks 130-1 to 130-3. Note that in FIG. 8, the symbols for the switch (SW) and the terminal (END) are omitted. Each of the networks 130-1 to 130-3 includes a core switch (also referred to as a core switch) serving as a backbone, and an edge switch (also referred to as an edge switch) connected to the core switch. At least one terminal is connected to each edge switch. Each terminal has various uses, such as one that communicates only with another terminal connected to the same network 130, one that connects to an external server (not shown), and one that is open to the outside as a server. Things are mixed. Individual terminals are also called nodes.

FIG. 9 is a conceptual diagram showing an example of grouping a plurality of switches included in each of the plurality of networks 130-1 to 130-3. The multiple networks 130-1 to 3 are divided into multiple groups 135-1 to 6. Note that, for example, assume that the load of any switch included in group 135-1 exceeds a threshold value. In this case, the management device 10 selects as a candidate a route included in group 135-1 that passes through at least one switch of groups 135-2 to 135-6 different from group 135-1. . Then, the management device 10 calculates the group load factor for each of the groups 135-1 to 135-6 when a network route is configured with the selected route candidates. If the group load factor calculated for each of the groups 135-1 to 135-6 falls below the threshold, the management device 10 updates the network route with that route candidate. On the other hand, if the group load factor calculated for each of groups 135-1 to 135-6 does not fall below the threshold, another route is selected as a candidate and the network route is optimized based on the group load factor.

(motion)
Next, the operation of the management device 10 of this embodiment will be explained with reference to the drawings. FIG. 10 is a flowchart for explaining an example of the operation of the management device 10. In the description of the processing along the flowchart of FIG. 10, the management device 10 will be described as the operating entity.

In FIG. 10, the management device 10 first generates a group including a plurality of nodes having a specific relationship among the plurality of switches 13 (also referred to as nodes) that constitute the network 130 (step S111).

The management device 10 collects statistical data from each of the plurality of nodes configuring the network 130 (step S112).

The management device 10 analyzes statistical data collected from each of the plurality of nodes (step S113).

If a highly loaded node occurs (Yes in step S114), the management device 10 executes route optimization processing (step S115). On the other hand, if a node with a high load has not occurred (No in step S114), the process returns to step S112.

In the route optimization process (step S115), if there is a route change between different groups (Yes in step S116), the management device 10 updates the group according to the new network route (step S117). On the other hand, in the route optimization process (step S114), if there is no route change between different groups (No in step S116), the groups are not updated. For example, if there is no route change between different groups (No in step S116), a terminal or the like (not shown) that manages the network system including the network 130 is notified that a high-load node has occurred.

[Route optimization processing]
Next, route optimization processing by the management device 10 will be explained with reference to the drawings. FIG. 11 is a flowchart for explaining an example of route optimization processing. In the description of the processing along the flowchart of FIG. 11, the management device 10 will be described as the operating entity.

In FIG. 11, the management device 10 first sets a group including a node with a high load to a group including selection targets for nodes that are candidates for route change (also referred to as a verification target group) (step S131).

Next, the management device 10 selects unverified nodes included in the verification target group as route change candidates (step S132).

Next, the management device 10 calculates the group load factor for the network route whose route has been changed to the selected node (step S133). In step S133, the management device 10 may calculate the group load factor of only the group including the highly loaded node, or may calculate the group load factor of all groups included in the network 130. Further, in step S133, the management device 10 may calculate the group load ratio of a group including a node with a high load and a group adjacent to the group.

Here, if the calculated group load ratio does not fall below the threshold (No in step S134) and there is an unverified node in the verification target group (Yes in step S135), the process returns to step S132. On the other hand, if the calculated group load rate does not fall below the threshold (No in step S134) and there are no unverified nodes in the verification target group (No in step S135), the management device 10 determines that there is an unverified group. The process is switched depending on whether or not (step S136).

If there is an unverified group (Yes in step S136), the management device 10 sets one of the unverified groups as a verification target group (step S137). After step S137, the process returns to step S132. On the other hand, if there is no unverified group (No in step S136), the process according to the flowchart in FIG. 11 is terminated, and the process proceeds to step S116 in the flowchart in FIG.

In step S134, if the calculated group load factor is less than the threshold (Yes in step S134), the management device 10 updates the network path to include the selected node (step S138 in FIG. 12). After step S138, the process advances to step S116 in FIG.

