JP2006319849A - Band guarantee communication system between end users - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To guarantee a requested band for over the entire edge nodes about traffic between each end user terminal. <P>SOLUTION: A communication starting request specifying the requested band with a receiving side end user terminal G is transmitted from a transmitting side end user terminal A toward an edge node I. In a network, band guarantee traffics are transferred by setting a path having the requested band between edge nodes I and L corresponding to the transmitting side and receiving side end user terminals A, G according to the communication starting request. Reserved setting or constant setting of the path between the edge nodes are possible. In each of the edge nodes I-L, the band guarantee traffics are monitored, a band guarantee traffic exceeding the requested band is marked, each node in the network preferentially transfers the band guarantee traffics and an amount of the band guarantee traffic of a buffer exceeds a threshold, discards the marked band guarantee traffic. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an end-user bandwidth guaranteed communication system, and more particularly to an end-user bandwidth guaranteed communication system that provides a bandwidth-guaranteed communication service to end users by applying a path technology based on the GGPLS protocol between edge nodes of a network. About.

  The configuration of a conventional bandwidth guarantee communication system is shown in FIG. In a network that provides a bandwidth-guaranteed communication service, an upper area and a lower area are hierarchically configured. Area boundary nodes M and N are arranged at the boundary between the upper area and the lower area, and edge nodes I to L are arranged in the lower area. A relay node is also provided in each area, but it is not shown. End user terminals A to H are connected to the edge nodes I to L.

Conventionally, bandwidth-guaranteed communication is realized by setting a path of a predetermined bandwidth in the upper area. That is, by collecting traffic between a plurality of end user terminals and accommodating the traffic in a path set in the upper area, the bandwidth of the traffic passing through the path is guaranteed as a whole. Further, the amount of traffic accommodated in the path is monitored at the area boundary nodes M and N, and the bandwidth of the path is controlled according to the amount of traffic. Such a technique is described in Non-Patent Document 1.
K.Shiomoto, E.Oki, W.Imajuku, S.Okamoto, and N.Yamanaka, "Distributed virtual network topology control mechanism in GMPLS-based multiregion networks," IEEE JSAC, vol.21, no.8, pp.1254 -1262 (Oct. 2003).

  However, the conventional bandwidth-guaranteed communication system guarantees the bandwidth of the traffic passing through the path set in the upper area of the network as a whole, so that the bandwidth is not necessarily guaranteed for the traffic between individual end-user terminals. There are challenges.

  Further, although the bandwidth is guaranteed in the upper area, there is a problem that the bandwidth is not guaranteed across the entire edge node including the lower area. Furthermore, in order to control the bandwidth of the path according to the traffic volume, it is necessary to monitor the traffic volume accommodated in each path at the area boundary node, and there is a problem that the processing load at the area boundary node increases.

  The object of the present invention is to solve the above-mentioned problems, and to guarantee the required bandwidth across the entire edge nodes for the traffic between individual end-user terminals, and reduce the processing amount and processing delay in the network accompanying the provision of bandwidth-guaranteed communication services. It is an object of the present invention to provide a bandwidth guarantee communication system between end users.

