JP2007336021A - Network topology design method and network topology design apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a technology capable of intuitionally, simply obtaining an optimum network topology with excellent scalability by taking into account a plurality of evaluation scales in different units at the same time. <P>SOLUTION: In the network topology design method for selecting a set of core nodes and links in use while taking into account the evaluation scales at the same time in a state that a set of edge nodes each accommodating a user, a set of candidates of the core nodes including only a relay function, and a set of candidates of the links tying them are obtained, the method calculates ranking of each topology candidate among all the candidates by the evaluation scales considered at the same time, calculates an average of the rankings obtained by each evaluation scale, and determines superiority and inferiority among the candidates on the basis of the average ranking. The method uses the four evaluation scales taking into account at the same time: the total number of nodes, the total length of links, the sum of products each between a length and a traffic amount of each path over all paths, and a traffic amount of a maximum load link. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a rope topology design method and a rope topology design apparatus, and in particular, in a network topology design problem, an optimum network topology is determined by simultaneously considering a plurality of evaluation scales having different units by evaluating with an average rank. It relates to the method of seeking.

How to configure the physical network topology for telecommunications carriers and some Internet Service Providers (ISPs) that manage and operate physical network resources such as optical fibers and routers The physical topology design is an important issue.
The physical topology is a factor that determines node and link design costs, network management and maintenance costs, reliability against failures, and the like.
In addition, when a logical path such as LSP (Label Switched Path) or wavelength path cannot be set up and the route through which the packet is transferred cannot be set explicitly, the physical topology directly affects the link load or the packet route. This is a factor that determines user quality such as transmission delay quality and throughput.
On the other hand, if a logical path is installed on a physical network, such as an MPLS (Multiprotocol Label Switching) network or a wavelength path network, and the path of the path between grounds can be explicitly designed, the user of the physical topology can be determined by logical topology design. It is possible to eliminate the influence on the quality to some extent.
In this case, in addition to the physical topology design, it is necessary to appropriately design the logical topology. Thus, the network topology design problem can be broadly divided into a physical topology design problem and a logical topology design problem. The outline of each is summarized below.

[Physical topology design problem]
As generalized components of the network, consider three types: edge nodes, core nodes, and links. The edge node is an incoming / outgoing node for AC traffic, and accommodates an end host directly or via a lower network.
On the other hand, the core node is a node having only a relay function and does not accommodate users. A link plays a role of connecting edge nodes and core nodes. Assume that an AC traffic matrix between arbitrary edge nodes is given.
Telecom operators already have civil infrastructure such as pipelines and stations that are arranged in consideration of geographical characteristics, and it is not realistic to redesign them. Therefore, it is assumed that the position of the edge node and the position where the core node or link can be installed are given.
It is a practical problem to select these existing pipelines and stations and study where to place equipment such as optical fibers and routers in accordance with changes in the requirements of the network. Therefore, here, the physical topology design is defined as “an act of selecting a candidate position to be actually installed for a given core node and link installation candidate position”. However, edge nodes are installed at all edge node installation positions.

[Logical topology design problems]
By using an MPLS router or OXC (optical cross-connect system), a virtual path (LSP or wavelength path) can be installed on a physical network, and a route through which traffic between each ground flows can be explicitly designed. .
The physical topology is difficult to change once constructed, and the design cycle ranges from one to several years, but the logical topology can be easily changed, and from a few minutes depending on changes in AC traffic Flexible operations such as dynamically changing paths at intervals of several hours are possible.
The logical topology design is what kind of route a virtual path is set up for a given physical topology and an AC traffic matrix between edge nodes. Since the design is based on physical network resources, it is usually necessary to design in consideration of various constraints such as link capacity.

When the logical path can be installed on the physical topology, the logical topology can be designed at the same time when the physical topology is designed. Since the design period of the logical topology is much shorter than that of the physical topology, only the logical topology is designed after the operation is started.
For the user, it is desirable to avoid congestion at the transit node and to reduce the transmission delay so that the total route distance is short.
Furthermore, from the viewpoint of the utilization efficiency and reliability of transmission resources, it is desirable to avoid the concentration of traffic on a specific link and to distribute it as evenly as possible to all links.
However, if the number of installed nodes and the number of links are reduced in order to reduce the equipment cost and management / maintenance cost, the flexibility of path setting decreases, making it difficult to properly distribute traffic and reduce the path length.
As described above, when designing a network topology, it is necessary to simultaneously consider a plurality of evaluation measures having different units such as a traffic distribution degree, an average path length, and a network cost, which are mutually contradictory.