As described above, the network system of this embodiment includes a network configured by a plurality of network devices and a management device. The management device collects statistical data of multiple network devices that make up the network. The management device generates a group made up of a plurality of network devices having a relationship among the plurality of network devices. The management device uses the statistical data to calculate the load factor of the network device and the group load factor of the group. When the load factor exceeds the threshold, the management device determines the load status of the group based on the group load factor. The management device. Update network routes based on group load status.

In one aspect of the present embodiment, when the load factor of a first network device included in the network exceeds a threshold, the management device calculates a group load factor of a first group including the first network device. The management device selects a second network device included in the network as a candidate for allocating at least one of the routes passing through the first network device, based on the load status of the first group.

In one aspect of the present embodiment, the management device calculates the group load factor of the first group with respect to network route candidates in which at least one of the routes passing through the first network device is assigned to the second network device. do. The management device updates the network route with the network route candidate when the group load factor of the first group in the network route candidate is less than the threshold value.

In one aspect of this embodiment, the management device selects a network route candidate in which at least one of the routes passing through the first network device is assigned to the second network device. The management device calculates a group load factor of a first group including the first network device and a group load factor of a second group including the second network device with respect to the network route candidate. The management device updates the network route with the network route candidate when the group load factors of the first group and the second group in the network route candidate are below the threshold. The management device regenerates a group made up of a plurality of related network devices.

In one aspect of the present embodiment, the management device generates a group that includes a plurality of network devices that relay flows with matching conversations to each other at close times.

In one aspect of this embodiment, the management device calculates the load factor of the network device using the transmission and reception traffic included in the management information base transmitted from the network device. In one aspect of the present embodiment, at least one of an average value and a maximum value of load factors of a plurality of network devices included in a group is calculated as a group load factor.

According to this embodiment, in a large-scale network environment where multiple redundant routes exist depending on performance and failure resistance, the load It can automatically determine the optimal route depending on the situation. For example, according to the present embodiment, even if there are many redundant routes in a large-scale network environment and there are more options than can be calculated for all route candidates, the The optimal route can be determined in a short time.

Furthermore, according to the present embodiment, in a network that includes various network devices with different performances, it is possible to dynamically change the network route and level the network load regardless of the performance difference between the network devices. Therefore, the method of this embodiment can be applied to technologies that are standardly implemented in general switches such as SNMP, NetFlow, and sFlow. The method of this embodiment is not limited to the type of switch and can be applied to various environments.

Here, the difference between the method of this embodiment and the method of related technology will be explained by giving an example. FIG. 12 is a conceptual diagram showing an example of the load status of a network composed of 12 nodes (SW1 to SW12). The network in FIG. 12 shows a high load state where the load on SW2 is 70% and the load on SW4 is 90%. For example, the threshold value for determining a high load is 70%.

FIG. 13 is an example in which the network route in FIG. 12 is changed based on the load factor of each node using a related technique. In related techniques, routes are changed based on the load factor of individual nodes. In the example of FIG. 13, in order to reduce the load on SW4, the node to be passed through on the route between SW1 and SW8 is changed from SW4 to SW5 based on the load factor of each node. As a result, although the load on SW4 decreased from 90% to 65%, the load on SW2 above SW4 and SW5 still remained at 70%. That is, when optimizing routes between nodes based on the load factor of each node, it is possible to reduce the load on a node with a high load, but the load on nodes above it is not reduced.

FIG. 14 is an example in which the network route in FIG. 12 is changed based on the group load factor for each group using the method of this embodiment. In the method of this embodiment, the route is changed based on the group load factor for each group. In the example of FIG. 14, in order to reduce the load on SW4, the route connecting SW1 and SW8 is changed from the route via SW2 and SW4 to the route through SW3 and SW5 based on the group load factor for each group. change. As a result, the load on SW4 decreased from 90% to 65%, and the load on SW2 above SW4 and SW5 decreased to 45%. That is, by changing the route based on the group load factor for each group, it is possible to reduce the load on a node with a high load, and it is also possible to reduce the load on nodes above it.

For example, in the case of a small-scale network, it is possible to deal with constantly occurring loads by calculating the amount of communication on all routes included in the network and simulating the optimal route. However, in the case of a large-scale network, it takes time to estimate the traffic of all routes included in the network and simulate the optimal route, and the load situation may change during the calculation. Therefore, the method of simulating the optimal route by calculating the traffic volume of all routes included in the network is useful not only in environments where the load is constantly high, but also in environments where the load suddenly occurs. is difficult to apply.

Since the method of this embodiment can simulate an optimal route based on the load factor for each group, calculation time is shortened and it can be applied to environments where loads suddenly occur.

