JP5833184B2 - Network congestion avoidance system and method - Google Patents

Description

  The present invention relates to a network congestion avoidance system and method, and in particular, when transmitting data that needs to ensure communication quality such as voice and video over an IP (Internet Protocol) network, the communication quality deterioration over the IP network. The present invention relates to a network congestion avoidance system and method for controlling traffic in order to avoid traffic.

  In an application that takes a server / client form such as content distribution, when one viewer views a certain content, usually the starting point is a single transmission server storing the content, and the viewer Data is delivered along one route on the IP network that ends with one client terminal to be used. In the case of such a one-to-one communication method, in applications where delivery time is restricted such as video viewing in real time, it is necessary to deliver the data in all parts of the route from the server to the client. It is necessary to secure a stable network bandwidth and perform stable data delivery at a constant bit rate. However, when the delivery data has a large capacity and requires a large bandwidth, it is not easy to secure the bandwidth as described above, and normal delivery of the data cannot be performed if there is a location where the bandwidth cannot be secured even at one location on the route.

  In order to solve the above problem, a technique has been proposed in which one content is divided and stored in storage devices on a plurality of physically different transmission servers (see, for example, Non-Patent Document 1). In such a method, when reproducing the content, the data of the content are transmitted all at once from the plurality of transmission servers (hereinafter referred to as the transmission server together with the storage device and the computer on which the storage device is mounted). A many-to-one communication method is used in which a plurality of flows are delivered to one receiving terminal. This communication method is shown in FIG.

  In the many-to-one communication method as shown in the figure, only one part of the content data is transmitted from one transmission server, and the data is gradually aggregated on the IP network, and finally one unit. All the data of the content is collected in the receiving terminal. Therefore, even when the content is played back in real time, the portion close to the transmission server in the IP network only needs to have a bandwidth that allows real-time distribution of a part of the data of the content. Compared to this, there is an advantage that it is easy to secure the bandwidth in the IP network. In other words, even if it is not always possible to secure the bandwidth required for real-time distribution of the entire content in all parts from the transmission server to the receiving terminal, the actual content of the content is effectively used in a plane that effectively uses the bandwidth resources of the IP network. There may be time distribution.

Kunitake Kaneko, et al., "Content Espresso: A Global Secure Large File Sharing System for Media Industries," Proc. Of the 13th ACIS International Conference on Software Engineering, Artifical Intelligence, Networking and Parallel / Distributed Computing (SNPD 2012), pp 649-654, August 2012. OPEN NETWORKING FOUNDATION, "OpenFlow Switch Specification Version 1.4.0 (Wire Protocol 0x05)," October 14, 2013. E. Rosen, et al., "Multiprotocol Label Switching Architecture," RFC3031, January 2001. Akihiro Nakao, "Network virtualization in the new generation network concept," IEICE Journal, Vol.94, No.5, pp.385-388, 2011. G. Dommety, "Key and Sequence Number Extensions to GRE," RFC2890, September 2000. Kazune Hasegawa, et al., "Proposal of Active Bandwidth Active Measurement Method Using Different Rate Packet Train Pairs," 251st Meeting of the Society of Instrument and Control Engineers, Tohoku Branch, No.251-4, 2009. Yoji Ozawa and Hideki Okita, "A Study on User-Oriented Network System with Multiple VLAN Access", Information Processing Society of Japan, 2007-DSM-45 (2), pp. 7-12, 2007.

  The many-to-one communication method has the advantages as described above. However, in order to ensure communication quality, transmission is performed at the start of data transmission in order to secure bandwidth before starting data transmission or to avoid locations where congestion or communication failure has occurred on the network. There is a problem that processing is complicated when it is attempted to explicitly specify a route or to perform traffic arrangement control that changes a transmission route during data transmission. Furthermore, there is a problem that setting of the device on the network service provider side is required to perform such control in the prior art. Details are described below.

  In a normal IP network that performs routing control using a general IP routing protocol such as IS-IS (Intermediate System to Intermediate System) or OSPF (Open Shortest Path First), each of the preconfigured networks Depending on the metric value of the link, the transmission path of the IP packet is determined, and other transmission paths cannot be selected. For this reason, even if a bandwidth is secured in advance for the purpose of securing communication quality, transmission cannot be started unless a desired bandwidth is secured in the transmission path determined by the path control protocol. Therefore, in order to perform communication with ensured communication quality, there is not only a network that can secure traffic, but also a network that can perform traffic placement control that allows the transmission route to be explicitly specified or changed independently of the route control protocol. desirable. Typical current techniques for realizing such a function include Openflow (for example, see Non-Patent Document 2) and MPLS (Multi Protocol Label Switching) (for example, see Non-Patent Document 3).

  In the following, there will be described problems in the case of trying to handle the traffic of the many-to-one communication system described in the background section above with these current technologies.

  First, Openflow is described. As shown in FIG. 2, Openflow is a technology that refers to a network system composed of an Openflow switch that transfers traffic data, an Openflow controller that controls the operation of multiple Openflow switches, and a control method that operates on the network system. .

  The Openflow switch internally has a flow table that is a collection of “flow entries”. A flow entry is a "flow" of received traffic (a set of communications having the same attributes determined by the combination of the contents of the Ethernet (registered trademark) frame header, the contents of various headers included in the IP packet, and the input physical port). Describes the contents of processing to be performed on the packet corresponding to the flow for each unit called. The processing contents include rewriting of various header values of a packet, addition and removal of a VLAN tag, transfer and discard of a packet to a designated physical port, and the like.

  When the Openflow switch receives a packet, it examines the input physical port of the packet and the contents of the various headers described above to identify the corresponding flow entry, and executes the processing contents described in the flow entry. The contents of the flow table are provided by the Openflow controller. In addition, the contents of the flow table of all Openflow switches in the network are basically centrally controlled by one Openflow controller.

  In this way, Openflow has a structure in which the control mechanism and the data transfer mechanism are separated, so that flexible control over the entire network can be performed centrally from the Openflow controller. Therefore, the above-described traffic arrangement control is also possible.

  However, since Openflow does not have a bandwidth guarantee function, communication quality can be ensured by setting a transmission path so as to avoid congestion or failure at the start of transmission, or in other traffic situations during transmission. Accordingly, it is necessary to perform the traffic placement control to change the transmission path accordingly.

  When trying to perform such traffic allocation control, Openflow does not have a technical element equivalent to a path such as MPLS LSP (Label Switch Path), which will be described later. Placement control cannot be performed, and an operation of rewriting the contents of the flow entry of the Openflow switch in the network is required. And since the Openflow switches that make up the network are equipment on the network service provider's side, and the Openflow controller that controls this is also the equipment on the network service provider's side, all traffic placement control by Openflow is network service providers. It will be realized by the equipment on the side.

  However, when the network service provider is different from the network users and accommodates a large number of users on the network, the network service provider controls the traffic arrangement while considering the communication quality of the individual traffic of the many users. Doing this increases the processing load on the network service provider's equipment due to an increase in the control frequency, and as a result, there is a risk that the performance of the entire network will be degraded. In particular, when dealing with traffic that tends to increase the number of flows, such as many-to-one communication, such a risk becomes significant. From this point of view, it is desirable that the user's individual traffic can be controlled by the user based on the user's own criteria. But Openflow can't do that.

  Next, MPLS (Multi Protocol Label Switching) will be described. MPLS is a technique for transmitting data on a path called an LSP based on the value of a label in an MPLS header newly assigned to the packet, not the destination address of the IP header.

  In MPLS, a routing protocol such as ISIS-TE (Traffic Engineering) or OSPF-TE, which is an extension of IS-IS or OSPF, can be used. In these routing protocols, information on the bandwidth of each link of the network can be advertised from the router. In MPLS, when a LSP (Label Switched Path) is established, a route is determined using this information, and signaling for LSP construction along the route is performed, so that the conventional IGP (Indicator Gateway Protocol) is dragged. Can be specified explicitly. With this function, traffic can be arranged along a route that satisfies a desired bandwidth requirement. In addition, the bandwidth for the LSP can be secured at the time of LSP construction signaling.

  The MPLS having such a function has the following problems when trying to transmit the traffic of the many-to-one type communication described in the background section while ensuring the communication quality.

  First, even if an attempt is made to secure a bandwidth in advance in order to ensure communication quality, a problem occurs in many-to-one type communication. When handling many-to-one communication traffic with MPLS, a method of constructing one LSP as a whole by merging multiple LSPs originating from multiple transmission sources in the middle (Fig. 3), and from the transmission source to the receiving base A method of handling traffic as a set of LSPs extending up to one (Fig. 4) can be considered. In the latter case, one LSP can be prepared for each server traffic, or the traffic of multiple transmission servers at the same site can be combined into one LSP, or the degree of subdivision of the LSP is free There is a degree.

  While the former method can suppress an increase in the number of LSPs, the required bandwidth differs for each section on the route, making it difficult to issue a bandwidth securing request. In the latter method, if the bit rate of traffic transmitted from one transmission server is known, the necessary bandwidth for each LSP can be easily calculated regardless of the degree of subdivision of the LSP. Since the necessary bandwidth is also constant regardless of the section, it is easy to issue a bandwidth securing request. However, the closer to the receiving base, the more LSPs handled by one MPLS router, and the more complicated the hardware resource management necessary to secure the bandwidth.

  As described above, it is not easy to secure a bandwidth when handling many-to-one type communication traffic in MPLS. Therefore, in order to ensure communication quality, traffic arrangement control such as changing the transmission path in order to avoid congestion during communication or performing load balancing among a plurality of paths is required.

  In MPLS, the operation of changing a part of a route of an LSP once constructed or constructing a new LSP while searching for a new route is heavy. The following control (1) and (2) is generally performed as the traffic relocation control during the communication as described above. The outline of each is shown in FIG. 5 and FIG.

  (1) Transfer traffic to another LSP with a different route.

  (2) Change the packet distribution rate in the load balance at the LSP entrance and branch points.

  The control of (1) is realized by changing the mapping between the IP packet and the LSP on the router at the edge part of the MPLS domain called LER (Label Edge Router). The control in (2) is the ratio of distributing IP packets to multiple LSPs on LER and the ratio of distributing labeled packets to multiple LSPs on a router called LSR (Label Switch Router) in the MPLS domain. It is realized by changing.

  In MPLS, not only LSR but also LER placed at the edge are facilities for network service providers. This means that MPLS-related communication protocols need to be able to communicate between the LER and the nearest LSR. If the LER is installed on the user base side, communication of the above communication protocol cannot be performed depending on the type of suspension line. Because.

  Accordingly, the control of either (1) or (2) is accompanied by a change in equipment setting of the network service provider, and the same problem as in Openflow occurs.

  The present invention has been made in view of the above points, and a network congestion avoidance system capable of avoiding congestion of content data distribution on an IP network without changing the setting of equipment of a network service provider, and It aims to provide a method.

According to one aspect, a network congestion avoidance system that avoids congestion at the time of content data distribution,
One or more transmission side terminals installed in the network service user base and transmitting content;
Installed in each viewer in the user base for the network service, the receiving terminal that receives the content,
One or more data transfer devices installed in the network service provider's site and transferring IP packets storing the content data;
Installed in the network service user base, connecting the transmitting terminal or the receiving terminal via a subnetwork, and further connecting the network service user base and the network service provider base A terminal accommodating device connected to the data transfer device via an underline that is a physical IP network,
A communication quality management device connected via the sub-network to the terminal accommodating device in a user base of any one of the network services;
A list of content names distributed and stored in the data transmission side terminal group, a list of transmission side terminals storing each content, and a data transmission bit when each transmission side terminal transmits data of each content A content database with rate information;
List of terminal accommodating devices for accommodating transmitting side terminals, list of transmitting side terminals, list of sub-networks used by terminal accommodating devices for accommodating transmitting side terminals to accommodate transmitting side terminals, transmitting side terminals and transmitting side terminals A network database having information on the correspondence between the terminal accommodating device that accommodates the sub-network, and
Have
The data transfer device
One or a plurality of virtual data transfer nodes (hereinafter referred to as “logical nodes”) are generated in the data transfer device, and the logical nodes on the other data transfer devices and their own logical nodes A network consisting of a set of logical nodes and logical links (hereinafter referred to as “virtual networks”) that connect logical nodes on a physical network to which one or more are connected. In a data transfer apparatus constructed by superimposing one or a plurality of terminals and physically connected directly to the terminal accommodating apparatus, the virtual network is provided between the logical node inside the terminal accommodating apparatus and the terminal accommodating apparatus. By constructing a logical link corresponding to the virtual network for each, a logical node, a logical link connecting the logical nodes, the logical node and the logical node Network composed of a set of logical links connecting end accommodating device (hereinafter, referred to as "logical network") is formed,
The terminal accommodating device that accommodates the receiving terminal is:
Measures the available bandwidth of the logical network based on a test packet for measuring the available bandwidth of the logical network between the own device and another terminal accommodating device sent from the terminal accommodating device that accommodates the transmitting terminal. And having usable bandwidth information notifying means for notifying the communication quality management device of usable bandwidth information,
The communication quality management device includes:
An available bandwidth measurement instruction means for sending an available bandwidth measurement instruction to a terminal accommodating device that accommodates each transmitting terminal;
A transmission-side terminal in each terminal accommodating device that accommodates the transmitting-side terminal so that the usable bandwidth information is received from a terminal accommodating device that accommodates the receiving- side terminal and the available bandwidth is equal to or greater than a predetermined margin value. An accommodation change instruction means for determining a logical network that accommodates each of the sub-networks that are connected to each other, and sending an accommodation instruction to change the accommodation to the determined logical network to each terminal accommodating apparatus that accommodates the transmitting side terminal; ,
Have
The terminal accommodating device that accommodates the transmitting terminal is:
An available bandwidth measuring means for transmitting the test packet based on the available bandwidth measurement instruction from the communication quality management device to a terminal accommodating device accommodating the receiving terminal;
Based on the receiving instruction from the communication quality management system, have a, and accommodating network changing means for dynamically changing the logical network which accommodates the respective sub-networks connecting the transmitting terminal,
The accommodation change instructing means of the communication quality management device is:
Based on the available bandwidth information acquired from the terminal accommodating device that accommodates the receiving terminal, the content database and the network database are referred to, a logical network that accommodates the transmitting terminal is determined, and the transmitting terminal is Provided is a congestion avoidance system including accommodation instruction means for transmitting an accommodation instruction to a logical network to a terminal accommodation apparatus to be accommodated .

  According to one aspect, traffic can be rearranged in many-to-one type communication by using a plurality of logical networks with different topologies prepared in advance only by operating the device at the user base and changing the settings. Therefore, as described in [Problems to be Solved by the Invention], the processing load of the equipment increases due to an increase in the frequency of control of the equipment of the network service provider, and the risk of degradation of the performance of the entire network caused by the processing load. This has the effect of solving the problems of existing technologies.