As a method for simultaneously considering a plurality of evaluation measures, the Pareto optimal concept used in the field of microeconomics is representative, and an example of application to logical topology design is also seen.
There are M evaluation scales V _{1 to} V _M, where m is an arbitrary value, and m-th evaluation scale value of a candidate x is V _{m, x} , V _{m, x} ≦ V for all m _{When m and y} and at least one _{m that satisfies} V _{m, x} <V _{m, y} exists, the candidate x is said to dominate the candidate y in a Pareto manner. However, it is assumed that the smaller the evaluation scale, the better. Then, it is assumed that all candidates for which there are no other controlled candidates are Pareto optimal.
As another method for simultaneously considering a plurality of evaluation scales, DEA (Data Envelopment Analysis) is known. DEA is an evaluation method that has been developed mainly as a method for evaluating investee entities, and it is possible to obtain efficiency from the attitude of evaluating each business entity's strength (evaluation scale). Is also being applied to network topology design. Refer to the following Non-Patent Document 1 for the above technique.

As prior art documents related to the invention of the present application, there are the following.
G. Huiban, and GRMateus, "A multiobjective approach of the virtual topology design and routing problem in WDM networks," ICT International Conference on Telecommunications, 2005.

However, when network design is performed by Pareto optimization, there are a large number of candidates judged to be optimal, and it is difficult to effectively narrow down candidates.
In addition, DEA has an intuitive concept of efficiency that makes it difficult to determine the optimality of the obtained solution. Furthermore, the efficiency calculation needs to solve a linear programming problem, and there is a problem in scalability.
The present invention has been made to solve the problems of the prior art, and an object of the present invention is to be intuitive, simple, and highly scalable in a rope topology design method and a rope topology design apparatus. In addition, it is an object of the present invention to provide a technique capable of obtaining an optimum network topology while simultaneously considering a plurality of evaluation scales having different units.
The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.
In order to achieve the above object, the present invention provides a set of edge nodes that accommodate users, a set of candidate core nodes having only a relay function, and a set of link candidates that connect them. A network topology design method for selecting a core node and a set of links to be used while simultaneously considering a plurality of evaluation measures, and for each topology candidate, rank among all candidates for each evaluation measure considered simultaneously. The average value of the ranking obtained for each evaluation scale is calculated, and the superiority or inferiority between candidates is determined by the average ranking.
Further, in the present invention, four evaluation scales that are simultaneously considered include the total number of nodes, the total link length, the product of the length of each path and the traffic amount for all paths, and the traffic amount of the maximum load link. .
In addition, the present invention provides a plurality of evaluation measures simultaneously when a set of edge nodes accommodating users, a set of candidate core nodes having only a relay function, and a set of link candidates connecting them are given. A network topology design device that selects a set of core nodes and links to be used while considering the topology information and demand matrix input device to which the positions of edge nodes, core nodes can be installed, and links can be installed , An initial topology generation device that generates an initial topology based on the edge node position, the core node installation position, and the link installation position input by the topology information / demand matrix input device, and the initial topology generation device Multiple candidate topologies by adding several links and core nodes to the initial topology For each candidate evaluation scale to be calculated at the same time with respect to a plurality of candidate topologies generated by the topology candidate generating apparatus and the topology candidate generating apparatus. An average rank calculation device that calculates an average value of ranks and determines superiority or inferiority among candidates based on the average rank, and an optimum topology output device that outputs a set of optimum topologies based on the evaluation result of the average rank calculation device.

The effects obtained by the representative ones of the inventions disclosed in the present application will be briefly described as follows.
According to the present invention, it is intuitive, simple and excellent in scalability, and an optimum network topology can be obtained in consideration of a plurality of evaluation measures having different units at the same time.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
In all the drawings for explaining the embodiments, parts having the same functions are given the same reference numerals, and repeated explanation thereof is omitted.
FIG. 1 is a block diagram showing a schematic configuration of a network topology design apparatus using an average rank according to an embodiment of the present invention. In FIG. 1, 101 is a topology information / demand matrix input device, 102 is an initial topology generation device, 103 is a topology candidate generation device, 104 is an average rank calculation device, and 105 is an optimum topology output device.
The topology information / demand matrix input device 101 inputs the position of the edge node, the installable position of the core node, and the installable position of the link.
The initial topology generation apparatus 102 generates an initial topology based on the position of the edge node, the installable position of the core node, and the installable position of the link input by the topology information / demand matrix input apparatus 101.
The topology candidate generation device 103 adds a number of links and core nodes to the initial topology generated by the initial topology generation device 102 to generate a plurality of candidate topologies. The average rank calculation device 104 evaluates superiority or inferiority among a plurality of candidate topologies generated by the topology candidate generation device 103.
The optimum topology output device 105 outputs a set of optimum topologies based on the evaluation result from the average rank calculation device 104.

