KR100744934B1 - Analysis Method for Routing Status based on Passive Measurement - Google Patents

Abstract

The present invention relates to a routing state analysis method using an active measurement method, comprising the steps of inputting configuration information, and periodically starting measurement of a target node according to a measurement period according to the configuration information, from the target node of the measurement performed. Collecting measurement data, storing the data, and analyzing the measurement data.

According to the present invention configured as described above, the bottleneck can be extracted by IP unit, and thus, there is an advantage of providing more raw analysis information, and the bottleneck can be extracted in the long term, thereby immediately improving the performance of the network. It can be used to establish a more efficient and economical BGP peering policy, and can freely adjust the measurement cycle, thus making the network more efficient.

Description

Analysis Method for Routing Status based on Passive Measurement

1 is a view schematically showing a connection state of an autonomous system, a router and a server according to an embodiment of the present invention.

2 is a view showing in detail the structure of a table configured in the analysis database shown in FIG.

3 is a view showing in detail the structure of a table configured in the Internet routing registry table database shown in FIG.

4 is a flowchart illustrating a routing state analysis method using an active measurement method according to an embodiment of the present invention in detail.

5 is a diagram showing an example of a result of performing a trace route command of the periodic measurement step shown in FIG. 4;

FIG. 6 is a diagram showing an example of a result of performing a ping command of the periodic measurement step shown in FIG. 4; FIG.

7 is a diagram illustrating in detail the analysis and bottleneck information extraction step shown in FIG.

<Description of Symbols for Main Parts of Drawings>

100, 200, 300, 410: IP network 400: autonomous system                 

310: IRR Server 320: IRR Database

420: routing status analysis server 430: analysis database

The present invention relates to routing state analysis, and more particularly, to an apparatus and method for analyzing a routing state using an active measurement method capable of periodically measuring a router using an active measurement method and analyzing the measured result.

In general, there are two methods of measuring data in a network: active measurement and passive measurement.

The active measurement method measures data by generating test packets in a network. Such data measurement tools include ping, traceroute, bing, pathchar, pchar, and mtr. Passive measurement methods can measure very precise data without generating test packets on the network. Such data measurement tools include OcxMON, NeTraMet, Ethereal, and tcpdump.

The ping measures the bidirectional delay and loss rate of the node, and the traceroute measures the list of hops to the node and the bidirectional delay to each hop. In addition, bing provides the function of the ping and bandwidth measurement function, and pathchar and pchar provide the bandwidth measurement function between each hop in the function of traceroute. Mtr performs the function of ping and traceroute at the same time.

The tools of the active measurement method as described above measure each individual node or a set of nodes and express or graph the measurement results in text.

However, such a conventional active measurement method is not suitable as a tool for analyzing the routing status and routing stability because the measurement becomes very cumbersome and the data increases when the set of nodes to be measured increases. In particular, in the case of measuring other ISPs from arbitrary information service providers (ISPs), the hops are increased and the amount of data is huge, which is not suitable.

Moreover, the conventional active measurement method can take full advantage of the information because the measured data simply provides the hop's IP address or host name, so even if the bottleneck is extracted, the information about the ISP or AS (autonomous system) to which they belong is not known. There is no problem.

Accordingly, an object of the present invention is to provide a routing state analysis method using an active measurement method that can measure and store a target node to be measured periodically in a database.

In addition, an object of the present invention is to provide a routing state analysis method using an active measurement method that can analyze the autonomous system by matching the measurement data and the autonomous system information.

Furthermore, an object of the present invention is to provide a routing state analysis method using an active measurement method that can analyze routing state and routing state bottlenecks, routing stability, and routing stability bottlenecks from measurement data.

According to a feature of the present invention proposed to achieve the above object, a routing state analysis method using an active measurement method comprising the steps of inputting the target node to be measured, the type of measurement and the configuration information of the measurement period; Starting measurement of the target node periodically according to a measurement period according to the input configuration information in the inputting of the configuration information; Collecting measurement data from the target node of the measurement performed in initiating the measurement; Storing the collected measurement data in a database; And analyzing the collected measurement data.

In a preferred embodiment of this aspect, the step of inputting the configuration information is characterized in that it comprises the step of receiving a measurement type to perform ping and traceroute for each target node.