Explanatory drawing of the conventional many-to-one type communication system. The figure for demonstrating the outline | summary of Openflow which is a prior art. An example of realizing many-to-one communication by merging LSPs in conventional MPLS. An example of realizing many-to-one communication using point-to-point LSP in conventional MPLS. The figure for demonstrating the control which transfers traffic to other LSP in the prior art MPLS. The figure for demonstrating the change control of the packet allocation rate in MPLS of a prior art. The physical structural example of the network congestion avoidance system in one embodiment of this invention. The figure explaining the definition of the logical network in one embodiment of this invention. The figure explaining the method of designing a logical network for the purpose of congestion avoidance in one embodiment of this invention. The figure for demonstrating the measurement of an available band in one embodiment of this invention, and the switching operation | movement of a logical network based on it. The structural example of the network congestion avoidance system in one embodiment of this invention. The figure which shows the detail of the logical network constructed | assembled on the physical network part which consists of the data transfer apparatus in one embodiment of this invention, and a terminal accommodating apparatus. The figure for demonstrating the management system of the response | compatibility of a virtual network interface and a real network interface in one embodiment of this invention, and a response | compatibility (logical link) between virtual networks. The figure which described the example of the setting value about the detail of the physical network part which consists of the data transfer apparatus and terminal accommodating apparatus in one embodiment of this invention, and the logical network built on it. The figure for demonstrating the format of an IP packet and the transfer method in one embodiment of this invention. The figure for demonstrating the connection of the terminal accommodating apparatus and data transmission apparatus in one embodiment of this invention. The example of an internal structure of the terminal accommodating apparatus in one embodiment of this invention. The structural example of the accommodation network change part and route setting management part of the terminal accommodation apparatus in one embodiment of this invention. The structural example of the usable band measurement part of the terminal accommodating apparatus in one embodiment of this invention. The flowchart (the 1) of the procedure which the path | route information public information process on the terminal accommodating apparatus in one embodiment of this invention publicizes the path | route information regarding the VLAN circuit which has connected the terminal with respect to a logical network. The flowchart (the 2) of the procedure which the path | route information public information process on the terminal accommodating apparatus in one embodiment of this invention publicizes the path | route information regarding the VLAN circuit which has connected the terminal with respect to a logical network. The flowchart (the 1) of the procedure in which the terminal accommodating apparatus in one embodiment of this invention outputs the packet transmitted from the terminal to the logical network of the backbone side. The flowchart (the 2) of the procedure in which the terminal accommodating apparatus in one embodiment of this invention outputs the packet transmitted from the terminal to the logical network of the backbone side. The flowchart of the procedure which the terminal accommodating apparatus in one embodiment of this invention transfers the packet received from the logical network of the backbone side to a terminal. The flowchart of the procedure which builds the management logical network in one embodiment of this invention. The flowchart of the procedure of the content delivery process in one embodiment of this invention. The flowchart of the procedure which builds the logical network for data transmission in one embodiment of this invention. The flowchart of the content delivery process in one embodiment of this invention. The figure which showed the control signal transmitted / received between each apparatus before starting the content delivery in one embodiment of this invention. The list | wrist of the data provided to each control signal of FIG. 29 in one embodiment of this invention. The flowchart in which the communication quality management apparatus in one embodiment of this invention determines the logical network of the first accommodation destination of each transmission server. The communication quality management apparatus in one embodiment of this invention is a flowchart (the 1) of the procedure in which the logical network of the accommodation destination of each transmission server is changed based on the available bandwidth information of each logical network. The communication quality management apparatus in one embodiment of this invention is a flowchart (the 2) of the procedure in which the logical network of the accommodation destination of each transmission server is changed based on the available bandwidth information of each logical network. The figure which showed the control signal transmitted / received between each apparatus when completing the content delivery in one embodiment of this invention. The list | wrist of the data provided to each control signal of FIG. 34 in one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, the outline of the present invention will be described.

  FIG. 7 is a physical configuration example of a network congestion avoidance system according to an embodiment of the present invention.

The system shown in the figure has a function of superposing and building a plurality of different logical networks on a physical network, and terminal accommodating devices 120 and 210 of a plurality of user bases 100 _{1 to} 100 ₃ and 200 ₁ to 200 _3. Are connected to the network service provider side base 20 via the suspension lines 10 ₁ , 10 ₂ . In the figure, a user site ₁₀₀ 1 to 100 ₃ to the transmission side, illustrating the user site ₂₀₀ 1-200 ₃ as a receiver.

The terminal accommodating devices 120 and 220 installed in the network service user bases 100 and 200 are suspensions that are general physical IP networks that connect the user bases 100 and 200 and the network service provider side base 20. The lines 10 ₁ and 10 ₂ are connected to one or more data transfer apparatuses 300.

  The transmission base 100 and the reception base 200 accommodate a terminal 110, a reception terminal 210, and one or more sub-networks 101 to 104 and 201 to 203 used for connection of these terminals 110 and 210.

The terminal accommodating devices 120 and 220 are installed in the network service user bases 100 and 200, and are suspension lines that are physical IP networks that connect the user bases 100 and 200 and the network service provider side base 20. 10 ₁ and 10 ₂ are connected to one or more data transfer apparatuses 300. The terminal accommodating devices 120 and 220 accommodate the subnets 101 to 104 and 201 to 203 used for connection with the terminals 110 and 210 in the user bases 100 and 200, and are virtual networks configured on the data transfer device 300. A logical link corresponding to 1 to 1 is constructed on the suspension lines 10 ₁ , 10 ₂ between the data transfer device 300 based on the encapsulation technique using the IP header, and is connected to the virtual network and the virtual network. In addition, a logical network constituted by a set of logical links possessed by the terminal accommodating devices 120 and 220 and other terminal accommodating devices is formed. Based on an instruction from the communication quality management device 150 (to be described later), the terminal accommodating device 120 is configured so that the terminal 110 in the sub-network for each of the sub-networks 101 to 104 accommodating the terminal 110 in the user base. Between a function for dynamically changing a logical network that can transmit and receive data (a “accommodating network changing unit” to be described later) and its own terminal accommodating device 120 and another terminal accommodating device 220 in each logical network The available bandwidth value of the data transfer path is measured by a measurement method using a probe packet, generally called active measurement. Further, the probe packet transmission operation has a function (an “available bandwidth measuring unit” described later) that is performed at a time interval specified from a time specified by the communication quality management device 150 described later.

Communication quality control device of the user site 100 ₃ 150 is connected via a sub-network 105 as well as other terminals in the terminal accommodating apparatus 120 of one of the transmission points, one or more accommodating terminal 110 of the transmission points The function of instructing the transmission start time and transmission time interval of the probe packet for active measurement to the available bandwidth measuring unit in the terminal accommodating device 120 of the transmission base, and the usable bandwidth measuring unit of the terminal accommodating device 120 of each transmission base A function for receiving the value of the usable bandwidth for each logical network and a logical network that accommodates each sub-network connected to the terminal 110 accommodated by the terminal accommodating device 120 at each transmission base are determined as needed, and the determined logical network It has a function of issuing an instruction to change accommodation to the terminal accommodation device 120 at the transmission base.

Content database 130 of the user site 100 ₃ is connected via the other terminal and a sub-network 105 similarly to the terminal accommodating apparatus 120 of one of the transmission points, content that is distributed and stored in the terminal groups of the data transmission side A list of names, a list of terminals 110 on the data transmission side in which each content is stored, and information on a data transmission bit rate when each data transmission side terminal 110 transmits data of each content are stored. .

User site 100 ₃ of the network database 140, either in the terminal accommodating apparatus 120 of one transmission points are connected via a sub-network 105 as well as other terminals, the transmission points which accommodates each terminal 110 of the user site List of terminal accommodating devices 120, list of terminals, list of sub-networks used by terminal accommodating device 120 of transmission base for accommodating terminals 110, terminal accommodating device 120 of transmitting base accommodating each terminal 110 and the terminal 110 And the correspondence between sub-networks.

  The network service provider side base 20 has a plurality of data transfer devices 300.

  The data transfer apparatus 300 generates a plurality of virtual data transfer nodes (logical nodes) in addition to a packet transfer function on a physical network to which a plurality of data transfer apparatuses are connected. A function that builds a logical link that connects the upper logical node and its own internal logical node and superimposes multiple different virtual networks on the physical network to which it is connected, and its internal A function for constructing a logical link corresponding to each virtual network between the ethics node and the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base, which will be described later, based on the encapsulation technique using the IP header. Have.

  Each of the data transfer apparatuses 300 has a function of superposing and creating a number of logical networks that are independent virtual networks that do not interfere with each other on a physical network. The structure assumed in the present invention for this logical network is shown in FIG. One logical network is a logical node that logically connects a logical node that is a virtual data transfer node and logical nodes to each other, or a terminal accommodating device 120 at a transmission base, a terminal accommodating device 220 at a receiving base, and a logical node. It shall be composed of links.

  Referring to the example of FIG. 8, “logical network 1” in FIG. 8 is a logical data transfer node “logical node 1” to “logical node 4”, and logically connects these logical nodes. It is composed of “logical links 2 to 4” and “logical link 1” and “logical link 5” that logically connect the terminal accommodating device 120 of the transmission base, the terminal accommodating device 220 of the receiving base and the logical node. Yes. The “logical network 2” includes “logical nodes 5” to “logical nodes 8” which are virtual data transfer nodes, and “logical links 7” to “logical links 9” which logically connect these logical nodes. ”And“ logical link 6 ”and“ logical link 10 ”that logically connect the terminal accommodating device 120 at the transmission base, the terminal accommodating device 220 at the receiving base, and the logical node.

  One logical node is a virtual node generated as a software module on one data transfer apparatus 300, and has a data transfer function similar to that of a normal IP router. In the example of FIG. 8, “logical node 1” to “logical node 4” of “logical network 1” are respectively generated on “data transfer device 1” to “data transfer device 4” (FIG. 8). 8, each logical node and a data transfer device that generates them are connected by dotted lines), and “logical node 5” to “logical node 8” of “logical network 2” are “data transfer devices 5, 6,” respectively. 3 and 4 ".

  One logical link is generated on the physical line that physically connects the data transfer apparatuses 300 or between the data transfer apparatus 300 and the terminal accommodation apparatus 120 at the transmission base and the terminal accommodation apparatus 220 at the reception base. This is a virtual link. The logical link that connects the logical nodes is generated on any one physical line that connects the data transfer apparatuses 300 in which the logical nodes at both ends exist. In addition, the logical link that connects the terminal accommodation device 120 at the transmission base, the terminal accommodation device 220 at the reception base, and the logical node connects the terminal accommodation device 120 at the transmission base and the data transfer device 300 in which the logical node exists. (In the present invention, the physical line connecting the terminal accommodating apparatus 120 at the transmission base, the terminal accommodating apparatus 220 at the receiving base, and the data transfer apparatus 300 is referred to as an “underline”).

  In the example of FIG. 8, “logical link 1” of “logical network 1” is generated on “physical line 1” connecting “terminal accommodating apparatus 1” and “data transfer apparatus 1”. . Further, “logical link 2” is generated on “physical line 2” connected to “data transfer apparatus 2”. “Logical link 3” to “logical link 5” are generated on “physical line 3” to “physical line 5”, respectively. Similarly for “logical network 2”, “logical link 6” to “logical link 10” are generated on “physical lines 6, 7, 9, 4, 5”, respectively.

  Note that the data transfer apparatus 300 may be an existing diversion as long as it has a function for creating a logical network as described above. The present invention also relates to a specific implementation method of a logical network, that is, a logical node or a logical link. However, a specific method is not assumed (as an example that can be used as a data transfer apparatus in the present invention, for example, the system of Non-Patent Document 4 is cited). Further, the physical line realizing technology for connecting the data transfer apparatuses 300 may be any technique that can communicate IP packets, and does not assume any specific technique.

Special functions are assumed for the physical lines that connect the terminal accommodating apparatus 120 at the transmission base, the terminal accommodating apparatus 220 at the receiving base, and the data transfer apparatus 300, that is, the underlines 10 ₁ and 10 ₂ shown in FIG. The only requirement is that IP packets can be communicated.

  Then, the data transfer apparatus 300 constructs a logical link corresponding to the logical network on the underline 10 using a technique that can encapsulate a packet with an IP header. Encapsulate packets sent and received within each logical network using a technology such as GRE (see Non-Patent Document 5, for example) that has a key that identifies the flow of packets to be encapsulated. In doing so, a virtual link corresponding to the logical network is realized by using this key as an identifier of the logical network.

  In the present invention, in the network congestion avoidance system described above, a plurality of terminals on the transmission side in the many-to-one type communication described in the background section are appropriately divided and accommodated in a plurality of different logical networks. Further, according to the change in the usable bandwidth of the data transmission path from the transmission base 100 to the reception base 200 in each logical network, the setting change of the terminal accommodation device 120 of the transmission base installed in the user base changes each setting. Change the data transmission path by dynamically changing the destination server's destination logical network during data communication, avoiding the congestion of the network, that is, the logical network including the transmission path where the available bandwidth is insufficient Thus, the communication quality is ensured.

  A logical network design method for realizing the above will be described with reference to FIG.

  In many-to-one type communication, in order to enable communication while dynamically changing the logical network to which the transmission-side terminal belongs in accordance with the change in available bandwidth, for each transmission base 100, the terminal at the transmission base of that base A logical network is prepared so that there is at least one logical network having a path from the accommodation device 120 to the terminal accommodation device 220 at the reception base. In the example of FIG. 9, “transmission side terminal accommodating apparatus 1” is included in “logical network 1” and “logical network 2”, and “transmission side terminal accommodating apparatus 2” is “logical network 2” and “logical network 3”. And “transmission side terminal accommodating device 3” is included in “logical network 4” and “logical network 5”.

  In the example of FIG. 9, in any logical network, there is a route from the terminal accommodating device of all connected transmission bases to the terminal accommodating device of the receiving base. Data can be transmitted to the receiving base using any of the logical networks that have been used. Therefore, the “transmission-side terminal accommodating device 1” appropriately transfers the data transmitted from the transmission-side terminal group accommodated by itself to the “logical network 1” and the “logical network 2” according to the traffic situation. Communication can be performed while allocating. Similarly, “transmission side terminal accommodating device 2” can use “logical network 2” and “logical network 3”, and “transmission side terminal accommodating device 3” can use “logical network 4” and “logical network 5”. .

  In addition, as shown in “logical network 2” in FIG. 9, a terminal accommodating apparatus (here, “transmitting terminal accommodating apparatus 1” and “transmitting terminal accommodating apparatus 2”) at a plurality of transmission bases is connected to one logical network. Then, they may use the same logical network.

  Further, the topology of the logical network is a topology in which a candidate for a route from the terminal accommodating apparatus 120 at one transmission base connected to the logical network to the terminal accommodating apparatus 220 at the receiving base is uniquely determined in the logical network. There is no need, and a logical network having a topology in which a plurality of route candidates exist may be constructed. Even if the topology has a plurality of route candidates, when a normal routing protocol is operated in the logical network, only one route that is determined to be the optimum route is actually data. This is because it is used during transmission.

  In the example of FIG. 9, the “logical network 5” is a network having a plurality of routes from the terminal accommodating device 120 at the transmission base to the terminal accommodating device 220 at the receiving base, but actually operates within this logical network. Depending on the routing protocol being used, only one route is selected and used.

  Further, even if the logical network of the receiving terminal 110 is switched to another network, the actual data transmission path does not change, and the switching of the logical network is prevented from functioning as a traffic relocation means. Therefore, it is desirable to design so that the transmission path from the terminal accommodation apparatus at the transmission base to the terminal accommodation apparatus at the reception base is clearly different for each logical network. In the example of FIG. 9, for “logical network 1” and “logical network 2”, there is only one path from “transmission side terminal accommodating device 1” to “reception side terminal accommodating device”. It has a different path. For this reason, the “transmission-side terminal accommodating device 1” can reliably perform traffic relocation only by switching the accommodating logical network of the transmission server. The same can be said for “transmitting terminal accommodating apparatus 2” and “transmitting terminal accommodating apparatus 3”.

  The above is the description of the design method of the logical network used for data transmission.

  As for the logical network, apart from the one for data transmission, one management logical network for connecting all the terminal accommodating devices 120 of the transmission bases and the terminal accommodating device 220 of the receiving bases is constructed. This logical network is used for communication between a communication quality management device 150 (to be described later) and each terminal accommodating device 120, 220.

  Also, on each logical network, it is assumed that traffic other than the content distribution of the above-mentioned many-to-one type communication is often superposed together, so that it can be used for many-to-one type communication. It is assumed that the available bandwidth of the existing transmission path varies.