Next, details of the network topology design method using the average rank of the embodiment of the present invention will be described. Here, as an example, the study will proceed with physical topology design as an example, but it is assumed that the logical path is installed at the upper level, and the logical path is also installed at the time of physical topology design (however, restrictions such as link capacity) Condition is not considered).
However, the network topology design method of the present invention focuses on how to consider a plurality of evaluation measures at the same time, and the constraint conditions are highly versatile. It can also be applied to topology design.
[Evaluation scale]
Consider the evaluation measures that should be considered when designing the physical topology. Considering a backbone network as a design object, the physical topology requires connectivity between all edge nodes and redundancy that can maintain connectivity even when a failure occurs.
Therefore, here, as constraints in the physical topology design, (i) connectivity between all edge nodes, and (ii) connectivity between all edge nodes is maintained by a detour path even when one link fails. Think about two things.

There are various evaluation scales, but here we consider the following four examples. The i-th evaluation scale is denoted as V _i .
(1) Total number of nodes:
The total number of installed edge nodes and core nodes is the total, where Ne is the number of edge nodes and Nc is the number of installed core nodes, the first evaluation measure (V ₁ ) is expressed by the following equation (1).
V ₁ = Ne + Nc (1)
Since the number of edge nodes (Ne) is the same in all candidate topologies, the difference in the number of installed core nodes (Nc) becomes the difference in the _first evaluation scale (V ₁ ) as it is. The first evaluation scale (V ₁ ) is preferably small.
(2) Total link distance:
The second evaluation scale (V ₂ ) is expressed by the following expression (2), where d ₁ is the sum of the distances of the installed links, d ₁ is the distance of the link 1 and Εα is the set of installed links. It is desirable that the _second evaluation scale (V ₂ ) is also small.

(3) Sum obtained by weighting the path distance by the traffic volume:
From the viewpoint of transmission delay, it is desirable that the path length is short, and a sum (ζ) obtained by weighting the path length by the traffic amount of the path shown in the following equation (3) is considered as an evaluation measure. The following, however, a path through which traffic directed from the edge node s to the edge node d _{P sd,} path _{P sd} traffic amount _{T sd,} path _{P sd} is a set of links through [Phi _sd, the length of the path _{P sd} It is set as _Dsd shown to (4) Formula.
If t _l is the total traffic amount of link l, ζ coincides with the sum of products of link distance d _l and traffic amount t _l , so the total sum (ζ) is expressed by the following equation (5).
Therefore, the third evaluation scale (V ₃ ) is expressed by the following equation (6). It is desirable that the _third evaluation scale (V ₃ ) is also small.

(4) Maximum link load:
When traffic is concentrated on a specific link, the transmission quality of the path through the bottleneck link is greatly deteriorated. Also, since the transmission bandwidth of the link is increased in discrete units such as optical fibers and wavelengths, it is desirable to suppress and disperse the link load in order to effectively use the transmission resources. Therefore, considering the traffic amount of the maximum load link as an evaluation measure, the fourth evaluation measure (V ₄ ) is expressed by the following equation (7). It is desirable that the _fourth evaluation scale (V ₄ ) is also small.

Create candidate topology
When links and core nodes are arranged freely at a given installable position, the obtained topology candidates include those that do not satisfy the constraint conditions. Therefore, a physical topology that satisfies the constraint conditions is first created as an initial solution, and a set of physical topologies created by additionally deploying arbitrary core nodes and links to the initial solution is used as a solution supplement.
As a result, since the constraint condition is already satisfied at the stage of creating the initial solution, it is possible to adopt an arbitrary candidate topology.
Let z be the number of remaining link installable positions excluding the links arranged in the initial topology. The total number of topologies obtained as a result of placing a link to any installable position is 2 ^z pieces.
In each topology created in this way, a core node is placed at the core node installable position where the order is 3 or higher, but no exchange function is required at the core node installable position where the order is 2, so the core node is placed. First, two placement links are directly connected. In addition, the topology in which the position where the core node where the degree is 1 is present is an inappropriate topology and is excluded from the candidates for examination.