  In order to solve the above-described problem, the present invention provides an end-user bandwidth guaranteed communication in a network in which area boundary nodes exist between hierarchically configured areas and end user terminals are connected in a star shape to the edge nodes. In the system, at least one of the on-demand setting of the inter-edge node path, the reservation setting of the inter-edge node path, and the constant setting of the inter-edge node path can be performed in the network. A communication start request or a communication end request designating a requested bandwidth with the called end user terminal is sent from the side end user terminal to the edge node, and in the network, in response to the communication start request, the calling side and the called end Set up a path with the required bandwidth between the edge nodes corresponding to the user terminals to maintain the bandwidth. The traffic is transferred, and the path set between the edge nodes is released in response to the communication end request. In the reservation setting of the path between the edge nodes, the communication start time, the communication end time, the calling side and the called side are set from the end user terminal. The end user terminal sends a communication request specifying the requested bandwidth, and when the communication start time specified in the communication request comes within the network, the request is made between the edge nodes corresponding to the specified originating and terminating end user terminals. A path with bandwidth is set and bandwidth guaranteed traffic is transferred. When the communication end time specified in the communication request is reached, the path set between the edge nodes corresponding to the specified caller and callee end user terminals is released. In the constant setting of the path between the edge nodes, a path of a predetermined band is set in advance between the edge nodes, and the called side from the calling end user terminal is set. In response to a communication start request and a communication end request specifying a requested band, or a communication start time, a communication end time, a calling side and a calling end user terminal from the end user terminal, a communication request specifying a requested band Accordingly, the bandwidth guaranteed traffic between the calling end user terminal and the called end user terminal is transferred using the preset path, and at each edge node, between the calling end user terminal and the called end user terminal. Bandwidth-guaranteed traffic is monitored and marked for bandwidth-guaranteed traffic that exceeds the required bandwidth, and each node in the network has two buffers prepared for bandwidth-guaranteed traffic and non-bandwidth-guaranteed traffic for each outgoing link. Used and set on the outgoing link among the bandwidth guaranteed traffic stored in the buffer for the bandwidth guaranteed traffic. Bandwidth-guaranteed traffic that exceeds the total amount of each path bandwidth accumulated in the buffer for bandwidth-guaranteed traffic, and non-bandwidth-guaranteed traffic that is preferentially transferred using each path, up to the total amount of each bandwidth The non-bandwidth guaranteed traffic accumulated in the traffic buffer is non-prioritized and is transferred using each path and the remaining link bandwidth, and the amount of bandwidth guaranteed traffic accumulated in the buffer for bandwidth guaranteed traffic. The first characteristic is that when the value exceeds the preset threshold value, the marked bandwidth guaranteed traffic is preferentially discarded without accumulating in the buffer.

  Further, the present invention provides each edge node with a leaky bucket for each band guaranteed traffic between the calling end user terminal and the called end user, and for the band guaranteed traffic when the leaky bucket exceeds a threshold value. There is a second feature in marking.

  Further, according to the present invention, when each node in the network is triggered by a trigger generated at regular intervals, the bandwidth guaranteed traffic stored in the bandwidth guaranteed traffic buffer is equal to a time corresponding to the total amount of the outgoing path bandwidth ( = Trigger generation interval x total of outgoing path bandwidth / outgoing link bandwidth), and the bandwidth guaranteed traffic corresponding to the total of outgoing path bandwidth or whether the bandwidth guaranteed traffic buffer becomes empty A third feature is that, if the transmission time has elapsed, the accumulated traffic is alternately transmitted from the bandwidth guarantee traffic buffer and the non-bandwidth guarantee traffic buffer until the next trigger occurs.

  Further, according to the present invention, when the bandwidth guarantee traffic buffer becomes empty before each node in the network passes the bandwidth guarantee traffic transmission time, the difference is added to the original bandwidth guarantee traffic transmission time. The fourth feature is that the time is used as the bandwidth-guaranteed traffic transmission time of the next cycle (however, the bandwidth-guaranteed traffic transmission time in each cycle is always less than or equal to the trigger generation interval).

  In addition, the present invention has a fifth feature in that the edge node performs the charging process based on the requested bandwidth and the duration of the bandwidth guaranteed traffic transfer.

  In the present invention, the originating area boundary node collects and accommodates a plurality of lower area paths heading to the same area boundary node in the upper area in one upper area path, and transfers the band guaranteed traffic collectively. The side area boundary node distributes the bandwidth-guaranteed traffic transferred through the upper area path to the plurality of lower area paths accommodated in the upper area path. There is a sixth feature in that the bandwidth is managed and the bandwidth of the upper area path that accommodates the bandwidth is controlled according to the total bandwidth of each lower area path.

  Further, the present invention increases the bandwidth of the upper area path when the originating area boundary node has a ratio of the total bandwidth of the accommodated lower area path to the bandwidth of the upper area path greater than a predetermined threshold, A seventh feature is that the bandwidth of the upper area path is reduced when the ratio of the total bandwidth of the accommodated lower area path to the bandwidth of the upper area path becomes smaller than another predetermined threshold.