(Second embodiment)
Next, a management device according to a second embodiment will be described with reference to the drawings. The management device of this embodiment has a simplified configuration of the management device 10 of the first embodiment. The management device of this embodiment optimizes network routes in a network configured by a plurality of network devices.

FIG. 15 is a block diagram showing an example of the configuration of the management device 20 of this embodiment. The management device 20 includes a collection section 21 , a generation section 22 , a calculation section 23 , a determination section 24 , and an update section 25 .

The collection unit 21 collects statistical data of a plurality of network devices that constitute a network. The generation unit 22 generates a group made up of a plurality of related network devices among the plurality of network devices. The calculation unit 23 uses the statistical data to calculate the load factor of the network device and the group load factor of the group. If the load factor exceeds the threshold, the determining unit 24 determines the load status of the group based on the group load factor. The updating unit 25 updates the network route based on the load status of the group.

For example, when the load factor of the first network device included in the network exceeds the threshold, the calculation unit 23 calculates the group load factor of the first group including the first network device. For example, the updating unit 25 selects the second network device included in the network as a candidate for allocating at least one of the routes passing through the first network device, based on the load status of the first group.

For example, the calculation unit 23 calculates the group load factor of the first group with respect to network route candidates in which at least one of the routes passing through the first network device is assigned to the second network device. For example, when the group load factor of the first group in the network route candidate is less than the threshold value, the updating unit 25 updates the network route with the network route candidate.

For example, the calculation unit 23 selects a network route candidate in which at least one of the routes passing through the first network device is assigned to the second network device. For example, the calculation unit 23 calculates the group load factor of the first group including the first network device and the group load factor of the second group including the second network device with respect to the selected network route candidate. For example, if the group load factors of the first group and the second group in the network route candidate are below the threshold, the updating unit 25 updates the network route with the network route candidate. For example, the generation unit 22 regenerates a group made up of a plurality of related network devices.

For example, the generation unit 22 generates a group including a plurality of network devices that relay flows with matching conversations at close times among the plurality of network devices.

For example, the calculation unit 23 calculates the load factor of the network device using the transmission/reception traffic included in the management information base transmitted from the network device. For example, the calculation unit 23 calculates at least one of the average value and the maximum value of the load factors of the plurality of network devices included in the group as the group load factor.

According to this embodiment, the network route is updated based on the statistical data of the network devices that make up the network, so the network route can be optimized without relying on the resources of the network devices.

(hardware)
Here, the hardware configuration for realizing devices such as a management device, a network device, and a terminal according to each embodiment of the present invention will be described using the information processing device 90 in FIG. 16 as an example. Note that the information processing device 90 in FIG. 16 is a configuration example for executing the processing of the device of each embodiment, and does not limit the scope of the present invention.

As shown in FIG. 16, the information processing device 90 includes a processor 91, a main storage device 92, an auxiliary storage device 93, an input/output interface 95, and a communication interface 96. In FIG. 16, the interface is abbreviated as I/F (Interface). Processor 91, main storage 92, auxiliary storage 93, input/output interface 95, and communication interface 96 are connected to each other via bus 98 so as to be able to communicate data. Furthermore, the processor 91, main storage device 92, auxiliary storage device 93, and input/output interface 95 are connected to a network such as the Internet or an intranet via a communication interface 96.

The processor 91 loads a program stored in the auxiliary storage device 93 or the like into the main storage device 92, and executes the loaded program. In this embodiment, a configuration using a software program installed in the information processing device 90 may be used. The processor 91 executes processing by the device according to this embodiment.

The main storage device 92 has an area where programs are expanded. The main storage device 92 may be, for example, a volatile memory such as DRAM (Dynamic Random Access Memory). Furthermore, a non-volatile memory such as MRAM (Magnetoresistive Random Access Memory) may be configured or added as the main storage device 92.

Auxiliary storage device 93 stores various data. The auxiliary storage device 93 is configured by a local disk such as a hard disk or flash memory. Note that it is also possible to adopt a configuration in which various data are stored in the main storage device 92 and omit the auxiliary storage device 93.

The input/output interface 95 is an interface for connecting the information processing device 90 and peripheral devices. The communication interface 96 is an interface for connecting to an external system or device via a network such as the Internet or an intranet based on standards and specifications. The input/output interface 95 and the communication interface 96 may be shared as an interface for connecting to external devices.