In a preferred embodiment of this aspect, the collecting of the measurement data comprises retrieving autonomous system information corresponding to the measurement data collected from the collecting of the measurement data.

In a preferred embodiment of this aspect, analyzing the measurement data comprises analyzing the routing status.

In a preferred embodiment of this aspect, analyzing the routing state includes analyzing node packet loss rate, node bidirectional delay, autonomous system packet loss rate, and autonomous system bidirectional delay.

In a preferred embodiment of this aspect, analyzing the routing state comprises analyzing an autonomous system routing state bottleneck.

In a preferred embodiment of this aspect, analyzing the measurement data comprises analyzing routing stability.

In a preferred embodiment of this aspect, the analyzing the routing stability comprises analyzing a node hop variation coefficient, a node bidirectional delay variation coefficient, an autonomous system hop variation coefficient, and an autonomous system bidirectional delay variation coefficient. It features.

In a preferred embodiment of this aspect, the analyzing of the routing stability comprises analyzing the autonomous system routing stability.

Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. 1 to 7.

1 is a diagram schematically illustrating a connection state between an autonomous system, a router, and a server according to an exemplary embodiment of the present invention.

In the drawings, reference numerals 100, 200, 300, and 410 denote IP networks. 400 represents an autonomous system, 310 represents an IRR server, 320 represents an IRR database, 420 represents a routing state analysis server, and 430 represents an analysis database.

The IP networks 100, 200, 300, and 410 belong to respective autonomous systems.

The routing state analysis server 420 providing routing state analysis information using an active measurement method according to an embodiment of the present invention is connected to the router 440 of the IP network 410. The IP network 410 is configured in the autonomous system 400. The routing state analysis server 420 is configured to include an analysis database 430 for storing the analyzed routing information.

On the other hand, an Internet Routing Registry (IRR) server 310 is connected to the router 330 of the IP network 300. The IRR server 310 provides information on a registered IP address block, and provides data stored in the IRR database 320 in response to a request from a user.

2 is a view showing in detail the structure of a table configured in the analysis database shown in FIG. 2A shows an IP information table, 2b shows a ping table, and 2c shows a traceroute table.

The IP information table indicates host and autonomous system information for the IP address included in the measurement data. The IP information table includes a date 581, a time 582, an IP address 583, a host name 584, and an AS number 585. ), And an AS name 586 field. In this case, the date field indicates a date of active measurement and the time field indicates a time when active measurement is performed. In addition, the IP address field represents all IP addresses and host name fields appearing when measuring ping or trace route, respectively, and represents host names of IP addresses stored in the IP address field. The AS number field indicates an AS number to which the IP address stored in the IP address field belongs, and the AS name field indicates an AS name to which the IP address stored in the IP address field belongs. The IP address field stores all the IP addresses measured in the ping or traceroute.

The ping table consists of a date 587, a time 588, an IP address 589, a loss rate 590, and an average delay 591 field. Here, the date field indicates the date when the ping is performed and the time field indicates the time when the ping is performed. The IP address field indicates the IP address of the target node of the ping. In addition, the loss rate field indicates the packet loss rate to the target node, and the average delay field indicates the bidirectional delay to the target node.

Finally, the traceroute table includes date (592), time (593), destination address (594), destination hop (595), first IP (596), first delay (597), Nth IP (598), and N-th delay 599 field. Here, the date field indicates the date of the trace route and the time field indicates the time of the trace route. The destination address field indicates the IP address of the target node and the destination hop field indicates the number of hops to the destination node. In addition, the first through Nth IP fields indicate IPs passed through the target node, and the first through Nth delay fields indicate bidirectional delays through the pass IP, respectively.

3 is a view showing in detail the structure of the autonomous system table configured in the analysis database shown in FIG. As shown in the figure, the autonomous system table includes a route 601, an AS number 602, an AS name 603, a mask 604, a start IP address 605, and a last IP address 606 fields. It is composed.

The autonomous system table consists of information about the Internet routing registry collected from the IRR server 310 by the routing status analysis server 420.