  The present invention also manages the required bandwidth of each end-user bandwidth guaranteed traffic transferred by the edge node using a preset inter-edge node path, and the total required bandwidth of each end-user bandwidth guaranteed traffic. Accordingly, an eighth feature is that the bandwidth of the path between the edge nodes that transfers them is controlled.

  Furthermore, the present invention provides an inter-edge node path bandwidth when the ratio of the required bandwidth guarantee traffic total bandwidth with respect to the pre-set inter-edge node bandwidth is greater than a predetermined threshold. The path bandwidth between edge nodes is reduced when the ratio of the required bandwidth guarantee traffic total bandwidth to the pre-set path bandwidth between edge nodes becomes smaller than another predetermined threshold. There is a ninth feature.

  According to the present invention, at least the required bandwidth of traffic between individual end-user terminals is guaranteed over the entire edge nodes of the network, so that a high-quality communication service can be provided to the end users.

  In addition, each node in the network transfers traffic using a single bandwidth-guaranteed traffic buffer for a plurality of paths passing through each outgoing link, so that the transfer does not depend on the number of bandwidth-guaranteed traffic transfer paths accommodated. Can be processed, and traffic transfer processing can be simplified.

  Also, if a path is always set between edge nodes, there is no need to set a path between edge nodes for every request to start bandwidth guarantee traffic communication. Delay can be reduced.

  In addition, the edge node can easily perform billing processing based only on the requested bandwidth and the communication duration time without monitoring the traffic volume every moment.

  In addition, it is not necessary to monitor the traffic volume at the area boundary node, it is only necessary to manage the bandwidth (requested bandwidth) of the accommodation lower area path, and the bandwidth control of the upper area path at the area boundary node is simplified, and the accommodation lower layer By setting an upper area path having a marginal band compared to the total band of area paths, the frequency of bandwidth control of the upper area path can be reduced.

  Furthermore, the edge node does not need to monitor the traffic volume, it is only necessary to manage the required bandwidth of the guaranteed bandwidth, and the bandwidth control of the path between the edge nodes in the edge node when the path between the edge nodes is always set. The frequency of frequency control of the path between the edge nodes can be reduced by setting the path between the edge nodes having a marginal band compared to the total required bandwidth of the accommodated band guaranteed traffic.

  The present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a network to which a band guarantee communication system between end users according to the present invention is applied. In addition, the same code | symbol is attached | subjected to the same or equivalent part as FIG.

  As shown in FIG. 1, individual end user terminals A to H are connected to the edge nodes I to L of the network in a star shape by a subscriber link having a certain physical band. The network is configured to have a plurality of areas in a hierarchical manner, for example, an upper area and a lower area. Area boundary nodes M and N exist between the two areas. Each node in the network including the edge nodes I to L and the area boundary nodes M and N can set and release a path of an arbitrary band using a signaling protocol.

  An end-user bandwidth guaranteed communication system according to the present invention will be described below on the premise of the network. First, the case of on-demand setting of a path between edge nodes, that is, a case where a path is set or released according to a request from an end user terminal will be described.

  When transferring traffic in response to the bandwidth guarantee communication service, the user sends a communication start request to the edge node from the calling end user terminal, specifying the requested bandwidth with the called end user terminal. The requested bandwidth is equal to or less than the subscriber link bandwidth that connects the calling end user terminal and the edge node, and is equal to or less than the subscriber link bandwidth that connects the called end user terminal to the edge node.

  On the network side, a GMPLS protocol or the like is used, a path having a requested bandwidth is set between specified edge nodes in response to a communication start request, and bandwidth guaranteed traffic between the calling end user terminal and the called end user is set. Forward.

  After the bandwidth guaranteed traffic transfer, when the user sends a communication end request from the calling end user terminal to the edge node toward the edge node, the network side sets it between the edge nodes according to the communication end request. Release the path that was left.

  For example, if the calling end user terminal is A and the called end user terminal is G, the user specifies the called end user terminal G and the requested bandwidth from the calling end user terminal A toward the edge node I. Send a communication start request. The requested bandwidth is equal to or less than the subscriber link bandwidth connecting the calling end user terminal A and the edge node I, and is equal to or less than the subscriber link bandwidth connecting the called end user terminal G and the edge node L.