FIG. 10 is a diagram for explaining the measurement of the available bandwidth and the switching operation of the logical network based on it according to the embodiment of the present invention. Hereinafter, the terminals 110 of the user bases 100 _{1 to} 100 ₃ will be described as transmission servers that transmit content.

  In contrast to the logical network designed and constructed as described above, the logical network for data transfer used for data transmission using the available bandwidth measuring units 121 and 221 in the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base. Every 30th, the available bandwidth on the data transmission path from the terminal accommodating device 120 at the transmission base to the terminal accommodating device 220 at the receiving base in the data transfer logical network 30 is measured. Then, based on this measurement value, the data transfer logical network 30 that accommodates each transmission server 110 is determined at the start of communication, or the data transfer logical network 30 that accommodates the transmission server 110 during communication execution. Control to change.

  As a method for measuring the available bandwidth, a technique called active measurement (see, for example, Non-Patent Document 6) that does not require a device on the network service provider side or its operation is used. In this method, test packets called probe packets are sent to the network, and the available bandwidth of the transmission path through which the probe packet passes is estimated based on the quality status when they reach the receiving base 200.

  In the present invention, from the available bandwidth measuring unit 121 in the terminal accommodation device 120 of each transmission base, the terminal accommodation device 120 of the transmission base is connected, and all the logical networks 30 used for the content distribution are probed. Send the packet. Then, the available bandwidth measuring unit 221 in the terminal accommodating device 220 of the receiving base that received them notifies the communication quality management device 150 of the estimated available bandwidth value of each logical network 30 via the management logical network 70. .

  At this time, the probe packets sent from the terminal accommodating devices 120 at a plurality of transmission bases are concentrated on the inside of the network, particularly in a narrow part near the reception base, during the transmission, and the probe packets themselves press the network bandwidth. The operation of the available bandwidth measuring unit 121 of the terminal accommodating device 120 at each transmission base is synchronized so that probe packets are transmitted at intervals.

  The communication quality management apparatus 150 of the transmission base 100 that has received the above notification from the terminal accommodation apparatus 220 of the reception base, and the content transmitted from each transmission server 110 and the notified value of the usable bandwidth of each data transfer logical network 30. Based on the value of the data bit rate, the data transfer logical network 30 that accommodates each transmission server 110 is determined so that the available bandwidth of the transmission path of each logical network 30 remains over a predetermined margin value. Then, an accommodation control notification for accommodating each transmission server 110 in the desired data transfer logical network 30 is transmitted to the terminal accommodating device 120 at each transmission base. The above-described management logical network 70 is also used for this notification.

  Note that the communication quality management device 150 includes the bit rate of content data necessary for execution of these controls, the transmission server 110, the reception terminal 210 and the terminal accommodation device 120 at the transmission base, and the terminal accommodation device 220 at the reception base involved in content distribution. And the content database 130 and the network information of the identifiers of the sub-networks 101 to 104 used by the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base to accommodate the transmission server 110 and the receiving terminal 210. Collect from database 140.

  The terminal accommodation device 120 at each transmission base that has received the above accommodation control notification from the communication quality management device 150 is instructed to each transmission server 110 using the accommodation network changing unit 122 in the terminal accommodation device 120 at the transmission base. It is accommodated in the logical network 30 for data transfer. Specifically, the subnetwork and the logical link that associate the ID of the subnetwork connecting each transmission server with the ID of the logical link 40 corresponding to each data transfer logical network 30 to the accommodation network changing unit 122. The correspondence table table 123 is provided and the contents of the correspondence table table 123 are set based on the contents of the control notification, thereby managing the correspondence between the two.

  The terminal accommodating device 120 at the transmission base refers to the correspondence table 123 of the subnetwork and the logical link when transferring the data transmitted from each transmission server 110 to the data transfer device 300, and accommodates the transmission server 110. Data is transmitted to the logical link 40 corresponding to the previous logical network 30. As a result, data is transferred to the desired data transfer logical network 30 determined based on the available bandwidth measurement result, and communication that avoids congestion points on the network becomes possible.

  Below, it demonstrates in detail based on the physical structure of the network congestion avoidance system shown in FIG. In addition, the same code | symbol is attached | subjected to the same component as FIG.

Network congestion avoidance system shown in FIG. 11, the transmission server ₁₁₀ 1 to 110 _n, the receiving terminal 210 to receive the content, the terminal accommodation apparatus of the transmission points to accommodate the transmission server ₁₁₀ 1 _{~110 n 120} 1 _{~120 n} content, The receiving terminal 210 includes a terminal accommodating device 220 that accommodates the receiving terminal 210, data transfer devices 300 _{1 to} 300 ₈ , a communication quality management device 150, a content database 130, and a network database 140.

  The terminal accommodation device 120 at the transmission base has an available bandwidth measurement unit 121 and an accommodation network change unit 122 therein. In addition, it is assumed that the terminal accommodating device 220 at the reception base has the same configuration. These functions are realized by a computer executing a predetermined program.

  Further, the communication quality management device 150, the terminal accommodation device 120 at the transmission base, and the terminal accommodation device 220 at the reception base are NTP servers (not shown) installed outside or in the congestion avoidance system by NTP (Network Time Protocol). (The installation method of the NTP server and the connection method with the NTP server are not specified).

In the description of the present embodiment, an example is used in which _one piece of content data divided and stored in transmission servers 110 ₁ to 110 n distributed to N transmission bases is distributed to the reception terminals 210 of one reception base 200. . The divided and stored content data is transmitted in the form of IP packets over the network to which the data transfer device 300 is connected, and gathers at the receiving terminal 210 installed at one receiving base 200. UDP (User Datagram Protocol) is used for the transport layer above the IP layer, packet order alignment and transmission rate control are not performed by the transport layer, and the receiving terminal 210 receives IP packets from a plurality of transmission bases. Reach out of order. The receiving terminal 210 internally buffers these IP packets and arranges and reproduces them in an appropriate order that constitutes the content.

The data transfer device 300 is a device installed in the network service provider side base 20. All the components other than the data transfer device 300 are installed in the transmission base 100 or the reception base 200. All transmission bases 100 and reception bases 200 are bases on the network service user side. These user bases and physical networks (hereinafter referred to as backbone networks) formed by a set of data transfer apparatuses 300 are connected by suspension lines 10 ₁ and 10 ₂ . The undercarriage lines 10 ₁ and 10 ₂ may be any network that can communicate IP packets, and there are no other requirements.

The data transfer apparatus 300 includes a physical backbone network formed by its own set, and suspension lines 10 ₁ and 10 ₂ used for connection with the terminal accommodation apparatus 120 at the transmission base and the terminal accommodation apparatus 220 at the reception base. A plurality of data transfer logical networks 30 having the above-described structure are superimposed on the physical network, and a function of creating these as independent networks that do not interfere with each other is provided.

<Physical network>
FIG. 12 shows details of a resource for realizing a physical network part composed of a data transfer apparatus and a terminal accommodating apparatus and a logical network constructed thereon according to an embodiment of the present invention. Details of the physical network and the logical network 30 will be described below with reference to FIG.

  First, the configuration of the physical network will be described.

  The data transfer device 300, the terminal accommodation device 120 at the transmission base, and the terminal accommodation device 220 at the reception base have physical network interfaces, which are referred to as “real network interfaces”. These interfaces are given an IP address and a MAC (Media Access Control) address of the link layer, which are referred to as “physical IP address” and “physical Mac address”, respectively. All of these address values are unique without duplication.

The terminal accommodation device 120 at the transmission base and the terminal accommodation device 220 at the reception base installed at the network service user side are physically connected to the data transfer device 300 via the suspension lines 10 ₁ and 10 _2. . The undercarriage lines 10 ₁ and 10 ₂ are lines that can connect the classic on the user side and the network service provider side base 20 on which the data transfer apparatus 300 is placed, as long as IP packets can be communicated. Similarly, the physical network connecting the data transfer apparatuses 300 only needs to be able to communicate IP packets.

<Logical network>
Next, the data transfer logical network 30 constructed on the physical network will be described.

  As described above, the logical network 30 for data transfer includes a logical node that is a virtual data transfer node and logical nodes, or a terminal accommodating device 120 at a transmission base, a terminal accommodating device 220 at a receiving base, and a logical node. Are logically connected to each other. In the present embodiment, one logical node is specifically realized as a VM (Virtual Machine) on one data transfer apparatus 300. Also, in one logical network, only one logical node is generated at most in one data transfer apparatus 300.

  This logical node can transmit and receive IP packets to and from other logical nodes belonging to the same data transfer logical network 30, the terminal accommodating device 120 at the transmission base, and the terminal accommodating device 220 at the receiving base. (This function is realized as VM software). When the logical node, the terminal accommodating device 120 at the transmission base, and the terminal accommodating device 220 at the receiving base perform transmission / reception of packets on the logical network 30, a virtual network interface (hereinafter referred to as a virtual network interface; a general virtual machine). The packet can be transmitted / received via the network. This virtual network interface is assigned an IP address and a link layer Mac address, which are called a logical IP address and a logical Mac address, respectively. The values of these addresses need only be unique within the data transfer logical network 30, and there may be overlap between different logical networks.

  Then, as shown in FIG. 12, this virtual network interface is linked to one real network interface, and further includes other logical nodes belonging to the same logical network, terminal accommodation device 120 at the transmission base, and terminal accommodation at the reception base. One-to-one correspondence with another virtual network interface included in the device 220 (however, only virtual network interfaces associated with real network interfaces directly connected by a physical network can be associated with each other). ).

  Since all the virtual network interfaces are associated with one real network interface, the one-to-one correspondence between the real network interfaces and the real network interfaces is also in accordance with the one-to-one correspondence between the virtual network interfaces. As a result, the data transmission path on the physical network when data transmission is performed between them is also uniquely determined under the path control method operating on the physical network. That is, the logical link 40 in the logical network 30 of the present embodiment is a logical network formed by a one-to-one correspondence between the virtual network interfaces and a data transmission path between them defined by the virtual network interface. One virtual communication link.

  The correspondence between the virtual network interface and the real network interface and the correspondence between the virtual networks are shown in FIG. 13 in the data transfer device, the terminal accommodating device 120 at the transmission base, and the terminal accommodating device 220 at the receiving base. Manage by management table. The virtual network interface management table and the opposite real interface management table are held for each data transfer logical network 30 built in the apparatus. Only one real network interface management table is stored in the apparatus.

  (A), (b), (c) in the upper part of FIG. 13 is an example of a management table held in “data transfer device 1” in FIG. 12, and (d), (d) in the lower part of FIG. e) and (f) are examples of management tables held in “data transfer apparatus 2” in FIG. Note that the setting values of each item in the table are based on the example shown in FIG.

  The virtual network interface management table (FIGS. 12A and 12D) includes the logical nodes on “data transfer device 1” and “data transfer device 2”, “terminal accommodating device 1”, and “terminal accommodating device 2”. Information about the virtual network interface in the data transfer logical network 30 is held. Specifically, each virtual network interface is identified by an ID, the logical IP address and logical Mac address assigned to each interface, the logical link ID corresponding to the interface, and the own device to which the interface is linked Holds the real network interface ID.

  The real network interface management table (FIGS. 12B and 12E) holds information related to the real network interface of the data transfer apparatus 300. Specifically, each real network interface is identified by an ID, and a physical IP address and a physical Mac address assigned to each interface are held.

  In the example of FIG. 14, the “data transfer device 1” has two IDs with an ID “1” and a physical IP address “192.168.2.1” and an ID “2” and a physical IP address “192.168.3.1”. Has a real network interface. “Data transfer device 2” has two real network interfaces with ID “1” and physical IP address “192.168.4.2”, and ID “2” and physical IP address “192.168.5.2”. Have. These values are recorded in the actual network interface management tables of FIGS. 13B and 13E (the physical Mac address values are omitted).

  In the example of FIG. 14, the logical node belonging to “logical network 1” in “data transfer apparatus 1” is associated with the real network interface with ID “1” and the virtual network with ID “1”. The interface has a virtual network interface with IDs “2” and “3” linked to the real network interface with ID “2”. Also, the logical node belonging to “logical network 1” in “data transfer device 2” has a virtual network interface with ID “1” linked to the real network interface with ID “1”. . In the virtual network interface management tables of FIGS. 13A and 13D, the IDs of these virtual interfaces and the IDs of the real network interfaces to which the respective virtual interfaces are linked (the real network of the corresponding own device) (Interface ID column) is recorded (logical IP address and logical Mac address values are omitted). Thus, the correspondence between the virtual network interface and the real interface is managed by the virtual network interface management table.

  Similarly, as shown in FIG. 14, between “data transfer device 1” and “data transfer device 2”, virtual network interfaces having IDs “2” and “1” are associated with each other. An ID “0x5232000b” (GRE key) is assigned to the logical link consisting of The value of this GRE key is the entry of the virtual network interface ID “2” in the virtual network interface management table of “data transfer device 1” and the virtual network interface management table of “data transfer device 2” in FIG. The virtual network interface ID is recorded in the entry “1”.

  Each virtual network interface entry also describes the real network interface ID of the corresponding local device. Therefore, the “data transfer” associated with the logical link is recorded by recording the GRE key value (0x5232000b). The actual network interfaces of “device 1” and “data transfer device 2” are associated with each other. In addition, the values of the physical IP address and physical Mac address of those real network interfaces can be obtained from the real network interface management table.

  The opposite real network interface management tables in FIGS. 13C and 13F are the physical IP addresses of the real network interfaces of the opposite devices to which the opposite virtual network interfaces of the respective logical links in the tables are linked. When sending a packet handled on a logical network to a physical network, the packet is encapsulated with an IP header containing the physical network interface IP address and an L2 header containing the physical Mac address. Use when. Details will be described later.

  The determination of the values described in the real network interface management tables of FIGS. 13B and 13E and registration in the tables are performed when the system of this embodiment is physically constructed. Further, the determination of the values described in the virtual network interface management table and the opposite real network interface management table and the registration in the table are performed when the data transfer logical network 30 described later with reference to FIGS. 23 and 25 is constructed.

  The above is the description of the logical network.

<Packet transfer>
Packet transfer in the above logical network is performed in the same manner as an IP router in a normal physical network based on an IP routing protocol operated in the logical network by a logical node in the data transfer apparatus 300. Is called. That is, only packets whose destination logical Mac address is the logical Mac address of the interface or the broadcast logical Mac address of the packets arriving at their virtual network interface are received (the description of the operation in the case of multicast is here) (Omitted). If the logical IP address of the destination of the packet matches the logical IP address of the interface, the packet is transferred to the upper layer application software based on the port number in the IP header of the packet. If the logical IP address of the destination does not match the logical IP address of the interface, the logical node on another data transfer apparatus 300 that becomes the next hop with reference to the route table of the IP packet transfer constructed by the route control protocol The virtual network interface of the terminal accommodating device 120 at the transmission base or the terminal accommodating device 220 at the receiving base is specified, the logical Mac address of the packet destination is replaced with the logical Mac address of the virtual network interface of the next hop, and next hop The packet is transmitted from its own virtual network interface connected to the virtual network interface. It is assumed that the logical Mac address of the next hop virtual network interface is known by an ARP (Address Resolution Protocol) operation in the data transfer logical network 30. Details of this ARP operation will be described later.

  The packet is transferred in the logical network 30 as described above. However, the packet transmitted from the virtual network interface of the logical node or the terminal accommodating device 120 at the transmitting base and the terminal accommodating device 220 at the receiving base is actually virtual. The data is transmitted to the physical network through the real network interface associated with the network interface, and transferred on the physical network. At this time, a GRE header including a GRE key for identifying the logical link 40 in the logical network 30 is added to the packet on the data transfer logical network 30, and further, transfer with the transmission source on the physical network Encapsulation with the IP header including the physical IP address of the previous real network interface. Then, a link layer header (hereinafter referred to as L2 header) including the physical Mac address of the destination real network interface in the destination address is added, and packet transfer is performed on the physical network based on these values.