Next, for each topology created as a candidate, a logical path is set between the edge nodes using a given AC traffic matrix. Here, for the traffic flowing from the node s to the node d, only one path P _sd is installed, and the path P _sd is constructed in descending order of the requested traffic amount T _sd (once the arranged path is not changed). ) To set up a pass.
When determining the route of each path, the route that minimizes the traffic amount (ζ ′) shown in the following equation (8) is selected. However, the traffic amount of the link 1 at the time of installing the corresponding path is assumed to be _tl . Link capacity constraints are not taken into account when setting up paths. As a result of setting all the world paths, the topology in which no link passes through the path is an inappropriate topology and is excluded from the candidates for examination.

[Topology evaluation by average rank]
For each candidate topology i, the rank in the candidate set of wrinkles of each evaluation scale is calculated. Since all the evaluation scales are preferably as small as possible, the rank can be calculated simply by sorting the evaluation scales V _{k, i} in ascending order (k = 1, 2, 3, 4).
Using the rank R (k, i) obtained for each evaluation measure V _{k, i} of topology i in this way, the average rank (AR) of topology i is calculated according to the following equation (9). Complementary topologies with small rank (AR) values are evaluated as excellent topologies. That is, among all N candidate topologies, the topology having the smallest average rank (AR) is determined as the optimum network topology.
ARi = {R (1, i) + R (2, i) + R (3, i) + R (4, i)} / 4
.... (9)
Usually, it is not possible to simply add a plurality of evaluation scales having different units for evaluation. However, by substituting each evaluation scale with the same scale that can be compared with each other, the final superiority can be evaluated based on the average value.
In the average rank method, an arbitrary number of evaluation scales can be considered at the same time. In this case, the average rank (AR) of topology i is expressed by the following equation (10).
Calculation of the average rank, which is an evaluation value, is very simple. Sorting is required to determine the rank of each evaluation scale, but if a sort algorithm such as Merge sort or Quick sort is used, the processing amount of O (ΝlogΝ) is sufficient, so it takes much less time than DEA. Network topology design is possible.

[Numeric evaluation]
Next, the numerical evaluation results obtained by designing the network topology by the average ranking method using a specific network model will be described.
All calculations were performed on a personal computer (PC) with a 3.2 GHz CPU and 1 GB memory.
[Evaluation network model]
FIG. 2 shows a network model (Japanese archipelago model) used for evaluation in this example. However, black circles represent edge nodes, and circles and lines drawn with dotted lines represent positions where core nodes and links can be installed.
FIG. 2 shows the population U _k accommodated by each edge node k. The traffic amount T _sd flowing from the edge node s to the edge node d is proportional to the product of the population of the edge nodes at both ends, and is a value obtained by the following equation (11). However, T is the traffic amount of the entire network, and here, T = 10 Tbps.

Considering a loop topology as an initial topology that can secure a detour for one link failure and ensure connectivity between all edge nodes, a link is always placed at a link installable position indicated by a solid line in FIG.
Since the order of the core node installable positions through which the loop passes is 2 and the exchange function is unnecessary, there is no need to arrange the core nodes at these installable positions.
Except for the links used in the initial topology, 17 link installation positions remain, so 2 ¹⁷ physical topologies can be configured, but the topologies with core nodes or unused links of order 1 are candidates. As a result, 83,868 complemented topologies were obtained. The calculation time required to obtain these candidate topologies was about 75 seconds.

[Results of applying the average ranking method]
The amount of calculation required to calculate the average rank (AR) is very small, and even if the average rank (AR) is calculated for all 83,868 complemented topologies, the time required is less than 1 second.
Since a topology with a smaller average rank (AR) is a desirable topology, the final evaluation rank of candidate topologies is in the order of smaller average rank (AR).
FIG. 3 is a graph in which the average rank (AR) calculated from the above-described equation (5) is plotted in the order of final evaluation for 83,868 candidate topologies.
The minimum value, maximum value, and average of the average rank (AR) in 83,868 candidate topologies were 3813.25, 75295.25, and 39915.0, respectively.
Since some of the four evaluation measures are contradictory, there is no candidate topology in which all evaluation measures are very good, and conversely, all candidate evaluation measures are very bad. Therefore, the minimum value of the average rank (AR) is larger than 1, and the maximum value of the average rank (AR) is smaller than the number of candidate topologies 83,868.