  On the network side, a path having a requested bandwidth is set between the edge nodes I and L in response to the communication start request, and the bandwidth guaranteed traffic between the calling end user terminal A and the called end user terminal G is transferred.

  After the bandwidth-guaranteed traffic transfer, when the user sends a communication end request from the calling end user terminal A to the edge node I to the destination end user terminal G, on the network side, in response to the communication end request, the edge node I And the path set between L and L is released.

  Next, the case of reservation setting for the path between edge nodes will be described. In this case, the user designates the communication start time, the communication end time, the calling and called end user terminals, and the requested bandwidth from the end user terminal, and sends out a communication request in advance. The requested bandwidth is equal to or less than the subscriber link bandwidth that connects the calling end user terminal and the edge node, and is equal to or less than the subscriber link bandwidth that connects the called end user terminal to the edge node.

  When the communication start time specified by the communication request is reached, the network side sets a path having the required bandwidth between the edge nodes corresponding to the specified calling and called end user terminals, and arrives at the calling end user terminal. Transfers bandwidth guaranteed traffic between end-users. Further, when the communication end time designated by the communication request comes, the path set between the edge nodes corresponding to the designated originating and terminating end user terminals is released.

  For example, if the calling end user terminal is A and the called end user terminal is G, the user can send a communication start time, a communication end time, a calling end user from the calling end user terminal A or another end user terminal. A communication request is transmitted in advance by designating the terminal A, the called end user terminal G, and the requested bandwidth. The requested bandwidth is equal to or less than the subscriber link bandwidth connecting the originating end user terminal A and the edge node I, and is equal to or less than the subscriber link bandwidth connecting the terminating end user terminal G to the edge node L.

  On the network side, when the communication start time specified by the communication request is reached, a path having a required bandwidth is set between the edge node I and the edge node L, and the bandwidth between the originating end user terminal A and the terminating end user terminal G is set. Forward guaranteed traffic. Further, when the communication end time designated by the communication request comes, the path set between the edge nodes I and L is released.

  Next, the case of always setting the path between edge nodes will be described. The path between edge nodes is always set by always setting a path of a predetermined band between edge nodes that are expected to receive a large amount of bandwidth-guaranteed traffic. Specified according to the communication start request and communication end request specifying the band, or according to the communication request from the end user terminal, communication start time, communication end time, originating and destination end user terminals and the requested band The guaranteed bandwidth traffic between the calling end user terminal and the called end user terminal is transferred using the path.

  For example, a predetermined bandwidth path is always set between the edge nodes I and L, which are expected to have a large amount of bandwidth guaranteed traffic to be transferred. Here, if there is a receiving end user terminal G from the calling end user terminal A, a communication start request and a communication end request specifying the requested band, the bandwidth between the calling end user terminal A and the called end user terminal G Forward guaranteed traffic using that path. Alternatively, if there is a communication request in which a communication start time, a communication end time, a calling end user terminal A and a calling end user terminal G, and a requested bandwidth are specified by the user, between the calling end user terminal A and the calling end user terminal G The bandwidth guaranteed traffic is transferred using the path from the communication start time to the communication end time.

  According to the constant setting of the path between edge nodes, there is no need to set a path each time a communication start request is made. Therefore, the amount of processing in the network can be reduced, and the processing delay can be reduced, so that there is an advantage that the service can be started immediately in response to the communication start request.

  Next, accounting processing at the edge node will be described. Unlike the best effort type service such as the Internet, the bandwidth guaranteed communication system according to the present invention secures a certain band resource and provides it to the end user. Therefore, the billing according to the resource usage is separate from the normal network connection fee. It is preferable to carry out.

  Each edge node can easily perform a billing process for a bandwidth-guaranteed communication service based on a bandwidth-guaranteed traffic transfer request bandwidth and a communication duration. Based on the result of this billing process, in addition to the normal network connection fee, the end-user bandwidth guarantee service fee may be billed to each end user terminal.