  FIG. 15 shows packet formats at various points from the content transmission server 110 to the reception terminal 210. Based on this figure, a method for transferring a packet in the data transfer logical network 30 or physical network will be described in detail.

  First, the application software on the transmission server 110 that transmits the content data is connected to the terminal accommodating apparatus 120 at the transmission base, as shown in (1) of FIG. 15, with the IP packet whose destination is the physical IP address of the reception terminal 210. To send on a subnetwork. The physical Mac address of the real interface connected to the subnetwork of the terminal accommodating apparatus 120 at the transmission base is inserted into the destination of the L2 header of this packet (this physical Mac address can be known by the ARP operation in the normal physical network). it can).

The terminal accommodation apparatus 120 at the transmission base of the transmission base 100 that has received the packet (1) in FIG. 15 uses the procedure shown in FIG. 21 to be described later, and the data stored in the transmission server 110 that has transmitted the packet at that time. The transfer logical network 30 is identified, and a GRE key representing the logical link 40 that can reach the virtual network interface of the data transfer apparatus 300 that becomes the next hop in the data transfer logical network 30 is obtained. Further, the logical Mac address of the virtual network interface opposite to the logical link 40, and the physical IP address of the real network interface corresponding to the virtual network interface at both ends of the logical link 40, that is, the terminal accommodating device of the own transmission base obtained 120 and the actual network interface underbody line 10 ₁ side, the physical IP address of the real network interface corresponding to the virtual network interface next hop.

  Upon obtaining these pieces of information, the terminal accommodating apparatus 120 at the transmission base adds the IP header and payload portion of the received packet (the packet received in (1), that is, the packet sent by the transmission server 110 to the “payload packet”. After the header is changed as shown in (2) of FIG. 15, the packet is transmitted to the virtual network interface of the next hop. The packet shown in (2) of FIG. 15 is the GRE key for identifying the L2 header that includes the logical Mac address of the virtual network interface that will be the next hop as the destination, and the logical link 40 that is used for the transfer. IP header including the physical network interface IP address (the physical network interface corresponding to the virtual network interface at both ends of the logical link 40) And an L2 header including the physical Mac address of the real network interface of the data transfer apparatus 300 that is the packet transfer destination as the destination, in order from the inside to the outside. That is, the received original packet added with the L2 header and GRE header used in the data transfer logical network 30 is encapsulated with the IP header including the physical IP address of the actual network interface.

  The packet transmitted from the terminal accommodating device 120 at the transmission base is transferred on the sub-network 101 based on the physical Mac address that is the destination of the outermost L2 header, and reaches the data transfer device 300 as the transfer destination.

  The data transfer apparatus 300 that has received this packet refers to the real network interface management table (FIG. 13) held by itself, and the destination of the outermost L2 header of the received packet is the address of the real network interface to which the packet has arrived. If it matches the physical Mac address, it refers to the destination address in the outermost IP header. Then, referring to the real network interface management table and confirming that this destination address is the physical IP address of its own real network interface, it identifies the internal data transfer logical network 30 that is the packet transfer destination. Then, refer to the value of the GRE key in the GRE header of the packet.

  Then, referring to the virtual network interface management table (FIG. 13) held by itself, the virtual network interface ID corresponding to the value of the GRE key is specified, and the virtual network interface is designated as the internal data transfer logic of the packet. It is determined as a transfer destination of the network 30.

  When the virtual network interface to which the packet is transferred is determined, the data transfer device 300 transfers the packet in which the IP header and GRE header used for encapsulation are deleted from the packet to the virtual network interface.

  The logical node that has received the packet via this virtual network interface refers to the virtual network management table, and the destination Mac address of the L2 header in the data transfer logical network 30 is the logical network interface logic that the packet has arrived at. If it matches with the Mac address, it is determined that processing at the IP layer is necessary, and the original packet that has been encapsulated, that is, the destination IP address of the payload packet is referred to. In the case of this description, since the destination IP address is the physical IP address of the receiving terminal 210, it is determined that the packet needs to be transferred to either the other data transfer device or the terminal accommodating device 220 at the receiving base. Search the routing table for IP packet transfer in the network (how to create the routing table contents will be described later). Then, the next hop virtual network interface (the most recent data transfer device or the virtual network interface of the terminal accommodating device 220 at the receiving base) serving as the packet transfer destination, and its own virtual network interface serving as the packet transmission source Identify.

  As described above, the physical IP of the terminal is used to identify the terminal outside the terminal accommodating device 120 at the transmission base (user base side) and the terminal outside the terminal accommodating device 220 at the receiving base (user base side). Use an address. As described above, in the logical network 30, the packet is forwarded to the destination based on this physical IP address (IP address information for reaching the terminal, that is, the IP of the user base VLAN that accommodates the terminal) The network address is advertised from the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base into each data transfer logical network 30 by the procedure of Fig. 20 described later. Packet transfer based on the physical IP address of the terminal is possible in the network 30). Therefore, it is necessary that the IP network address of the VLAN of the user base does not overlap with the IP network address in all the data transfer logical networks 30.

  When the logical node that has searched the routing table identifies the above information, next, for the payload packet, a new L2 header in the logical network 30 for data transfer, a GRE header, an IP header for encapsulation, and Then, a physical network L2 header is added and a packet to be transmitted to the next hop device is created. Information necessary for creating these headers is obtained as follows.

  First, the value of the destination Mac address of the L2 header in the data transfer logical network 30 is obtained through ARP processing in the data transfer logical network 30. Specifically, the ARP process, which is a software process operating on the logical node, performs the logical IP address of the virtual network interface of the next hop in the data transfer logical network 30 (this address is obtained from the IP packet transfer route table). An ARP query packet for knowing the logical Mac address corresponding to is broadcast to the subnetwork in the data transfer logical network 30. That is, a packet in which the destination IP address is the IP broadcast address in the subnetwork and the destination of the L2 header in the data transfer logical network 30 is the broadcast address is transmitted. When this packet is sent to a physical network, a GRE header including a GRE key for logical link identification, an IP header for encapsulation including the IP broadcast address of the subnetwork in the physical network as a destination IP address, Then, an L2 header including a broadcast address is added to the destination Mac address. The ARP process of the logical node of the data transfer device that has the virtual network interface to which the logical IP address that is the target of this ARP query is given returns the logical Mac address of the virtual network interface to the ARP query source. ARP reply packet is sent back. The sender of the original ARP query knows the value of the logical Mac address of the virtual network interface of the next hop in the data transfer logical network 30 by this ARP reply.

  The value of the GRE key included in the GRE header is an ID for identifying the logical link 40 that is the packet transmission destination, that is, the logical link that is obtained from the routing table and is associated with its own virtual network interface that is the packet transmission source. 40 IDs. Therefore, this value can be known as the value of the GRE key in the entry of the virtual network interface ID in the virtual network interface management table.

  The real network interface ID in this entry is an ID for identifying the real network interface associated with the virtual network interface that is the source of the packet, and the physical IP address and physical Mac address of this real network interface are respectively indicated. It is necessary to store the IP header for encapsulation and the source address of the L2 header in the physical network. These values are obtained from the actual network interface management table (FIGS. 13B and 13E).

  In addition, as for the value of the IP header for encapsulation and the destination address value in the L2 header in the physical network, the physical IP address and physical Mac address of the real network interface of the opposite device ahead of the logical link 40 that is the packet transmission destination Must be stored. These addresses can be obtained from the entry including the above GRE key with reference to the opposite real network interface management table (FIGS. 13C and 13F).

  A new transmission packet created by giving various headers through the above processing is sent out from its own virtual network interface as a transmission source. This new transmission packet is as shown in (3) in FIG. 15 when the transfer destination device corresponding to the next hop is the data transfer device 300, and in the case of the terminal accommodating device 220 at the receiving base in FIG. It becomes like (4).

  The payload packet sent from the transmission server 110 is transferred between the data transfer devices 300 as described above, and eventually reaches the terminal accommodating device 220 at the receiving base. The terminal accommodating apparatus 220 at the receiving base that has received the packet performs the processing shown in FIG. 22 (details will be described later), and “transfer from the terminal accommodating apparatus at the receiving base to the receiving terminal” in (5) in FIG. “Packet” is transferred to the receiving terminal 210. As described above, the payload of the “transfer packet from the transmission server to the transmission side terminal accommodating device” in FIG. 15 (1) sent from the transmission server 110 reaches the reception terminal 210. The above is the description of the packet format in various places from the content transmission server to the reception terminal 210 and the mechanism of packet transfer in the data transfer logical network 30 and physical network based on FIG.

<Route table construction method>
Next, a method for constructing a routing table for transferring IP packets in the data transfer logical network 30 will be described. For this, a routing protocol is used in the same way as in the case of a normal IP network. In the data transfer device 300, the route control protocol is executed by a route information exchange process, which is a software process in the VM constituting the logical node, and in the terminal accommodation device 120 at the transmission base and the terminal accommodation device 220 at the reception base. It is assumed that it is performed by a route information publicizing process (described later) existing for each data transfer logical network 30 to which it is connected. A route based on the address system assigned to the virtual network interface by the execution subject of these route control protocols of the terminal accommodation device 120 at the transmission base, the terminal accommodation device 220 at the reception base, and the data transfer device 300 in the data transfer logical network 30. By exchanging information (subnet address information), a communication path between the transmission server 110 and the reception terminal 210 between the user bases is established.

  Note that the subnet address assigned to the subnetwork to which the user terminal (transmission server, receiving terminal) is connected is the logical IP of the logical link 40 in all the data transfer logical networks 30 that can accommodate the subnetwork. It is assumed that the subnet address corresponding to the address and the subnet address corresponding to the physical IP address of the physical link of the backbone network are assigned in a form that does not overlap. In addition, the subnet address corresponding to the logical IP address of the logical link 40 in all the data transfer logical networks 30 is given in a form that does not overlap with the subnet address corresponding to the physical IP address of the physical link of the backbone network. Shall. However, the subnet address in the data transfer logical network 30 may overlap between the data transfer logical networks 30.

  The logical node route information public information processes exchange route information in the same way as is done on a normal IP network. The route information publicity process (details will be described later in FIG. 18) of the terminal accommodation device 120 at the transmission base and the terminal accommodation device 220 at the reception base is performed by the transmission server 110 at the user base connected to the terminal accommodation device at its own transmission base. The subnet address of the subnetwork accommodated in the data transfer logical network 30 to which the receiving terminal 210 belongs is advertised in the data transfer logical network 30. As will be described later, in this embodiment, for simplicity, the terminal accommodation device 120 at the transmission base and the terminal accommodation device 220 at the reception base have only one logical link 40 connected to one logical network 30 for data transfer. Therefore, the route information publicizing process of the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base need not receive the route information broadcast from the route information transferring process on the data transfer device 300. . Details of the operation of the route information publicizing process of the terminal accommodation device 120 at the transmission base and the terminal accommodation device 220 at the reception base will be described later. This completes the description of the routing table construction method for forwarding IP packets within the data transfer logical network 30.

<Configuration processing of terminal accommodating device>
Next, the configuration of the terminal accommodating device 120 (220) will be described. The terminal accommodation device 120 at the transmission base is a device installed in the base of the user of the network, and dynamically changes the logical network 30 for data transfer on the backbone network side that accommodates the subnetwork on the user base side. It is a device that realizes the function. In this embodiment, an L2 VLAN (hereinafter referred to as VLAN) is accommodated as a sub-network on the user base side, and one terminal is connected to each VLAN. In the following description, the subnetwork on the user base side is referred to as a VLAN line.

  Regarding the physical connection method with the data transfer device 300, it is possible to connect the terminal accommodating device 120 (220) of one transmission base to a plurality of data transfer devices 300, or to connect the terminal accommodating device of a plurality of transmission bases. 120 (the terminal accommodating device 220 at the reception base) can be accommodated in one data transfer device 300.

  In connection with the connection with the data transfer logical network 30, the logical links 40 corresponding to the respective data transfer logical networks 30 are constructed on the physical links between the data transfer apparatuses 300. For simplicity, the terminal accommodating device 120 (220) at one transmission base is designed to have only one logical link connected to one data transfer logical network 30.

  FIG. 16 shows a physical connection between the terminal accommodating device 120 (220) and the data transfer device 300, and a configuration example of the data transfer logical network 30. As in this example, it is possible to physically connect one “terminal accommodating apparatus 1” to a plurality of “data transfer apparatuses 1” and “data transfer apparatuses 2”, or one “data transfer apparatus 2”. A plurality of “terminal accommodating apparatus 1” and “terminal accommodating apparatus 2” can be physically connected to “.” However, each of “terminal accommodating apparatus 1” and “terminal accommodating apparatus 2” has only one logical link connected to one logical network. For example, “terminal accommodating device 1” has a logical link that is an entrance to “logical network 1” built on a physical link that connects to “data transfer device 1”. Such a logical link is the same physical link. It is assumed that the above two are not constructed or constructed on a physical link connected to “data transfer apparatus 2”.

  FIG. 17 shows an internal configuration of the terminal accommodating device 120 (220) in the present embodiment. The terminal accommodating apparatus 120 (220) is connected to a terminal in the user base via a plurality of VLAN lines, and is connected to the data transfer apparatus 300 via a plurality of backbone lines. The terminal accommodating device 120 (220) includes a VLAN interface 160, a backbone interface 165, a data switch unit 123, an available bandwidth measuring unit 124, an accommodated network changing unit 125, and a route setting managing unit 126 therein. Hereinafter, the terminal accommodation device 120 at the transmission base will be described, but the same applies to the terminal accommodation device 220 at the reception base.

  The VLAN interface 160 terminates the VLAN line 170 used for connection with the transmission server 110 in the user base, and performs input / output processing of packets via the VLAN line 170.

  The backbone interface 165 terminates the backbone line 180 used for connection with the data transfer apparatus 300 and performs input / output processing of packets via the backbone line 180.

  The data switch unit 123 plays a role of a switch in packet transfer between the VLAN interface 160 and the backbone interface 165, whereby the packet is transferred to an appropriate transfer destination as needed.

  The usable bandwidth measuring unit 124 and the accommodation network changing unit 125 indicate the above-described usable atmosphere measurement function and accommodation network changing function, and are both part of the means that characterize the present invention. The route setting management unit 125 sets and manages route information necessary for communication between bases via the data transfer logical network 30 on the backbone network, that is, IP routing. The packet is transferred on the data switch unit 123 and the data signal line 127 shown in FIG. Separately, the common control signal line 128 described in FIG. 17 is used for communication between the respective units associated with various controls.

  The configuration of the route setting management unit 125 and the accommodated network changing unit 126 is shown in FIG. 18, and the configuration of the available bandwidth measuring unit 124 is shown in FIG. The available bandwidth measuring unit 124 includes therein a time information providing unit 310 that can generate time information synchronized with an NTP server installed outside or in the system of the present embodiment. The entity of the time information providing unit 310 may be a time management system of a general computer system synchronized with an NTP server. The available bandwidth measurement process 320 inside the available bandwidth measuring unit 124 uses the time information provided by the time information providing unit 310 to perform an operation based on the time synchronized with the NTP server.

  Below, the detail of operation | movement of the terminal accommodation apparatus 120 of the transmission base which has said structure is demonstrated. First, a method for setting route information necessary for communication between bases via the data transfer logical network 30 will be described with reference to FIG.

  The route setting management unit 125 advertises the route information of the IP network address assigned to the VLAN accommodated by the terminal accommodating device 120 of the transmission base to the data transfer logical network 30 accommodating each VLAN. The route information publicizing process 511, the ARP process 513 for executing the MAC address resolution in the data transfer logical network 30, and the logical IP address of the virtual network interface of the next hop data transfer device in the data transfer logical network 30 Each data transfer logical network 30 has an ARP table 512 that holds pairs of logical Mac addresses.