The route field here represents a block of IP addresses. The AS number field indicates the AS number of the IP address block included in the route field, and the AS name field indicates the AS name of the IP address block included in the route field. In addition, the mask field indicates a mask of the IP address block included in the route field. Finally, the start IP address field indicates the start IP address of the IP address block included in the route field, and the last IP address field indicates the last IP address of the IP address block included in the route field.

4 is a flowchart showing in detail a routing state analysis method using an active measurement method according to an embodiment of the present invention.

First, the operator inputs the node to be measured through the routing state analysis server 420, configuration information to be measured, such as time (or period), and a measurement tool (ping or trace route) (S110). At this time, the operator inputs configuration information reflecting the topology of the network. The configuration information about the time indicates the time at which the measurement is started, and can be set separately for each node.

Then, the routing state analysis server 420 periodically starts the measurement for the target node based on the input measurement configuration information (S120). In this case, measurement of hop related information for each target node and information on bidirectional delay and packet loss rate for each target node are started.

Subsequently, the routing state analysis server 420 collects measurement data from the target node according to the measurement disclosed as described above (S130). The collected measurement data is stored in a memory or a file. The routing state analysis server 420 extracts hop related information for each target node based on the collected measurement data. In addition, the routing status analysis server 420 extracts the bidirectional delay and packet loss rate for each target node based on the collected measurement data.

Subsequently, the routing state analysis server 420 searches for the AS information by referring to the autonomous system table based on the IP address of the measurement data (S140). The routing state analysis server 420 collects the measurement data as described above (S130), and when the AS information is retrieved (S140), and stores the result in the analysis database (430) (S150).

Finally, the routing status analysis server 420 analyzes the routing status and the routing stability based on the data stored in the analysis database 430. At this time, the routing state analysis server 420 extracts the bottleneck information (S160). The analysis information and the bottleneck information is stored in the analysis database 430. The routing state analysis is performed for each measurement data, and the routing stability is performed using the measurement data for one day or one week. Bottleneck information is then made for measurements that affect routing status and routing stability.

FIG. 5 is a diagram illustrating an example of a result of performing a trace route command of the periodic measurement step illustrated in FIG. 4.

When the routing state analysis server 420 performs a traceroute command on the target node (eg, www.yonsei.ac.kr) based on the measurement configuration information, the execution result is returned as shown in the figure.

In the drawing, reference numeral 510 denotes a hop number, 520 denotes a hop address, 530 denotes a host name of a hop address, 540 denotes a hop-by-hop bidirectional delay, and 550 denotes the number of hops.                     

FIG. 6 is a diagram illustrating an example of a result of performing a ping command of the periodic measurement step illustrated in FIG. 4.

In the figure, reference numeral 560 denotes a packet loss rate, and 570 denotes a bidirectional average delay.

7 is a diagram illustrating in detail the analysis and bottleneck information extraction step shown in FIG.

First, the packet loss rate and the bidirectional delay for the routing state of each node to be analyzed are set, and the hop variation coefficient within the predetermined period and the bidirectional delay variation coefficient within the period for the routing stability of each node are set (S161).

Subsequently, the routing state analysis server 420 extracts routing state analysis and routing state bottleneck information (S162). Here, the packet loss rate and the bidirectional delay of the node corresponding to the date field 592 and the time field 593 of FIG. 2C are calculated. The autonomous system packet loss rate and the bidirectional delay corresponding to the date field and the time field are calculated. When the bidirectional delay is calculated, the routing state analysis server 420 calculates bottleneck information corresponding to the date field and the time field.

The packet loss rate of the node is equal to the value of the loss rate field 590 of FIG. 2B. The node bidirectional delay is equal to the value of the average delay field 591 of FIG. 2B.

The autonomous system packet loss rate is equal to the average of the values of the loss rate field 590 of the IP address fields 589 of FIG. 2B included in the value of the AS number field 585 of FIG. 2A. The autonomous system bidirectional delay is equal to the average of the average delay field 591 of the IP address fields 589 of FIG. 2B included in the value of the AS number field 585 of FIG. 2A.