  For example, charging processing is performed at the edge node I based on the required bandwidth for the bandwidth guaranteed traffic transfer between the calling end user terminal A and the called end user terminal G and the duration of communication, and based on the result of the charging processing, In addition to a normal network connection fee, a bandwidth guarantee service fee between end users can be charged to, for example, the calling end user terminal A.

  Next, the marking process at the edge node will be described. The marking is performed in order to enable a measure that, when a bandwidth-guaranteed traffic exceeding the requested bandwidth is transmitted, traffic for the requested bandwidth is guaranteed but traffic exceeding the requested bandwidth is not necessarily guaranteed.

  The edge node monitors the bandwidth guaranteed traffic between the calling end user terminal and the called end user, and performs marking for the bandwidth guaranteed traffic (packet) exceeding the requested bandwidth. For example, the edge node I monitors the bandwidth guaranteed traffic between the calling end user terminal A and the called end user terminal G, and performs marking for the bandwidth guaranteed traffic exceeding the requested bandwidth.

  As shown in FIG. 2, the marking process for the bandwidth guaranteed traffic exceeding the requested bandwidth is provided with a leaky bucket for each bandwidth guaranteed traffic between the calling end user terminal and the called end user at each edge node. This can be achieved by marking when the distance is exceeded. For example, in the edge node I, a leaky bucket is provided for the bandwidth guaranteed traffic between the calling end user terminal A and the called end user terminal G, and marking is performed when the leaky bucket exceeds a threshold value.

  FIG. 3 is a diagram showing a traffic transfer process at each node in the network. As shown in FIG. 3, two buffers for bandwidth guaranteed traffic and non-bandwidth guaranteed traffic are prepared at each node for each outgoing link, and the bandwidth guaranteed traffic is preferentially transferred. In other words, of the bandwidth guaranteed traffic stored in the buffer for bandwidth guaranteed traffic, traffic up to the total of the path bandwidths set on the outgoing link is preferentially transferred using each path.

  Bandwidth-guaranteed traffic that exceeds the total amount of each path bandwidth accumulated in the buffer for bandwidth-guaranteed traffic and non-bandwidth-guaranteed traffic accumulated in the buffer for non-bandwidth-guaranteed traffic are non-prioritized. Transfer using the remaining link bandwidth. Further, when the amount of bandwidth guaranteed traffic accumulated in the buffer for bandwidth guaranteed traffic exceeds a preset threshold value, the marked bandwidth guaranteed traffic is preferentially discarded without being accumulated in the buffer.

  In this way, by preparing one band-guaranteed traffic buffer for a plurality of paths passing through each outgoing link, a transfer process independent of the number of paths passing through can be performed, and the transfer process can be simplified.

  For example, in the relay node in the lower area located between the edge node I and the area boundary node M, the preferential transfer process is performed as follows.

  The relay node in the lower area includes a path set between the edge nodes I and L in order to pass bandwidth guaranteed traffic between the calling end user terminal A and the called end user terminal G, and the calling end user terminal B The path set between the edge nodes I and L in order to pass the bandwidth guaranteed traffic between the called end user terminals H is accommodated in the same outgoing link. Therefore, two buffers for bandwidth guaranteed traffic and other non-bandwidth guaranteed traffic transferred using these two paths are prepared corresponding to the outgoing link. Of the bandwidth guaranteed traffic stored in the bandwidth guaranteed traffic buffer, traffic up to the total of the two accommodated path bandwidths is preferentially transferred using each path.

  Bandwidth-guaranteed traffic that exceeds the total path bandwidth accumulated in the buffer for bandwidth-guaranteed traffic and non-bandwidth-guaranteed traffic that is accumulated in the buffer for non-bandwidth-guaranteed traffic are non-prioritized and each has two paths. And transfer using the remaining link bandwidth. In addition, when the amount of bandwidth guaranteed traffic accumulated in the buffer for bandwidth guaranteed traffic exceeds a preset threshold, it does not depend on which path the marked bandwidth guaranteed traffic was transferred to. And do not accumulate in the buffer and discard it preferentially.