  The logical IP address of next hop in the ARP table 512 is determined in S308 in FIG. 25 and step S508 in FIG. 27, which will be described later, and the value is written into the ARP table 512 by the user of the network service. The ARP process 513 starts ARP processing for resolving the logical Mac address of the virtual network interface when the value of the logical IP address is written in the ARP table 512. Specifically, similarly to the ARP process on the logical node of the data transfer apparatus 300, the ARP query packet is broadcast to the subnetwork in the data transfer logical network 30 (this broadcast address is the logical IP address). The desired logical Mac address information is obtained from the ARP reply returned from the data transfer device 300 in the subnet, and written in the ARP table 512.

  The accommodated network changing unit 126 has six managements: a VLAN management table 410, a backbone interface management table 420, an opposed real network interface management table 430, an accommodated network management table 440, a logical network connection management table 450, and a virtual network interface management table 460. Has a table.

  The VLAN management table 410 is assigned to the VLAN interface ID, which is the identifier of each VLAN line 170 of the terminal accommodating apparatus 120 at the transmission base, the ID of the L2 VLAN used on the VLAN line 170, and the VLAN. Keep the IP network address (subnetwork address). These values all correspond one to one. These values are all values that are determined when the network congestion avoiding system is physically constructed, and values are written to the table by the network service user.

  The backbone interface management table 420 corresponds to the real network interface management table (FIG. 15) in the data transfer device 300, and the backbone interface ID that is the identifier of each backbone interface 165 of the terminal accommodating device 120 at the transmission base, and the backbone. The physical IP address and physical Mac address assigned to the interface 165 are held. These values all correspond one to one. These values are all values that are determined when the network congestion avoiding system is physically constructed, and the network service user writes the values in this table.

  The opposite real network interface management table 430 is the same as that of the data transfer device 300 (FIG. 15), and the data transfer device 300 in each data transfer logical network 30 to which the terminal accommodating device 120 of the transmission base is connected. Of the real network interface of the data transfer apparatus 300 to which the value of the GRE key that is the identifier of the logical link 40 used for connection to the virtual link interface and the opposite virtual network interface ahead of the logical link 40 is linked. Holds the physical IP address. These values can be identified from the information regarding the logical network 30 for data transfer provided at the time of notification in step S304 in FIG. 25 and step S504 in FIG. Performed by network service users.

  The accommodated network management table 440 is a list in which logical network IDs that are identifiers of logical networks accommodating the VLAN lines are arranged for each VLAN line on the user base side identified by the VLAN ID in the VLAN management table 410. Hold. The reason for holding the list of logical network IDs is that in the present embodiment, it is assumed that one VLAN is accommodated not only in one logical network but also in a plurality of logical networks. The ID value associated with each data transfer logical network 30 is determined and registered in the network database when the construction process of the data transfer logical network 30 shown in FIGS. 25 and 27 described later is performed. Use things. The contents of the accommodation network management table 440 are set by the accommodation network changing unit 126 in the terminal accommodation apparatus 120 at the transmission base based on an instruction from the communication quality management apparatus in the middle of transmission of content data or during transmission.

  The logical network connection management table 450 is the data transfer logical network 30 corresponding to each logical network ID in the accommodation network management table 440, and the terminal accommodation apparatus 120 at the transmission base in the data transfer logical network 30. And a GRE key which is an identifier of the logical link 40 used for connection between the data transfer apparatus 300 and the data transfer apparatus 300. As described above, since the terminal accommodating apparatus at one transmission base has only one logical link connected to one data transfer logical network 30, the logical network ID and GRE key values in the table are 1. Corresponds to pair 1. The association between each logical link 40 and the value of the GRE key is determined at the time of constructing the data transfer logical network 30 in step S303 in FIG. 25 and step S503 in FIG. The specific method for issuing the key value is not defined in this specification), and the correspondence is the logic for data transfer provided at the time of the notification in step S304 in FIG. 25 and step S504 in FIG. It is assumed that the information regarding the network 30 can be extracted (the specific contents of the information regarding the logical network provided in the notification are not specified in this specification). The contents of the logical network connection management table are performed by the network service user based on this notification.

  The virtual network interface management table 460 is the same as that of the data transfer apparatus 300 (FIG. 15). A table that holds information related to the virtual network interface associated with its own backbone interface 165 that the terminal accommodating device 120 of the transmission base has, an ID that identifies the virtual network interface, and a logical IP that is assigned to the virtual network interface An address, a logical Mac address, a GRE key for identifying the logical link 40 associated with the virtual network interface, and a backbone interface ID of the own apparatus associated with the virtual network interface are stored. These values all correspond one to one. The value of the GRE key and the ID of the backbone interface associated with the virtual network interface are related to the data transfer logical network 30 provided at the time of the notification in step S304 of FIG. 25 and step S504 of FIG. It can be identified from the information.

  Further, the values of the logical IP address and the logical Mac address are determined in step 308 in FIG. 25 and step 508 in FIG. 27, and writing to the table is performed by the network service provider.

  The above is the description of the configuration of the terminal accommodating device.

<Route information PR process>
Next, the route information publicizing process 511 on the terminal accommodation device 120 at the transmission base accommodates the route information of the IP network address assigned to the VLAN accommodated by the terminal accommodation device 120 at the transmission base. A method for publicizing the data transfer logical network 30 will be described.

  The charts shown in FIG. 20 and FIG. 21 show the procedure of the operation in which one route information public information process 511 publicizes the route information in the corresponding data transfer logical network 30. All the VLANs accommodated in the terminal accommodation device 120 at the transmission base are obtained by the route information publicizing process 511 responsible for route exchange in each data transfer logical network 30 performing the operations of FIGS. 20 and 21 independently. Publicity of routing information on IP network addresses is achieved.

  The procedure shown in the charts of FIGS. 20 and 21 will be described.

  In step S1, a variable indicating an entry in the accommodated network management table 440 of the accommodated network changing unit 126 to be referred to in the subsequent processing is initialized to 1.

  In step S2, whether or not the VLAN circuit 70 represented by the VLAN ID of each entry of the accommodated network management table 440 is accommodated in its own data transfer logical network 30, that is, the IP network assigned to the VLAN ID. It is determined whether or not the route information related to the address needs to be advertised in the own data transfer logical network 30. For this determination, it is checked whether or not its own logical network ID is included in the logical network ID list of the entry being referred to. If it is included, it is necessary to publicize it, so the processing for that is executed after step S3. If it is not included, there is no need to publicize it, so it is confirmed whether or not the processing has been completed up to the last entry in the accommodation network management table 440 (step S13), and if it is completed, the processing of this chart is terminated. If not completed, the value of the variable indicating the entry in the accommodation network management table is increased by 1 (step S14), and the process returns to step S2.

  Steps S3 and S4 are processes for determining an IP network address that is assigned to the VLAN and needs to be advertised as route information. This is achieved by finding an entry including the VLAN ID value of the entry referenced in step S2 from the VLAN management table 410 and obtaining the IP network address described in the entry.

  In step S5, an entry having its own logical network ID is searched from the logical network connection management table 450, and a corresponding GRE key, that is, a logical link 40 used for transmission of route information is determined.

  Steps S6 to S12 are processes for sending a packet of route information to the logical link 40 determined in step S5.

  In step S6, the entry of the virtual network interface management table 460 having the GRE key obtained in step S5 is specified, and the logical IP address of the virtual network interface corresponding to the logical link 40 represented by the value of the GRE key is determined from the entry. And the logical Mac address and the backbone interface ID representing the backbone line on which the logical link is superimposed.

  In step S7, the route setting management unit 125 refers to the ARP table 512 of its own logical network resource to obtain the next hop logical IP address and logical Mac address. If this logical Mac address is not written in the ARP table, it waits for setting by the ARP process.

  In step S8, an entry in the backbone interface management table 420 having the backbone interface ID obtained in step S6 is specified, and a physical IP address and a physical ac address are obtained from the entry.

  In step S9, the entry of the opposite real network interface management table 430 having the GRE key obtained in step S5 is specified, and the physical IP address is obtained from the entry.

  In step S10, an IP packet used in the data transfer logical network 30 for publicizing the route information is generated using the information obtained in the previous steps. The payload of this packet is the IP network address of the VLAN circuit to be publicized determined in step S4. Then, an IP header including the logical IP address obtained in step S6 as the transmission source IP address and the logical IP address including the logical IP address in step S7 is assigned to this payload as the destination IP address. Further, the L2 header in the data transfer logical network 30 including the logical Mac address obtained in step S6 as the source Mac address and the logical Mac address obtained in step S7 as the destination Mac address is assigned.

  In step S11, a header necessary for transmission on the physical network is added to the IP packet generated in step S10. Specifically, the GRE header including the GRE key obtained in step S5, the physical IP address obtained in step S8 as the source IP address, and the physical IP address obtained in step S9 as the destination IP address are included. The physical Mac of the real network interface of the opposite device that contains the IP header for encapsulation, the physical Mac address obtained in step S8 as the source Mac address, and the physical IP address obtained in step S9 as the destination Mac address An L2 header in the physical network including the address (this Mac address is obtained by ARP in the physical network) is added.

  In step S12, the packet generated in step S11 is transmitted to the backbone line 180 corresponding to the backbone interface ID obtained in step S6. Then, the process proceeds to step S13.

  As described above, the route information PR process 511 of the terminal accommodation device 120 of the transmission base accommodates the route information of the IP network address assigned to the VLAN accommodated by the terminal accommodation device 120 of the transmission base. Is a method of publicizing the data transfer logical network 30.

  As described above, in the present embodiment, for simplicity, the terminal accommodation apparatus 120 at the transmission base is designed to have only one logical link connected to one logical network. The route information public information process 511 does not need to receive the route information publicized from the route information public information process on the data transfer apparatus 300.

<Transmission and reception of packets for data communication>
Next, a method for transmitting / receiving normal data communication packets to / from the data transfer logical network 30 by the terminal accommodation apparatus 120 at the transmission base will be described. First, a procedure for inputting a packet transmitted by the transmission server 110 accommodated in the terminal accommodating device 120 at the transmission base from the VLAN line 170 to which the transmission server 110 is connected and outputting the packet to the data network 30 on the backbone side. This will be described with reference to the charts of FIGS. 22 and 23, the input packet processing unit 162 of each VLAN interface 160 of the terminal accommodating apparatus 120 at the transmission base receives one packet from the VLAN line and outputs it to the backbone line 180. The procedure of is shown. The input packet processing unit 162 on each VLAN interface 160 independently executes the processing shown in the chart of this drawing every time a packet is received.

  Step S101 is initialization of variables used in this chart.

  When a packet having its own Mac address as the destination address of the L2 header arrives at its own VLAN interface 160 (step S102), the terminal accommodating apparatus 120 at the transmission base enters an entry having its own VLAN interface ID from the VLAN management table 410. And the VLAN ID is confirmed (step S103). Next, using the accommodated network management table 440, a list of IDs of the logical networks accommodating the VLAN at that time is obtained (step S104).

  It is assumed that the terminal accommodating apparatus 120 at the transmission base accommodates one VLAN in a plurality of logical networks at the same time. The reason is that when performing many-to-one communication in the background section above, the VLAN accommodating the receiving terminal 210 is accommodated in a plurality of logical networks for data transfer at the same time via different logical networks. This is because it is necessary to be able to receive a packet that arrives. In this many-to-one type communication, there is no need to accommodate the transmission server 110 in a plurality of logical networks at the same time. However, in order to make the explanation general, there are a plurality of logical networks that accommodate the VLAN. A description will be given assuming the case.

  Steps S105 to S109 are processes for transmitting the packet to the i-th logical network among the logical networks obtained in step S104.

  In step S105, the entry having the i-th logical network ID is specified from the logical network connection management table 450, and the corresponding GRE key, that is, the logical link 40 used for packet transmission is determined.

  In step S106, the entry of the virtual network interface management table 460 having the GRE key obtained in step S105 is specified, and the logical IP address of the virtual network interface corresponding to the logical link 40 represented by the value of the GRE key is determined from the entry. The logical Mac address and the backbone interface ID representing the backbone line on which the logical link is superimposed are obtained.

  In step S107, the ARP table 512 of the logical network resource currently being processed by the path setting management unit 125 is referenced to obtain the next hop logical IP address and logical Mac address. If this logical Mac address is not written in the ARP table 512, the setting by the ARP process 513 is awaited.

  In step S108, an entry in the backbone interface management table 420 having the backbone interface ID obtained in step S106 is specified, and a physical IP address and a physical Mac address are obtained from the entry.

  In step S109, the entry of the opposite real network interface management table 430 having the GRE key obtained in step S105 is specified, and the physical IP address is obtained from the entry.

  In step S110, using the information obtained in the previous steps, an L2 header used in the data transfer logical network 30 is added to the packet received from the VLAN line 170 in step S102. The logical Mac address obtained in step S106 is inserted into the source Mac address of this header, and the logical Mac address obtained in step S107 is inserted into the destination Mac address of the header.

  In step S111, a header necessary for transmission on the physical network is added to the IP packet generated in step S110. Specifically, the GRE header including the GRE key obtained in step S105, the physical IP address obtained in step S108 as the source IP address, and the physical IP address obtained in step S109 as the destination IP address are included. Physical Mac interface physical network of the opposite device that has the IP header for encapsulation and the physical Mac address obtained in step S108 as the source Mac address, and further has the physical IP address obtained in step S109 as the destination Mac address An L2 header in the physical network including the address (this Mac address is obtained by ARP in the physical network) is added.

  In step S112, the packet is transmitted to the backbone line specified in step S106.

  In step S113, the value of the variable is incremented by 1. In step S114, it is determined whether or not the transmission of the packet to all the data transfer logical networks 30 constituting the list obtained in step S104 is completed. If it has been completed, the processing of the chart shown in this figure is terminated. If it has not been completed, the processing returns to step S105.

  The above is the description of the procedure for outputting the packet sent from the terminal to the data transfer logical network 30 on the backbone side.

<Packet transfer from terminal accommodating device to terminal on VLAN line>
Next, a procedure for transferring a packet input from the data transfer logical network 30 on the backbone side by the terminal accommodating apparatus 120 at the transmission base to the terminal on the VLAN line 170 as the destination will be described with reference to the chart of FIG. The chart of FIG. 24 shows a procedure from when the input packet processing unit 171 of each backbone interface 165 of the terminal accommodating apparatus 120 at the transmission base receives one packet from the backbone line 180 to output it to the VLAN line 170. Is shown. The input packet processing unit 171 on each backbone interface 165 independently executes the processing shown in the chart of this drawing every time a packet is received.

  First, when a packet having its own Mac address as the destination address of the L2 header arrives at its own backbone interface 165 (step S201), the value of the GRE key in the GRE header of the packet is examined (step S202). Then, an entry including the value of the GRE key is searched from the logical network connection management table 450 of the accommodated network changing unit 126, and the logical network ID of the entry is obtained (step S203). Thereby, it is specified from which logical network the packet has arrived.

  In step S204 and step S205, the destination IP address in the IP header portion of the payload packet of the packet received in step S201 is viewed, and the VLAN ID of the VLAN circuit 170 that is the destination of the packet, and the VLAN at that time are determined. A list of IDs of the accommodated data transfer logical network 30 is obtained.

  Then, it is checked whether or not the ID of the logical network that is the transmission source of the packet specified in step S203 is included in the list of logical network IDs obtained in step S205 (step S206). This is a check whether or not the logical network that is the source of the packet matches the logical network that currently accommodates the VLAN that connects the destination terminal of the packet. If the two match, there is no problem even if the packet is output to the destination VLAN line 170, and the process is moved to (step S207 and step S208).