Meanwhile, the routing state bottleneck information of the autonomous system is the first delay (597) to the Nth time beyond the bi-hop hop threshold for the destination address fields 594 of FIG. 2C included in the value of the AS number field 585 of FIG. 2A. The first IP 596 to the Nth IP 598 corresponding to the delay 599 and a pair of AS information. Here, the hop-by-hop reference value is an average of values of the first delay 597 to the Nth delay 599 shown in FIG. 2C.

Finally, the routing status analysis server 420 analyzes routing stability and extracts routing stability bottleneck information (S163). That is, the routing state analysis server 420 calculates a node hop variation coefficient, a node bidirectional delay variation coefficient, an autonomous system hop variation coefficient, and an autonomous system bidirectional delay variation coefficient, and extracts bottleneck information using the variation coefficient. .

The node hop variation coefficient is the standard deviation of the value of the destination hop field 595 with respect to the destination address fields 594 belonging to the period included in the combination of the date field 592 and the time field 593 shown in FIG. 2C. Same as

The node bidirectional delay variation coefficient is a standard deviation of the value of the average delay field 591 with respect to the IP address fields 589 belonging to the period included in the combination of the date field 587 and the time field 588 of FIG. 2B. Same as

In addition, the hop variation coefficient for each autonomous system is equal to the average of the node hop variation coefficients for the destination address fields 594 of FIG. 2C included in the AS number field 585 of FIG. 2A.

The bidirectional delay variation coefficient for each autonomous system is equal to the average of the node bidirectional delay variation coefficients for the IP address field 589 of FIG. 2B having the same AS number field 585 of FIG. 2A.

Finally, the autonomous system routing stability bottleneck information includes the autonomous system hop state coefficient and the autonomous system routing state bottleneck IP and autonomous system information pairs of the corresponding autonomous system. It is like a set of pairs that appear a lot. The reference value is equal to the total mean of the autonomous system hop variation coefficient or the total mean of the autonomous system bidirectional delay variation coefficient.

According to the present invention, when analyzing a routing table of a conventional router, a considerable data collection time and analysis time are required, and a load is placed on a router in operation, and router information and BGP routing information belonging to another ISP are formed in an autonomous system unit. Solving the problem that it is difficult to solve the problem of the autonomous system unit when extracting the bottleneck, there is an advantage that can provide a more fundamental analysis information because the bottleneck can be extracted by the IP unit.

In addition, since the present invention uses a database, it is possible to extract bottlenecks in the long term, thereby enabling immediate response to improve network performance, and be useful for establishing a more efficient and economical BGP peering policy.

Furthermore, according to the present invention, since the measurement period can be freely adjusted, the operator can be notified immediately of the problems occurring in the network, thereby making it possible to efficiently operate the network.

Claims (9)

In the active measurement method of the IP network, Inputting configuration information of a target node to be measured, a measurement type, and a measurement period; Starting measurement of the target node periodically according to a measurement period according to the input configuration information in the inputting of the configuration information; Collecting measurement data from the target node of the measurement performed in initiating the measurement; Storing the collected measurement data in a database; And And analyzing the collected measurement data. The routing state analysis method using an active measurement method. The method of claim 1, The step of inputting the configuration information includes the step of receiving a measurement type to perform a ping and trace route for each target node routing status analysis method using an active measurement method. The method of claim 1, The collecting of the measurement data may include retrieving autonomous system information corresponding to the collected measurement data from the collecting of the measurement data. The method of claim 1, And analyzing the measurement data comprises analyzing the routing status. The method of claim 4, wherein The analyzing of the routing state includes analyzing a node packet loss rate, a node bidirectional delay, an autonomous system packet loss rate, and an autonomous system bidirectional delay. The method of claim 4, wherein And analyzing the routing state comprises analyzing an autonomous system routing state bottleneck. The method of claim 1, The analyzing of the measurement data comprises the step of analyzing the routing stability, routing state analysis method using an active measurement method. The method of claim 7, wherein The analyzing of the routing stability may include analyzing a node hop variation coefficient, a node bidirectional delay variation coefficient, an autonomous system hop variation coefficient, and an autonomous system bidirectional delay variation coefficient. Condition analysis method. The method of claim 7, wherein The analyzing of the routing stability comprises the step of analyzing the routing stability of the autonomous system routing status analysis method using an active measurement method.

Images