  FIG. 4 is a diagram specifically illustrating the traffic transfer process in each node. At each node, triggered by a trigger generated at regular intervals, the bandwidth guaranteed traffic stored in the buffer for bandwidth guaranteed traffic is equal to the time corresponding to the total of the outgoing path bandwidth (= trigger generation interval x outgoing side) Only the total path bandwidth / outgoing link bandwidth) is sent to each outgoing path. Here, if the bandwidth guarantee traffic buffer becomes empty or if the bandwidth guarantee traffic sending time corresponding to the total of the outgoing path bandwidth has elapsed, the bandwidth guarantee traffic buffer and the buffer are used until the next trigger occurs. The accumulated traffic is alternately transmitted from the non-bandwidth guaranteed traffic buffer.

  If the bandwidth guarantee traffic buffer becomes empty before the bandwidth guarantee traffic transmission time elapses, the time obtained by adding the difference and the original bandwidth guarantee traffic transmission time is used as the bandwidth guarantee traffic transmission time for the next period. You may make it do. However, the bandwidth guaranteed traffic transmission time in each cycle is always less than or equal to the trigger generation interval.

  Next, path multiplexing processing and path bandwidth change processing at the area boundary node will be described. In the originating area boundary node, a plurality of lower area paths heading to the same area boundary node in the upper area are collected and accommodated in one upper area path, and the band guaranteed traffic is collectively transferred. The destination area boundary node distributes the bandwidth-guaranteed traffic transferred through the upper area path to a plurality of lower area paths accommodated in the upper area path. Furthermore, the originating area boundary node manages the bandwidth of each lower area path, and controls the bandwidth of the upper area path that accommodates the bandwidth according to the total bandwidth of each lower area path.

  In the area boundary node and the upper area node, it is not necessary to individually manage many paths from the lower area. In addition, since it is only necessary to control the bandwidth of the path based on the requested bandwidth at the area boundary node, it becomes unnecessary to observe the traffic fluctuation and control the bandwidth of the path accordingly, thereby reducing the management load. .

  For example, the area boundary node M includes a lower area path between two edge nodes I and L and a lower area path between one edge node J and K toward the area boundary node N in the upper area. Are collected and accommodated in one upper area path between N and N, and the bandwidth guarantee traffic is collectively transferred. In the destination area boundary node N, the bandwidth guaranteed traffic transferred through the upper area path between the area boundary nodes M and N is distributed to the three lower area paths accommodated in the upper area path. Furthermore, the originating area boundary node M manages the bandwidth of the three lower area paths, and according to the total bandwidth of each lower area path, the upper area between the area boundary nodes M and N that accommodate them. Control the bandwidth of the path.

  The originating area boundary node increases the upper area path bandwidth when the ratio of the total accommodated lower area path bandwidth to the upper area path bandwidth is greater than a predetermined threshold, When the ratio of the total path bandwidth becomes smaller than another predetermined threshold, the upper area path bandwidth is decreased.

  For example, in the originating area boundary node M, the ratio of the total of the three lower area path bands accommodated by this path to the upper area path band between the area boundary nodes M and N is larger than a predetermined threshold value. Then, the upper area path bandwidth between the area boundary nodes M and N is increased. Conversely, when the ratio of the total of the three lower area path bands accommodated by this path to the upper area path band between the area boundary nodes M and N becomes smaller than another predetermined threshold, The upper area path band between the boundary nodes M and N is reduced.

  Next, bandwidth guaranteed traffic multiplexing processing and path bandwidth change processing in the edge node in the case of always setting the path between edge nodes will be described. In the case of always setting the path between the edge nodes, the edge node manages the required bandwidth of each inter-user bandwidth guarantee traffic transferred using the path between the edge nodes set in advance. The bandwidth of the path between the edge nodes that transfers them is controlled according to the total required bandwidth of the guaranteed traffic. The edge node can be easily managed because the bandwidth of the path may be controlled based on the requested bandwidth.