  In step S207, the IP header and GRE header for encapsulation given outside the packet received in step S201 are deleted. In step S208, the remaining payload packet is output to the VLAN line corresponding to the VLAN interface ID obtained in step S204, that is, the VLAN line 170 to which the terminal that is the destination of the packet is connected. finish.

  If the logical networks do not match in the inspection in step S206, it is determined that the packet is not to be transmitted to the VLAN circuit 170, the packet is discarded (step S209), and the processing of this chart is terminated.

  The above is the description of the procedure for the terminal accommodating apparatus to output the packet input from the backbone side logical network to the terminal on the VLAN line as the destination.

  Above, description of the whole physical structure of this embodiment and operation | movement of the data transfer apparatus 300 and the terminal accommodation apparatus 120 of a transmission base is complete | finished.

<Real-time content delivery in a many-to-one type of communication>
Hereinafter, in the system according to the present embodiment, processing until real-time content distribution is performed in the many-to-one communication mode described in the background art described above will be described.

  As a premise for explanation, it is assumed that the physical construction of the system of this embodiment has been completed in advance. That is, physical connections among all the data transfer devices 300, the terminal accommodation device 120 at the transmission base, and the terminal accommodation device 220 at the reception base, allocation of physical IP addresses to their actual network interfaces, physical subnetwork Assume that the IP subnetwork address assignment for is complete. Furthermore, connection of all terminals of each user base to the terminal accommodating device 120 of the transmission base and the terminal accommodating device 220 of the receiving base has been completed, and the assignment of the IP address to the network interface of the terminal, the transmission server 110, and the reception It is assumed that the VLAN ID assignment and the IP subnetwork address assignment to the VLAN line 170 connecting the terminal 210 to the terminal accommodation apparatus 120 at the transmission base and the terminal accommodation apparatus 220 at the reception base are also completed. Then, the IP address of each terminal, the VLAN ID of the VLAN circuit 170 to which each terminal is connected, and the terminal accommodation apparatus 120 of the transmission base that physically accommodates each transmission server 110 and reception terminal 210, the reception base It is assumed that the physical IP address of the terminal accommodating device 220 is stored in the network database 140 shown in FIG.

  Assuming the above, the process until content distribution is implemented is roughly divided into four stages: management logical network construction processing, content placement processing, data transmission logical network construction processing, and content distribution processing. Consists of. FIG. 25 to FIG. 28 are charts showing the processing in each of these stages.

<Management logical network construction processing>
The construction process of the management logical network 70 is performed as the first stage until content distribution is performed. In this process, the control shown in FIG. 29 and FIG. 34 among the communication quality management device 150, the content database 130, the network database 140, and the terminal accommodation device 120 at the transmission base of each base and the terminal accommodation device 220 at the reception base. The management logical network 70 for performing communication for the purpose is constructed so that communication can be performed.

  Details of this processing will be described below with reference to FIG.

  First, the topology design of the management logical network 70 is performed (step S301). This is performed by a user who uses the network service. The topology design is to determine the position of the logical node in the logical network, that is, the data transfer apparatus 300 in which the logical node is arranged, and further decide the arrangement of the logical link 40 that connects the logical nodes. In the case of management logical network construction processing, a topology is prepared so that the communication shown in FIGS. 29 and 34 can be executed.

  When the topology design is completed, the network service provider is requested to perform processing for actually constructing the management logical network 70 (step S302). At this time, information regarding the topology of the network is provided to the network service provider, but the format of data describing this is not defined in this specification.

  Upon receiving this request, the network service provider performs an operation of actually constructing the management logical network 70 having the requested topology (step S303). By this operation, the VM corresponding to the logical node and the resource corresponding to the logical link are allocated inside the related data transfer apparatus 300. Specifically, the GRE key is assigned to the logical link 40.

  When the processing in step S303 is completed, the network service provider device (not shown) notifies the requesting network service user that the construction of the management logical network 70 is completed (step S304). In this notification, the ID for identifying the constructed management logical network 70 and the terminal accommodating device 120 at the transmission base on the user's base side, the terminal accommodating device 220 at the receiving base, and the data transfer device 300 are constructed. The GRE key of the logical link 40 that has been added is notified to the user (note that the specific operation content for constructing the logical network in step S303 and the specific information for determining the GRE key and logical network ID are determined. This method is not specified in this specification).

  Upon receiving this notification, the user device (not shown) registers in the network database 140 information related to the constructed management logical network 70, that is, a set of an ID for identifying the management logical network 70 and network topology information. (Step S305).

  Further, using the notified logical network ID and GRE key information, the logical network connection management table of the accommodated network changing units 122 and 222 in the terminal accommodating device 120 of each related transmission base and the terminal accommodating device 220 of the receiving base. 450 is described (step S306).

  Then, a subnet IP address is assigned to each sub-network in the management logical network 70 (step S307), and the data transfer device 300, the terminal accommodation device 120 at the transmission base, and the terminal accommodation device at the reception base included in the network. A logical IP address and a logical Mac address are assigned to the virtual network interface 220 (step S308).

  In step S309, the communication quality management device 150 can communicate with the terminal accommodating device 120 at each transmitting base and the terminal accommodating device 220 at the receiving base via the management logical network 70 in the processing of FIG. 29 and FIG. In order to do so, the logical IP address of the virtual network interface of each terminal accommodating apparatus assigned in step S308 is registered in the network database 140.

  In step S310, the value of the GRE key notified in step S304, the backbone interface ID corresponding to the logical link 40 indicated by the GRE key, the logical IP address and the logical Mac address determined in step S308 are determined based on the topology information. Information is described in the virtual network interface management table of the accommodation network changing units 122 and 222 in the terminal accommodation device 120 at each transmission base and the terminal accommodation device 220 at the reception base.

  In step S311, in order to connect the content database 130, the network database 140, and the communication quality management apparatus 150 to the constructed management logical network, the terminals of the receiving base that physically accommodate these apparatuses are accommodated. The VLAN line 170 of the device 220 is accommodated in the management logical network 70 (as shown in FIG. 29 and FIG. 34), the VLAN line 170 of the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base are connected to a certain logical network. The operation of accommodating is basically performed via the communication quality management device 150, but the operation of step S311 is performed by directly operating the terminal accommodating device 120 at the transmission base and the terminal accommodating device 220 at the receiving base. To do).

  When these operations are completed, the route information publicizing process 511 and the ARP process 513 in the logical node in the management logical network 70, the terminal accommodating device 120 at the transmission base, and the terminal accommodating device 220 at the receiving base are activated (step S312). . The route information publicizing process 511 exchanges route information related to the IP sub-network in the logical network, and when this is completed, communication in each logical network becomes possible.

  Thus, the logical network construction process is completed.

<Content arrangement processing>
When the construction processing of the management logical network 70 is completed, the network service user performs the content placement processing shown in FIG. In this process, content data to be distributed is stored in a plurality of transmission servers 110 in a distributed manner. The procedure will be described with reference to FIG.

  First, it is determined which content is distributed and stored in which transmission server at which location (step S401). When it is determined, the content data is divided and stored in the transmission server group determined in step S401 (step S401). S402). When the data storage is completed, the contents are registered in the content database 130 (step S403). Specifically, an entry that registers a content name for identifying the stored content and a list representing a server group storing the content is registered.

<Construction processing of logical network for data transmission>
Subsequent to the content arrangement processing, the construction processing of the data transfer logical network 30 is performed. FIG. 27 is a chart showing the procedure of this processing. The difference from FIG. 25 is that the processing object is a plurality of logical networks 30 for data transfer used for data transmission, and step S309 from the chart of FIG. Since only the step S311 is deleted, the description is omitted.

  When the above processing is completed, content can be distributed. FIG. 28 shows the flow of processing from the start to the end of content distribution.

  Steps S601 to S604 are processing until content distribution is started. When the user activates content distribution (step S601), the available bandwidth measurement process 321 in the terminal accommodating device 120 at the transmission base is activated, and measurement of the available bandwidth of the data transfer logical network 30 used for distributing the content is started. (Step S602) After the transmission server 110 holding the content is assigned to the data transfer logical network 30 (Step S603), the content data transmission starts (Step S604).

  FIG. 29 shows details of control signals transmitted and received between the devices in the processing corresponding to steps S601 to S604 in FIG. Further, for each control signal, data simultaneously transmitted from the transmission side apparatus to the reception side apparatus of the control signal are collectively shown in FIG. These control signals and control data are transmitted between devices via the management logical network 70.

<Processing until the start of content distribution>
The details of the processing up to the start of content distribution will be described below with reference to FIG.

  (1) First, a user of a network service that starts content distribution transmits a content distribution start request to the communication quality management apparatus 150 via the operation terminal 1.

  Data of the content name, a list of all logical network IDs used for distribution, and the physical IP address of the receiving terminal 210 of the content are attached to the control signal (Note that the operation terminal 1 is responsible for communication quality management). It is necessary to be able to communicate with the device 150, the transmission server 110, and the reception terminal 210, but the connection method between the operation terminal 1 and these devices is not defined in this specification).

  (2) Upon receiving this notification, the communication quality management device 150 receives a list of physical IP addresses of the transmission server 110 storing the content, and the terminal storage device 120 of the transmission base of the transmission base 100 that stores them. List of logical IP addresses, VLAN IDs used by the terminal accommodating devices 120 of the respective transmission bases for accommodating the respective transmission servers 110, transmission bit rates when the respective transmission servers 110 transmit the data of the contents, the contents Content database in order to obtain information of the logical IP address of the terminal accommodating device 220 at the receiving base accommodating the receiving terminal 210 and the VLAN ID used by the terminal accommodating device 220 at the receiving base for accommodating the receiving terminal 210 A facility information request is transmitted to 130.

  In this request, the content name and the physical IP address of the receiving terminal are given as data.

  (3) Upon receiving the above equipment information request, the content database 130 receives a list of logical IP addresses of the terminal accommodation devices 120 at the transmission bases that accommodate the transmission server group storing the content, and the transmission bases thereof. VLAN IDs used by each terminal server 120 for accommodating each transmission server 110, the logical IP address of the terminal accommodating device 220 at the receiving base accommodating the receiving terminal 210 for the content, and the terminal accommodating at the receiving base In order to obtain information on the VLAN ID used by the device 220 to accommodate the receiving terminal 210, a terminal accommodation device information request is transmitted to the network database 140.

  In this request, the list of physical IP addresses of the transmission server 110 storing the content held by the content database 130 itself, and the receiving terminal 210 notified by the facility information request from the communication quality management device 150 are received. Is assigned as data.

   (4) When the network database 140 receives the terminal accommodation device information request, the network database 140 uses the device data already registered by the processing shown in FIG. 25 to identify the transmission base that accommodates the designated transmission server group. A list of logical IP addresses in the management logical network 70 of the terminal accommodating device 120, the VLAN IDs used by the terminal accommodating devices 120 at the respective transmission bases to accommodate the respective transmission servers 110, and the designated receiving terminals 210 are accommodated. The logical IP address in the management logical network 70 of the terminal accommodating device 220 at the receiving base and the VLAN ID used by the terminal accommodating device 220 at the receiving base for accommodating the receiving terminal 210 are specified. And these are transmitted to the content database 130 as a terminal accommodation apparatus information notification.

  (5) The content database 130 that has received the terminal accommodation device information notification is a list of physical IP addresses of the transmission server 110 storing the content, and the terminal accommodation device 120 of each transmission base that accommodates the list. List of logical IP addresses in the logical network 30 for data transfer, VLAN ID used by the terminal accommodating device 120 at each transmitting base for accommodating each transmitting server 110, and transmission when each transmitting server 110 transmits data of the content The bit rate, the logical IP address in the management logical network 70 of the terminal accommodating device 220 at the receiving base accommodating the receiving terminal 210 of the content, and the terminal accommodating device 220 at the receiving base are used for accommodating the receiving terminal 210 The VLAN ID is transmitted to the communication quality management apparatus 150 as a facility information notification.

  (6) The communication quality management device 150 that has received the facility information notification sends the content to the terminal accommodation device 220 at the receiving base of the receiving base 200 identified by the logical IP address in the management logical network 70 received from the content database 130. Send a reception preparation start instruction.

  In this content reception preparation start instruction, the VLAN ID used by the terminal accommodating device 220 at the reception base to accommodate the content receiving terminal 210 and a list of logical network IDs used for the content distribution are given as data. .

  Receiving this content reception preparation start instruction, the terminal accommodating device 220 at the receiving base performs an operation of accommodating the VLAN used for accommodating the content receiving terminal 210 in all the logical networks 30 for data transfer used for content distribution. . That is, the logical network ID notified by the reception preparation start instruction is added to the logical network ID list part of the entry including the VLAN ID in the accommodated network management table 440 in the accommodated network changing unit 126 shown in FIG. Write a list. By this operation, the IP packet of the content data transmitted from the backbone network side of these logical networks can be transferred to the receiving terminal 210.

  Further, in addition to this operation, the terminal accommodating device 220 at the reception base performs all of the available bandwidth measurement process (for probe packet reception / available bandwidth measurement) 322 shown in FIG. The data transfer logical network 30 is generated. Each generated available bandwidth measurement process 322 is assigned an IP address in the data transfer logical network 30 to which it belongs.

  (7) Upon completion of these operations, the terminal accommodating device 220 at the receiving base on the receiving side transmits a content reception preparation completion notification to the communication quality management device 150.

  In this content reception preparation completion notification, a list of pairs of the generated IP address of the available bandwidth measurement process 322 and the logical network ID to which the process belongs is assigned as data.

  (8) Upon receiving the content reception preparation completion notification, the communication quality management device 150 stores the content determined from the list of logical IP addresses in the management logical network 70 received from the content database 130. An available bandwidth measurement process activation instruction is transmitted to the terminal accommodating devices 120 at all transmission bases accommodating the server 110.

  The usable bandwidth measurement start instruction includes a list of pairs of the IP address of the usable bandwidth measurement process 322 and the logical network ID to which the process belongs, which is notified in the content reception preparation completion notification from the terminal accommodating device 220 at the reception base. Grant as data.

  (9) Upon receiving the usable bandwidth measurement process activation instruction, the terminal accommodating device 120 at each transmission base lists a pair of the IP address of the usable bandwidth measurement process 322 notified in the instruction and the logical network ID to which the process belongs. , The usable bandwidth measurement process (for probe packet transmission) 321 shown in FIG. 19 is generated for the data transfer logical network 30 to which it is connected. Each generated usable bandwidth measurement process 321 is given an IP address in the data transfer logical network 30 to which it belongs.

  When the terminal accommodating device 120 at each transmission base completes the above operation, it transmits an available bandwidth measurement process activation completion notification to the communication quality management device 150.

  The usable bandwidth measurement process activation completion notification includes a logical IP address representing the terminal accommodating device 120 of its own transmission base, and a pair of the IP address of the usable bandwidth measurement process 321 generated by itself and the logical network ID to which the process belongs. A list is assigned as data.

  (10 When the communication quality management device 150 receives the above-mentioned usable bandwidth measurement process activation completion notification from the terminal accommodating devices 120 of all transmission bases, it instructs the terminal accommodating devices 120 of all transmission bases to start the usable bandwidth measurement. At this time, probe packets sent from the terminal accommodating devices 120 at a plurality of transmission bases are concentrated in the middle of the transmission, particularly in a narrow part near the reception base, and the probe packets themselves reduce the bandwidth of the network. In order to prevent compression, the operation of the available bandwidth measuring unit 124 at each transmission base is synchronized, and an instruction is given to transmit probe packets at an interval from each other. The time Sn at which the nth process starts sending probe packets among all the processes received in the completion notification, and each The process informs the time interval Δ from the transmission of one probe packet to the transmission of the next probe packet, the latter being a value common to all processes, and if the total number of processes is X, The time difference between Sn-1 and Sn is Δ / X, and the probe packet transmission start time S1 of the first process that starts probe packet transmission earliest is the first available bandwidth measurement start instruction. Assuming the transmission time to reach the terminal accommodating apparatus at the transmission base having the process, a time that is sufficiently later than the time at which this instruction arrives is designated. Since the process of the bandwidth measuring unit 124 and the communication quality management device 150 are synchronized by NTP, the transmission operation of the available bandwidth measuring unit 124 at each transmission base is specified by such time designation. There are synchronously controlled, as a result, it is possible to each process for each transmission site transmits a probe packet at intervals from one another.