It can be seen that the average rank (AR) of the majority of candidate topologies is in the range of 20,000-60,000, and the topologies whose average rank (AR) is close to the minimum value (361.25) are limited to a small number. .
This means that a small number of excellent network topologies can be selected by narrowing the selection candidates to only a small number of topologies having a small average rank (AR).
FIG. 4 is a table that summarizes the average rank (AR), the values of the four evaluation measures, and ranks in all the candidate topologies for the top ten top-ranked topologies with the small average rank (AR) in the graph shown in FIG. It is. These top 10 topologies are shown in FIG. It can be seen that a well-balanced complement topology with all four rating scales being highly appreciated by the average rank method.

FIG. 6 is a graph in which four evaluation measures are respectively plotted in order of candidate topologies having a high average evaluation rank (AR) and a high final evaluation rank. One point corresponds to one candidate topology. However, the compensated topology having an average rank (AR) of 10 or less is plotted with a large point.
It can be confirmed that candidate topologies having good evaluation scales are evaluated highly in the average ranking method. Further, it can be confirmed that the average rank (AR) of the physical topology having very good evaluation scales of V ₁ , V ₂ , and V ₄ is large and the evaluation by the average rank method is low.
This means that when a rope topology design is performed considering only a single evaluation scale, the other evaluation scales deteriorate, and the effectiveness of the average ranking method can be confirmed.
As described above, according to the present embodiment, when a location where a core node or link can be installed and an AC traffic matrix between edge nodes and edge nodes are given, a plurality of evaluation measures are taken into consideration. The optimal rope topology can be evaluated and determined by the average rank method.
As mentioned above, the invention made by the present inventor has been specifically described based on the above embodiments. However, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the scope of the invention. Of course.

It is a block diagram which shows schematic structure of the network topology design apparatus using the average order of the Example of this invention. In the Example of this invention, it is a figure which shows the network model used for evaluation. 3 is a graph showing a final evaluation rank of candidate topologies generated from the network model shown in FIG. 2. FIG. 4 is a table summarizing the average rank (AR), four evaluation scale values, and ranks among all the candidate topologies for the top 10 top-ranked topologies with the lowest average rank (AR) in the graph shown in FIG. 3. It is a figure which shows the top 10 topologies of the graph shown in FIG. It is a graph which shows each evaluation scale value plotted in order of the last evaluation order of the candidate topology produced | generated from the network model shown in FIG.

Explanation of symbols

101 topology information / demand matrix input device 102 initial topology generation device 103 topology candidate generation device 104 average rank calculation device 105 optimum topology output device

Claims (3)

When a set of edge nodes that accommodate users, a set of candidate core nodes that have only a relay function, and a set of link candidates that connect them, are used while simultaneously considering multiple evaluation measures A network topology design method for selecting a set of core nodes and links,
For each topology candidate, calculate the rank among all candidates for each evaluation measure considered simultaneously,
A network topology design method, comprising: calculating an average value of ranks obtained for each evaluation scale, and determining superiority or inferiority among candidates based on the average rank.   The evaluation scale to be considered at the same time uses the total number of nodes, the total link length, the sum of the length of each path and the traffic volume for all paths, and the traffic volume of the maximum load link. Item 2. The network topology design method according to Item 1. When a set of edge nodes that accommodate users, a set of candidate core nodes that have only a relay function, and a set of link candidates that connect them, are used while simultaneously considering multiple evaluation measures A network topology design device for selecting a set of core nodes and links,
Topology information / demand matrix input device for inputting the position of the edge node, the installable position of the core node, and the installable position of the link;
An initial topology generation device for generating an initial topology based on the position of the edge node, the installable position of the core node, and the installable position of the link, which are input by the topology information / demand matrix input device;
A topology candidate generation device that generates a plurality of candidate topologies by adding several links and core nodes to the initial topology generated by the initial topology generation device;
For a plurality of candidate topologies generated by the topology candidate generation device, the rank among all candidates is calculated for each evaluation scale to be considered simultaneously, the average value of the ranks obtained for each evaluation scale is calculated, and the average An average rank calculation device that judges superiority or inferiority among candidates by rank;
A network topology design device, comprising: an optimum topology output device that outputs a set of optimum topologies based on an evaluation result in the average rank calculation device.

Images