  For example, in the edge node I, the bandwidth-guaranteed traffic between the calling end user terminal A and the called end user terminal G and the calling end user using one path between the edge nodes I and L set in advance. The bandwidth guarantee traffic between the terminal B and the called end user terminal H is transferred. The edge node I manages the required bandwidth of the bandwidth guaranteed traffic between the two end users to be transferred, and the edge nodes I and L that transfer the bandwidth according to the total required bandwidth of the bandwidth guaranteed traffic between the end users. The bandwidth of the path set in advance is controlled.

  The edge node increases the path bandwidth between edge nodes when the ratio of the total bandwidth required for the guaranteed bandwidth for traffic to the path bandwidth between the edge nodes set in advance is larger than a predetermined threshold. On the other hand, when the ratio of the required bandwidth guarantee traffic total bandwidth to the preset inter-edge node path bandwidth becomes smaller than another predetermined threshold, the inter-edge node path bandwidth is decreased.

  For example, the edge node I is configured such that the bandwidth-guaranteed traffic between the calling end user terminal A and the called end user terminal G for the path band set in advance between the edge nodes I and L, and the calling end user terminal B and the called side When the ratio of the required bandwidth of the bandwidth guaranteed traffic between the end user terminals H becomes larger than a predetermined threshold, the path bandwidth between the edge nodes I and L is increased. Conversely, bandwidth guarantee traffic between the calling end user terminal A and the called end user terminal G, the calling end user terminal B, and the called end user terminal H for the path band set in advance between the edge nodes I and L. When the ratio of the total required bandwidth of the bandwidth guaranteed traffic between the two becomes smaller than another predetermined threshold, the path bandwidth between the edge nodes I and L is decreased.

  According to the present invention, since a requested bandwidth from an end user is guaranteed over the entire edge node, a high-quality communication service can be provided to the end user.

It is a block diagram which shows one Embodiment of the network to which the band guarantee communication system between end users which concerns on this invention was applied. It is explanatory drawing of the marking process in an edge node. It is explanatory drawing of the bandwidth guarantee traffic transfer process in each node. It is explanatory drawing of ensuring of the bandwidth guarantee traffic transfer band in each node. It is a block diagram which shows the structure of the conventional band guarantee communication system.

Explanation of symbols

A to H: End user terminal, I to L: Edge node, M, N: Area boundary node

Claims (9)