  The value of the above time interval Δ is a time interval at which each process measures the value of the available bandwidth of the transmission path, but this value is used by the communication quality management device 150 to receive an available bandwidth information notification described later. Even if it is shorter than the processing time required to transmit the accommodation instruction to the data transfer logical network 30 after that, it does not make sense, and if the value is excessively long, the change in the usable bandwidth of the transmission path can be quickly made. I can't catch it. As a guideline of the value, the processing time of the communication quality management apparatus 150 is assumed to be about several hundred milliseconds, and a value around 1 second is set. At this time, with respect to the time difference Δ / X between Sn-1 and Sn, the upper limit of X is about several hundred to thousands in order to secure a value of 1 millisecond or more which is the limit of time synchronization accuracy of NTP. It is considered to be a sufficiently large value. Also, if 1 / several milliseconds are secured as Δ / X, when the nth process sends out a probe packet, the probe packet sent out by the n−1th process takes into account the transmission rate of light. It is considered that it is possible to avoid that the probe packets transmitted by each process are concentrated in a narrow part of the network because it is propagated through the network for several hundred km.

  The terminal accommodating apparatus 120 at each transmission base that has received the above usable bandwidth measurement start instruction is designated by the above usable bandwidth measurement start instruction for each usable bandwidth measurement process 321 generated by itself within the usable bandwidth measurement unit 124. The operation of transmitting the probe packet at the specified time is started.

  (11) When probe packet transmission is started in all the available bandwidth measurement processes 321 of the terminal accommodating devices 120 at all transmission bases, the availability corresponding to each data transfer logical network 30 of the terminal accommodating device 220 at the receiving base The bandwidth measurement process 321 uses the probe packet transmitted from each transmission base in the data transfer logical network 30 from the transmission base 100 in the data transfer logical network 30 to its own base (reception base 200). Measure the available bandwidth of the transmission path up to. This process is also time-synchronized with the NTP server, so accurate measurement of available bandwidth is possible. When the measurement of the available bandwidth of all the transmission paths in all the data transfer logical networks 30 is completed, the result is transmitted to the communication quality management apparatus 150 as an available bandwidth information notification.

  This measurement result includes the source address of the packet that transmitted the probe packet, that is, the IP address of the available bandwidth measurement process 322 in the terminal accommodating device 120 at the transmission base that transmitted the probe packet, and the logic that received the probe packet. It is notified as a list of triples of network ID and available bandwidth measurement.

  The communication quality management device 150 that has received this usable bandwidth information notification receives the IP of the usable bandwidth measurement process generated by the terminal accommodating device 120 at each transmission base notified in the usable bandwidth measurement process activation completion notification in (9) in the figure. Using the list of logical network ID pairs to which the process belongs, a list of three sets of the logical IP address of the terminal accommodating apparatus 120 at the transmission base, the logical network ID, and the measured value of the available bandwidth is generated. Then, using this information, the data transfer logical network 30 of the first accommodation destination of each transmission server 110 is determined by the procedure shown in FIG. The procedure of FIG. 31 will be described later.

  (12) When the data transfer logical network 30 of the first accommodation destination of each transmission server 110 is determined, an accommodation instruction to the data transfer logical network 30 is transmitted to the terminal accommodation devices 120 of all transmission bases. In the accommodation instruction to the logical network 30 for data transfer, the VLAN ID used for accommodation of the transmission server 110 and the ID of the logical network of the accommodation destination of the VLAN are individually specified for each terminal accommodation device 120 of the transmission destination. A set of duplicates is assigned as data.

  (13) Upon receiving the accommodation instruction for the data transfer logical network 30, the terminal accommodation apparatus 120 at the transmission base accommodates the VLAN used for accommodation of the transmission server 110 in the designated data transfer logical network 30. Perform the operation. That is, the instructed logical network ID is written in the logical network ID list portion of the entry including the VLAN ID in the accommodated network management table 440 in the accommodated network changing unit 126 shown in FIG. By this operation, the content data transmitted from the transmission server 110 accommodated in the VLAN is transferred into the data transfer logical network 30 on the backbone network. When the terminal accommodation device 120 at the transmission base completes this operation, it sends a notice of completion of accommodation to the data transfer logical network 30 to the communication quality management device 150).

  (14) When the communication quality management device 150 receives the notification from the terminal accommodation devices 120 at all transmission bases, the communication quality management device 150 transmits a content distribution start permission to the operation terminal 1. The content name is given as data to the content distribution start permission.

  (15) The network service user who has received the content distribution start permission through the operation terminal 1 first transmits a data reception start instruction to the content receiving terminal 210. As a result, application software for receiving content is activated in the receiving terminal 210, and content data can be received.

  (16) Next, a data transmission start instruction is transmitted to all transmission servers 110. The content name is given as data to the data transmission start instruction. As a result, application software for transmitting content in the transmission server 110 is activated, and transmission of content data is started.

  The above is the detailed description of the processing up to step S604 in FIG.

  When the transmission of the content data is started, the communication quality management device 150 thereafter responds to the change in the value based on the available bandwidth information periodically transmitted from the terminal accommodating device 220 at the receiving base. The operation of accommodating the data transfer logical network 30 of the transmission server (step S605) is executed until the data transmission of the content is completed.

<Logical network decision processing>
The procedure for determining the data transfer logical network 30 that accommodates the transmission server 110, executed by the communication quality management device 150, will be described below.

  FIG. 31 is a flowchart showing the procedure of the process of determining the data transfer logical network 30 of the first accommodation destination of the transmission server, which is the process of step S603 of FIG. 28. FIG. 32 shows the start of data transmission. A procedure of processing for changing the data transfer logical network 30 at the accommodation destination based on the available bandwidth information periodically transmitted from the terminal accommodation device 220 at the reception / reception base (processing at step S605 in FIG. 28) will be shown later. It is a chart.

The meanings of symbols used in FIGS. 31 and 32 are as follows. For a specific value of the available bandwidth margin value T, refer to the past traffic history of the network, and the available bandwidth at a time interval equal to the available bandwidth measurement time interval for each of the data transfer logical networks 30 described above. The maximum reduction value may be determined as a guide.
N: Total number of transmission bases (total number of terminal accommodating devices at transmission bases) (constant)
L (k): Of the logical networks included in the list of logical networks specified in “(1) Content distribution start request” in FIG. 29, the terminal accommodating device 120 of the transmission base k at the transmission base k is included as the connection destination. The total number of logical networks that are open (constant)
M _k : the total number (constant) of transmission servers physically connected to the terminal accommodation apparatus 120 at the transmission base of the transmission base k
S _{k_i} : i-th transmission server (synonymous with i-th VLAN) physically connected to the terminal accommodation apparatus 120 at the transmission base of the transmission base k
L _{k_i} : i-th logical network among L (k) logical networks
MC _{k_i} : Number of transmission servers accommodated in L _{k_i} at transmission base k (variable)
B _{k_i:} L in _{k_i,} leading to the terminal accommodation apparatus of the receiving site from the terminal accommodation apparatus of the transmission points of the transmission points k, measured values of usable bandwidth of the transmission path to the End-to-End Routing protocols in the L _{k_i} stipulated ( variable)
R: Bit rate (constant) at which each transmission server sends the content data to the network
T: Available bandwidth margin for the logical network (constant)
First, processing for determining the data transfer logical network 30 of the first accommodation destination of the transmission server 110 will be described with reference to FIG.

  Steps S701 and S702 are initialization processes. In step S701, the value of a counter indicating the number of transmission servers accommodated in all data transfer logical networks is initialized to 0 for all transmission bases. In step S702, a subscript k indicating a transmission base, a subscript i indicating a logical network, and a subscript m indicating a transmission server are initialized to 1, respectively.

  After the initialization process is completed, look at the available bandwidth value of the data transfer logical network of interest, calculate the number of transmission servers that can be accommodated in the data transfer logical network, and determine the number of transmission servers that can be accommodated. The data is accommodated in the logical network for data transfer (steps S703 to S707).

  In step S703, the value of the available bandwidth of the data transfer logical network of interest is referred to from the value of the available bandwidth of each data transfer logical network notified from the terminal accommodating device 210 at the receiving base. .

In step S704, the value obtained by subtracting the bandwidth to be left as a margin value from the value of the available bandwidth notified in step S703 is compared with the data transmission bit rate per transmission server, and one or more transmission servers 110 are compared. It is inspected whether it can be accommodated in the logical network for data transfer. If the latter value is small, no new transmission server can be accommodated, and the processing moves to the processing focusing on the next logical network for data transfer after step S710 (described later). If the latter value is greater than or equal to the former value in the inspection in step S704, the transmission server 110 can be newly accommodated in the data transfer logical network, and therefore the number of transmission servers that can be accommodated in step S705. In step S706, an accommodation instruction for accommodating the number of transmission servers in the logical network for data transfer is transmitted to the terminal accommodation apparatus 120 at the transmission base currently focused on, and then the counter is obtained in step S707. _Update the value of MC _{k_i} .

  In step S708, it is checked whether or not the accommodation destination logical networks of all the transmission servers 110 connected to the terminal accommodation apparatus 120 at the transmission base currently focused on have been determined. If the processing has been completed for all transmission servers, the process proceeds to step S714.

  In step S714, it is checked whether or not the processing for all transmission bases has been completed. If there is a transmission base that has not been processed yet, the process proceeds to step S715, where the value of the subscript k indicating the transmission base is increased by 1, and the processing for the next transmission base is started from step S703. In step S714, if the processing for all the transmission bases has been completed, the processing of this chart ends.

  In step S708, if there remains a transmission server whose accommodation destination logical network has not yet been determined, the process proceeds to step S709. In step S709, the value of the subscript m indicating the transmission server under consideration is incremented by 1, and the process proceeds to step S710.

  In step S710, whether or not there is an available bandwidth that can accommodate the transmission server 110 is checked for all data transfer logical networks including the transmission base. If completed, there is no data transfer logical network candidate that can accommodate the remaining transmission servers 110 while leaving a margin in the available bandwidth, so the remaining transmission servers are currently focused on the data transfer logical network. After transmitting the accommodation instruction for accommodation in the terminal accommodation apparatus 120 at the transmission base currently focused on (step S712), the value of the accommodation transmission server counter is updated (step S713), and step S714 Move on. If there is a data transfer logical network that has not yet been checked in step S710 for whether there is an available bandwidth that can accommodate the transmission server 110, the value of the subscript i indicating the logical network is incremented by one. The processing target is changed to the next logical network, and the process returns to step S703.

  The above is the description regarding the processing for determining the data transfer logical network 30 of the first accommodation destination of the transmission server 110 in the communication quality management apparatus 150.

<Changing process of accommodation destination logical network>
Next, after the data transmission is started, the communication quality management device 150 changes the accommodation destination data transfer logical network 30 based on the available bandwidth information periodically transmitted from the terminal accommodation device 220 at the reception base. The process to perform is demonstrated based on FIG. 32, FIG.

  In step S801, a subscript k indicating a transmission base and a subscript i indicating a logical network are each initialized to 1.

  When the initialization process is completed, the communication quality management device 150 looks at the latest value of the available bandwidth of the data transfer logical network of interest (step S802), and an available bandwidth that is equal to or greater than a predetermined margin value remains. It is inspected whether or not (step S803). If the available bandwidth equal to or more than the margin value remains, the transmission server accommodated in the data transfer logical network is replaced with another data transfer logical network to increase the available bandwidth of the data transfer logical network. It is determined that it is not necessary, and the process proceeds to step S814.

  In step S814, it is checked whether or not the inspection of the available bandwidth has been completed for all the data transfer logical networks including the transmission base. If not completed, the value of the subscript i indicating the data transfer logical network is incremented by 1 (step S815), and the process returns to step S802 to check the available bandwidth of the next data transfer logical network. If the inspection of available bandwidth has been completed for all data transfer logical networks including the transmission base, the process proceeds to step S816, and if the series of processes has been completed for all transmission bases, the processing shown in this chart is terminated. To do.

  If there is a transmission base that has not been processed, the value of the subscript k indicating the transmission base is increased by 1 (step S817), and the subscript i indicating the logical network for data transfer is initialized (step S818). Return to step S802.

  If there is no remaining usable bandwidth equal to or greater than the margin value in the inspection in step S803, a part of the transmission server accommodated in the data transfer logical network is replaced with another data transfer logical network, and the data The available bandwidth of the transfer logical network needs to be increased. In step S804, the number m ′ of transmission servers that need to be accommodated in another logical network for data transfer is obtained.

  In steps S805 to S807, initial setting of the subscript n indicating the candidate for the data transfer logical network of the accommodation change destination is performed. A data transfer logical network different from the original data transfer logical network is examined in ascending order of subscript number (steps S806 and S807).

Then, by referring to the usable bandwidth information of a receiving destination change candidate data transfer logic network L _{k - n,} whether from the logical network L _{k_i} for the original data transfer can accommodate replacement of one or more transmission server to L _{k - n} Is inspected (step S808). If accommodation cannot be changed, the process moves to step S819, and the process moves to a process of searching for a data transfer logical network that becomes a new accommodation change destination candidate.

In step S819, it is checked whether or not the data transfer logical network as the accommodation change destination candidate has been exhausted, and if there is no more data transfer logical network as a candidate, not only the data transfer logical network L _{k_i} , Since it is impossible to recover the usable bandwidth of all the data transfer logical networks including the transmission base k that does not satisfy the marginal value, the process moves to step S816, and the process changes to the processing target transmission base.

  If there is still a candidate data transfer logical network in step S819, the value of the subscript n indicating the candidate data transfer logical network candidate of the accommodation change destination is incremented by one (step S820), and the data indicated by the subscript After guaranteeing that the transfer logical network is different from the data transfer logical network of the accommodation change source (step S821, step S823), the process returns to the inspection of step S808.

If previously accommodated replacement was possible in the test of step S808 described, obtains the upper limit number m "capable of accommodating the transmission server to L _{k - n} from the logical network L _{k_i} for data transfer (step S809), the number An accommodation instruction for relocating the transmission server to the above-described data transfer logical network is transmitted to the terminal accommodation apparatus 120 at the transmission base that is currently focused on (step S810). Among the VLANs that connect the transmission servers accommodated in the network L _{k_i} , the accommodation instruction for setting the accommodation destination for the number m ”with the _smaller number to the logical network L _{k_n} for the data transfer Transmit to the terminal accommodating device.

  Then, after updating the value of the counter representing the number of transmission servers connected to the data transfer logical network of the accommodation source and the accommodation destination (step S811 and step S812), the number of transmission servers determined in step S804 It is inspected whether accommodation to another data transfer logical network has been completed (step S813).

  If the accommodation change has not been completed, the process moves to step S819. If completed, the process moves to step S814.

  The above is the description of the processing for changing the data transfer logical network 30 at the accommodation destination based on the available bandwidth information periodically transmitted from the terminal accommodation apparatus at the reception base after the data transmission is started.

  The communication quality management device 150 performs the operation shown in the flowchart of FIG. 32 every time there is an available bandwidth information notification from the terminal accommodation device 220 at the receiving base, and (18) Content distribution end of FIG. Repeat until a notification (described later) is sent.