The information processing device 90 may be configured to connect input devices such as a keyboard, a mouse, and a touch panel as necessary. These input devices are used to input information and settings. Note that when a touch panel is used as an input device, the display screen of the display device may be configured to also serve as an interface for the input device. Data communication between the processor 91 and the input device may be mediated by the input/output interface 95.

Further, the information processing device 90 may be equipped with a display device for displaying information. When equipped with a display device, the information processing device 90 is preferably equipped with a display control device (not shown) for controlling the display of the display device. The display device may be connected to the information processing device 90 via the input/output interface 95.

The above is an example of the hardware configuration for realizing the apparatus according to each embodiment of the present invention. Note that the hardware configuration in FIG. 16 is an example of the hardware configuration for executing the arithmetic processing of the device according to each embodiment, and does not limit the scope of the present invention. Further, a program that causes a computer to execute processing related to the apparatus according to each embodiment is also included within the scope of the present invention. Furthermore, a program recording medium on which a program according to each embodiment is recorded is also included within the scope of the present invention. The recording medium can be realized by, for example, an optical recording medium such as a CD (Compact Disc) or a DVD (Digital Versatile Disc). Further, the recording medium may be realized by a semiconductor recording medium such as a USB (Universal Serial Bus) memory or an SD (Secure Digital) card, a magnetic recording medium such as a flexible disk, or other recording media. When a program executed by a processor is recorded on a recording medium, the recording medium corresponds to a program recording medium.

The components of the device of each embodiment can be arbitrarily combined. Furthermore, the components of the apparatus of each embodiment may be realized by software or by circuits.

Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above embodiments. The configuration and details of the present invention can be modified in various ways that can be understood by those skilled in the art within the scope of the present invention.

10, 20 management device 11 router 13 switch 15 terminal 21 collection unit 22 generation unit 23 calculation unit 24 determination unit 25 update unit 130 network

Claims (10)

a collection means for collecting statistical data of a plurality of network devices constituting the network;
generating means for generating a group formed by a plurality of network devices having a relationship among the plurality of network devices;
calculation means for calculating a load factor of the network device and a group load factor of the group using the statistical data;
determining means for determining the load status of the group based on the group load rate when the load rate exceeds a threshold;
A management device comprising: updating means for updating a network route based on the load status of the group. The calculation means is
If the load factor of a first network device included in the network exceeds the threshold, calculating the group load factor of a first group including the first network device;
The updating means includes:
2. A second network device included in the network is selected as a candidate for allocating at least one of the routes passing through the first network device based on the load status of the first group. management device. The calculation means is
calculating the group load factor of the first group with respect to network route candidates in which at least one of the routes passing through the first network device is assigned to the second network device;
The updating means includes:
The management device according to claim 2, wherein when the group load factor of the first group in the network route candidate is less than the threshold value, the management device updates the network route with the network route candidate. The calculation means is
With respect to a network route candidate in which at least one of the routes passing through the first network device is assigned to the second network device, the group load factor of the first group including the first network device; and the group load factor of a second group including a second network device;
The updating means includes:
If the group load factors of the first group and the second group in the network route candidate are below the threshold, updating the network route with the network route candidate;
The generating means is
The management device according to claim 2, wherein the management device regenerates the group made up of a plurality of related network devices. The generating means is
The management device according to any one of claims 1 to 4, which generates the group including a plurality of network devices that relay flows having matching conversations at close times among the plurality of network devices. The calculation means is
The management device according to any one of claims 1 to 5, wherein the load factor of the network device is calculated using transmission and reception traffic included in a management information base transmitted from the network device. The calculation means is
The management device according to any one of claims 1 to 6, wherein at least one of an average value and a maximum value of the load factors of the plurality of network devices included in the group is calculated as the group load factor. A management device according to any one of claims 1 to 7,
A network system comprising: the network configured by the plurality of network devices. The computer is
Collect statistical data of multiple network devices that make up the network,
generating a group constituted by a plurality of related network devices among the plurality of network devices;
calculating a load factor of the network device and a group load factor of the group using the statistical data;
If the load factor exceeds a threshold, determining the load status of the group based on the group load factor,
A management method that updates a network route based on the load status of the group. A process of collecting statistical data of multiple network devices that make up the network;
A process of generating a group constituted by a plurality of related network devices among the plurality of network devices;
a process of calculating a load factor of the network device and a group load factor of the group using the statistical data;
If the load factor exceeds a threshold, determining the load status of the group based on the group load factor;
A program that causes a computer to execute a process of updating a network route based on the load status of the group.

Images