In an end-user bandwidth guarantee communication system in a network in which area boundary nodes exist between hierarchically configured areas and end user terminals are connected in a star shape to edge nodes,
In the network, at least one of on-demand setting of the path between edge nodes, reservation setting of the path between edge nodes, and constant setting of the path between edge nodes is possible,
In the on-demand setting of the path between the edge nodes, a communication start request or a communication end request designating a requested bandwidth with the callee end user terminal is transmitted from the calling end user terminal to the edge node. In response to the request, a path having a requested bandwidth is set between the edge nodes corresponding to the originating and terminating end user terminals to transfer the bandwidth guarantee traffic, and set between the edge nodes in response to the communication end request. Free the path,
In the reservation setting of the path between edge nodes, the end user terminal sends a communication request specifying the communication start time, communication end time, originating and terminating end user terminals, and the requested bandwidth. When the specified communication start time is reached, a path having the requested bandwidth is set between the edge nodes corresponding to the designated originating and terminating end user terminals, and the bandwidth guarantee traffic is transferred, and the communication end time designated by the communication request The path established between the edge nodes corresponding to the specified originating and terminating end user terminals is released,
In the always setting of the path between the edge nodes, a path of a predetermined band is set in advance between the edge nodes, and a call start request and a communication end request specifying the requested band from the caller end user terminal Depending on the communication start time from the end user terminal, communication end time, originating and terminating end user terminals, communication request specifying the requested bandwidth, between the originating end user terminal and the terminating end user terminal Bandwidth guaranteed traffic is forwarded using the preset path,
At each edge node, the bandwidth guaranteed traffic between the originating end user terminal and the called end user terminal is monitored and marked for bandwidth guaranteed traffic exceeding the requested bandwidth,
Each node in the network uses two buffers prepared for band-guaranteed traffic and non-bandwidth-guaranteed traffic corresponding to each outgoing link, and among the band-guaranteed traffic stored in the buffer for band-guaranteed traffic. Bandwidth guarantee that exceeds the total of each path bandwidth stored in the buffer for bandwidth guarantee traffic by preferentially forwarding the traffic up to the total of each path bandwidth set on the outgoing link using each path. Traffic and non-guaranteed traffic stored in the buffer for non-bandwidth-guaranteed traffic are non-prioritized, are transferred using each path and the remaining link bandwidth, and are accumulated in the buffer for bandwidth-guaranteed traffic. If the amount of bandwidth guaranteed traffic exceeds the preset threshold, the marked bandwidth guaranteed traffic is backed up. End users between bandwidth guaranteed communication system and discards preferentially not accumulate in §. Each edge node has a leaky bucket for each bandwidth guaranteed traffic between a calling end user terminal and a called end user, and marks the bandwidth guaranteed traffic when the leaky bucket exceeds a threshold value. The end-user bandwidth guaranteed communication system according to claim 1. Each node in the network is triggered by a trigger generated at a constant interval, and the bandwidth guaranteed traffic stored in the bandwidth guaranteed traffic buffer is calculated as a time corresponding to the total of the outgoing path bandwidth (= trigger generation interval × (Total outgoing path bandwidth / outgoing link bandwidth) is sent to each outgoing path, and the bandwidth guaranteed traffic buffer becomes empty or the bandwidth guaranteed traffic sending time corresponding to the total outgoing path bandwidth has elapsed. 2. The inter-user bandwidth according to claim 1, wherein the accumulated traffic is alternately transmitted from the bandwidth guarantee traffic buffer and the non-bandwidth guarantee traffic buffer until the next trigger occurs. Guaranteed communication system. If the bandwidth guarantee traffic buffer becomes empty before the bandwidth guarantee traffic transmission time elapses, each node in the network adds the difference and the original bandwidth guarantee traffic transmission time to the next period. 4. The inter-end-user bandwidth-guaranteed communication system according to claim 3, wherein the bandwidth-guaranteed traffic is transmitted as a bandwidth-guaranteed traffic transmission time (however, the bandwidth-guaranteed traffic transmission time in each cycle is always equal to or less than a trigger generation interval). 2. The end-user bandwidth guaranteed communication system according to claim 1, wherein the edge node performs charging processing based on a requested bandwidth and duration of bandwidth-guaranteed traffic transfer. The originating area boundary node collects and accommodates a plurality of lower area paths that go to the same area boundary node in one upper area path in a higher area, and transfers the band guaranteed traffic together. The bandwidth guaranteed traffic transferred through the upper area path is distributed to a plurality of lower area paths accommodated by the upper area path, and the originating area boundary node manages the bandwidth of each lower area path, The end-user band guaranteed communication system according to claim 1, wherein the bandwidth of the upper area path that accommodates the lower area path is controlled according to the total bandwidth of the lower area path. The originating area boundary node increases the bandwidth of the upper area path and increases the bandwidth of the upper area path when the ratio of the total bandwidth of the accommodated lower area path to the bandwidth of the upper area path is greater than a predetermined threshold. The end-user bandwidth guarantee communication according to claim 6, wherein when the ratio of the total bandwidth of the accommodated lower area path becomes smaller than another predetermined threshold, the bandwidth of the upper area path is decreased. system. The edge node manages the required bandwidth of each end-user bandwidth guaranteed traffic that is transferred using the path between the edge nodes set in advance, and determines them according to the total required bandwidth of each end-user bandwidth guaranteed traffic. 2. The inter-end-user bandwidth guaranteed communication system according to claim 1, wherein a bandwidth of an inter-edge node path to be transferred is controlled. The edge node increases the inter-edge node path bandwidth when the ratio of the required bandwidth guarantee traffic total bandwidth to the pre-established inter-edge node pass bandwidth exceeds a predetermined threshold value. The inter-edge node path band is reduced when the ratio of the total bandwidth required for the guaranteed bandwidth to the inter-edge node path band is smaller than another predetermined threshold. The end-user bandwidth guaranteed communication system according to claim 8.

Images