  Finally, the completion of data transmission from each transmission server 110 and the contents of the processing accompanying it will be described with reference to FIG. The process shown in this figure is a process corresponding to steps S606 and S607 in the chart of FIG. Further, for each control signal in FIG. 34, data simultaneously transmitted from the transmission side device to the reception side device of the control signal is collectively shown in FIG.

  (17) All the transmission servers at all the transmission bases complete the transmission of the content data, and then the reception terminal 210 that has detected reception of the last packet of the content data notifies the operation terminal 1 of the data reception completion. Send. The content name is assigned as data to this data reception completion notification.

  (18) Receiving this data reception completion notification, the operation terminal 1 transmits a content distribution end notification to the communication quality management device 150 via the management logical network 70. The content distribution end notification is assigned with the content name as data.

  (19) Upon receiving this notification, the communication quality management device 150 stops the operation of determining the data transfer logical network of the transmission server accommodation destination according to the value of the available bandwidth in the distribution of the content (FIG. 32). . Furthermore, an available bandwidth measurement end instruction is transmitted to each terminal accommodating device 120 at the transmission base and terminal accommodating device 220 at the receiving base. A list of available bandwidth measurement process IP addresses included in the content reception preparation completion notification data of (7) in FIG. 29 is given as data to the available bandwidth measurement end instruction to the terminal accommodation device 220 at the reception base. In addition, in the available bandwidth measurement end instruction for the terminal accommodation device 120 at the transmission base, the available bandwidth measurement process notified by the availability bandwidth measurement start completion notification of (9) in FIG. 29 is individually provided for each terminal accommodation device. A list of IP addresses is assigned as data.

  (20) The terminal accommodating apparatus 120 at each transmission base that has received this available bandwidth measurement end instruction receives a series of available bandwidth measurement processes (probe packets) generated when the available bandwidth measurement start instruction in (8) in FIG. 29 is received. After the transmission 321 is stopped and the process is erased, an available bandwidth measurement end notification is transmitted to the communication quality management device 150.

  In addition, the terminal accommodating apparatus 220 at the reception base that has received the usable bandwidth measurement end instruction receives a series of usable bandwidth measurement processes (probe packet reception / usable) generated when the content reception preparation start instruction (6) in FIG. 29 is received. After the operation of 322 (for bandwidth measurement) is stopped and the process is deleted, an available bandwidth measurement end notification is transmitted to the communication quality management device 150.

  When the communication quality management apparatus 150 transmits the above available bandwidth measurement end notification to the terminal accommodation apparatus 120 at all transmission bases and the terminal accommodation apparatus 220 at the reception base, the communication quality management apparatus 150 ends all operations. Thus, a series of content distribution operations are completed.

  Since there is no special condition for the physical network connecting the terminal accommodating apparatuses 120 and 220 and the data transfer apparatus 300 in the present invention as long as it is a normal IP network, the user bases 100 and 200 and the network can be connected as long as these conditions are satisfied. The terminal accommodating device can be installed in the user base regardless of the type of the suspension line connecting the service provider side base 20 and the data communication method. Also, the communication quality management device 150, the content database 130, and the network database 140 can be installed in the user base because there are no restrictions on the installation location.

  The many-to-one type based on the measurement operation of the available bandwidth by the available bandwidth measuring units 121 and 221 in the terminal accommodating devices 120 and 220 and the measurement result for the plurality of logical networks 30 for data transfer constructed in advance. The operation of changing the accommodation of the transmission server 110 to the data transfer logical network 30 in this communication can be realized only by the terminal accommodation devices 120 and 220, the communication quality management device 150, the content database 130, and the network database 140. There is no need to operate the service provider or change settings.

  Furthermore, although it is a problem that is not described in the section of the problem to be solved by the invention, the available bandwidth measurement by the active measurement, which has conventionally been necessary to be performed by the user terminal itself, as seen in Non-Patent Document 7 The logical network (VLAN in this document) switching process realized by changing the configuration of the user terminal (virtual interface construction in the same document), the processing on the user terminal and the configuration of the user terminal by the method of the present invention There is also a side effect that it can be executed without any changes.

  The operations of the terminal accommodation device 120 at the transmission base, the terminal accommodation device 220 at the reception base, and the communication quality management device 150 in the congestion avoidance system shown in FIGS. 7 and 10 are constructed as programs and used as these devices. It can be installed on a computer and executed, or distributed via a network.

  The present invention is not limited to the above-described embodiments, and various modifications and applications are possible within the scope of the claims.

DESCRIPTION OF SYMBOLS 10 Undercarriage line 20 Network service provider side base 30 Data transfer logical network 40 Logical link 70 Management logical network 100 User base (transmission base)
101 to 105 Subnetwork 110 Terminal, transmission server 120 Terminal accommodation device 121 at transmission base 121 Available bandwidth measurement unit 122 Accommodation network change unit 123 Subnetwork and logical link correspondence table table 124 Available bandwidth measurement unit 125 Route setting management unit 126 Accommodation network Change unit 127 Data signal line 128 Common control signal line 130 Content database 140 Network database 150 Communication quality management device 160 VLAN interface 161 Input packet buffer 162 Input packet processing unit 170 VLAN line 171 Backbone interface side input packet processing unit 172 Backbone interface side input Packet buffer 200 User base (receiving base)
201-203 Subnetwork 210 Receiving terminal 220 Terminal accommodating device 221 at receiving base Available bandwidth measuring unit 310 Time information providing unit 320 Logical network resource 321 Available bandwidth measuring process (for probe packet transmission)
322 Available bandwidth measurement process (for probe packet reception / available bandwidth measurement)
410 VLAN management table 420 Backbone interface management table 430 Opposite real network interface management table 440 Accommodating network management table 450 Logical network connection management table 460 Virtual network interface management table 510 Logical network resource 511 Route information public information process 512 ARP table 513 ARP process

Claims (6)

A network congestion avoidance system that avoids congestion during content data distribution,
One or more transmission side terminals installed in the network service user base and transmitting content;
Installed in each viewer in the user base for the network service, the receiving terminal that receives the content,
One or more data transfer devices installed in a network service provider's site and transferring IP packets storing the content data;
Installed in the network service user base, connecting the transmitting terminal or the receiving terminal via a subnetwork, and further connecting the network service user base and the network service provider base A terminal accommodating device connected to the data transfer device via an underline that is a physical IP network,
A communication quality management device connected via the sub-network to the terminal accommodating device in a user base of any one of the network services;
A list of content names distributed and stored in the data transmission side terminal group, a list of transmission side terminals storing each content, and a data transmission bit when each transmission side terminal transmits data of each content A content database with rate information;
List of terminal accommodating devices for accommodating transmitting side terminals, list of transmitting side terminals, list of sub-networks used by terminal accommodating devices for accommodating transmitting side terminals to accommodate transmitting side terminals, transmitting side terminals and transmitting side terminals A network database having information on the correspondence between the terminal accommodating device that accommodates the sub-network, and
Have
The data transfer device
One or a plurality of virtual data transfer nodes (hereinafter referred to as “logical nodes”) are generated in the data transfer device, and the logical nodes on the other data transfer devices and their own logical nodes A network consisting of a set of logical nodes and logical links (hereinafter referred to as “virtual networks”) that connect logical nodes on a physical network to which one or more are connected. In a data transfer apparatus constructed by superimposing one or a plurality of terminals and physically connected directly to the terminal accommodating apparatus, the virtual network is provided between the logical node inside the terminal accommodating apparatus and the terminal accommodating apparatus. By constructing a logical link corresponding to the virtual network for each, a logical node, a logical link connecting the logical nodes, the logical node and the logical node Network composed of a set of logical links connecting end accommodating device (hereinafter, referred to as "logical network") is formed,
The terminal accommodating device that accommodates the receiving terminal is:
Measures the available bandwidth of the logical network based on a test packet for measuring the available bandwidth of the logical network between the own device and another terminal accommodating device sent from the terminal accommodating device that accommodates the transmitting terminal. And having usable bandwidth information notifying means for notifying the communication quality management device of usable bandwidth information,
The communication quality management device includes:
An available bandwidth measurement instruction means for sending an available bandwidth measurement instruction to a terminal accommodating device that accommodates each transmitting terminal;
A transmission-side terminal in each terminal accommodating device that accommodates the transmitting-side terminal so that the usable bandwidth information is received from a terminal accommodating device that accommodates the receiving- side terminal and the available bandwidth is equal to or greater than a predetermined margin value. An accommodation change instruction means for determining a logical network that accommodates each of the sub-networks that are connected to each other, and sending an accommodation instruction to change the accommodation to the determined logical network to each terminal accommodating apparatus that accommodates the transmitting side terminal; ,
Have
The terminal accommodating device that accommodates the transmitting terminal is:
An available bandwidth measuring means for transmitting the test packet based on the available bandwidth measurement instruction from the communication quality management device to a terminal accommodating device accommodating the receiving terminal;
Based on the accommodation instruction from the communication quality management device, accommodation network changing means for dynamically changing a logical network that accommodates each sub-network connecting the transmission side terminal;
I have a,
The accommodation change instructing means of the communication quality management device is:
Based on the available bandwidth information acquired from the terminal accommodating device that accommodates the receiving terminal, the content database and the network database are referred to, a logical network that accommodates the transmitting terminal is determined, and the transmitting terminal is A congestion avoiding system comprising accommodation instruction means for transmitting an accommodation instruction to a logical network to a terminal accommodation apparatus to be accommodated . The usable bandwidth measurement instructing means of the communication quality management device is:
When active probe packet measurement using probe packets is used as the available bandwidth measurement method, the probe packet transmission start time and transmission time interval for the available bandwidth measuring means of the terminal accommodating device accommodating the transmitting terminal. Including packet transmission instruction means for instructing
The usable bandwidth measuring means of the terminal accommodating device that accommodates each transmitting side terminal,
Based on the transmission start time and the transmission time interval of the probe packet acquired from the communication quality management device, the terminal accommodating device accommodating each transmitting terminal synchronizes and transmits the probe packet while being spaced apart from each other Including
The terminal accommodating device that accommodates the receiving terminal is:
Means for measuring an available bandwidth of a transmission path from the terminal accommodating apparatus accommodating the transmitting terminal in the logical network to the own apparatus using the probe packet transmitted from the terminal accommodating apparatus accommodating each transmitting terminal. The congestion avoidance system according to claim 1, comprising: The accommodation network changing means of the terminal accommodation device that accommodates the transmission side terminal,
A table that associates the IDs of the sub-networks connecting each transmitting terminal with the IDs of the logical links corresponding to each logical network;
Based on the accommodation instruction from the communication quality management device, the correspondence between the ID of the sub-network in the table and the ID of the logical link is determined, and based on the correspondence, the logical network that is the accommodation destination of each transmitting terminal is determined. Means for sending data to the corresponding logical link;
The congestion avoidance system according to claim 1, comprising: A network congestion avoidance method for avoiding congestion at the time of content data distribution,
One or more transmission side terminals installed in the network service user base and transmitting content;
Installed in each viewer in the user base for the network service, the receiving terminal that receives the content,
One or more data transfer devices installed in the network service provider's site and transferring IP packets storing the content data;
Installed in the network service user base, connecting the transmitting terminal or the receiving terminal via a subnetwork, and further connecting the network service user base and the network service provider base A terminal accommodating device connected to the data transfer device via an underline that is a physical IP network,
A communication quality management device connected via the sub-network to the terminal accommodating device in a user base of any one of the network services;
A list of content names distributed and stored in the data transmission side terminal group, a list of transmission side terminals storing each content, and a data transmission bit when each transmission side terminal transmits data of each content A content database with rate information;
A list of terminal accommodating devices that accommodate the transmitting terminal, a list of transmitting terminals, a list of sub-networks that the terminal accommodating device that accommodates the transmitting terminal uses to accommodate the transmitting terminals, the transmitting terminal and the transmitting side A network database having information on correspondence between a terminal accommodating device accommodating a terminal and a sub-network;
In a system having
The data transfer device is
One or a plurality of virtual data transfer nodes (hereinafter referred to as “logical nodes”) are generated in the data transfer device, and the logical nodes on the other data transfer devices and their own logical nodes A network consisting of a set of logical nodes and logical links (hereinafter referred to as “virtual networks”) that connect logical nodes on a physical network to which one or more are connected. In a data transfer apparatus constructed by superimposing one or a plurality of terminals and physically connected directly to the terminal accommodating apparatus, the virtual network is provided between the logical node inside the terminal accommodating apparatus and the terminal accommodating apparatus. By constructing a logical link corresponding to the virtual network for each, a logical node, a logical link connecting the logical nodes, the logical node and the logical node Network composed of a set of logical links connecting end accommodating device (hereinafter, referred to as "logical network") assumes that form a,
The communication quality management device sends an available bandwidth measurement instruction to each terminal accommodating device that accommodates the transmitting side terminal; and an available bandwidth measurement instruction step;
A terminal accommodating device that accommodates the transmission side terminal, based on the usable bandwidth measurement instruction from the communication quality management device, measures the usable bandwidth of the logical network between the own device and another terminal accommodating device. An available bandwidth measuring step for transmitting a test packet to a terminal accommodating apparatus accommodating the receiving terminal;
The terminal accommodating device accommodating the receiving terminal measures the usable bandwidth of the logical network based on the test packet sent from the terminal accommodating device accommodating the transmitting terminal, and transmits the usable bandwidth information to the communication An available bandwidth information notification step for notifying the quality management device;
Each terminal that accommodates the transmission-side terminal so that the communication quality management device receives the usable bandwidth information from a terminal accommodating device that accommodates the receiving- side terminal, and the usable bandwidth is equal to or greater than a predetermined margin value. In the accommodation apparatus, a logical network that accommodates each sub-network connected to the transmission side terminal is determined, and an accommodation instruction to change the accommodation to the determined logical network is sent to each terminal accommodation apparatus that accommodates the transmission side terminal. A containment change instruction step to be sent out;
Accommodating a terminal accommodating apparatus accommodating the transmitting terminal dynamically changing a logical network accommodating each sub-network connecting the transmitting terminal based on the accommodating instruction from the communication quality management apparatus A network change step;
The stomach line,
In the accommodation change instruction step,
Based on the available bandwidth information acquired from the terminal accommodating apparatus accommodating the receiving terminal, the communication quality management apparatus refers to the content database and the network database and determines a logical network accommodating the transmitting terminal. Then, a congestion avoidance method , comprising: transmitting an accommodation instruction to a logical network to a terminal accommodation apparatus that accommodates the transmission side terminal . In the usable bandwidth measurement instruction step,
When the communication quality management device uses active probe packet measurement using a probe packet as a method for measuring available bandwidth, the transmission start time and transmission time interval of the probe packet are transmitted to the terminal accommodating device accommodating each transmitting terminal. Instruct
In the usable bandwidth measurement step,
Each terminal accommodating apparatus that accommodates the transmitting terminal transmits the probe packet in synchronization with each other based on the transmission start time and the transmission time interval of the probe packet acquired from the communication quality management apparatus. And
In the available bandwidth information notification step,
From each terminal accommodating apparatus accommodating the transmitting terminal in the logical network, the terminal accommodating apparatus accommodating the receiving terminal uses the probe packet transmitted from each terminal accommodating apparatus accommodating the transmitting terminal. 5. The congestion avoidance method according to claim 4, wherein the available bandwidth of the transmission path to the own device is measured. In the accommodating network changing step,
Based on the accommodation instruction from the communication quality management device, the ID of the subnetwork connected to each terminal on the transmitting side, the ID of the subnetwork in the table that associates the ID of the logical link corresponding to each logical network, and the logical The congestion avoiding method according to claim 4, wherein a correspondence with a link ID is determined, and data is transmitted to a logical link corresponding to a logical network of an accommodation destination of each transmitting terminal based on the correspondence.

Images