JP3848067B2 - Route setting method, route management method and active router in network - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a route setting method in a network including an active router, and more particularly to a route setting method for setting a communication path between terminals that perform communication via a network including an active router.
[0002]
The present invention also relates to a route management method set by such a route setting method.
[0003]
Furthermore, the present invention relates to an active router included in a network where a communication path is set by such a route setting method.
[0004]
[Prior art]
In recent Internet technologies, a router performs functions of route selection, information storage, and transfer, which are network layer processing. On the other hand, there is a new network technology called an active net in which a host terminal and a router operate cooperatively in order to realize more advanced services on the Internet and effectively use network resources. This active net is characterized in that the router shares the processing of higher layers including the application layer higher than the network layer, and the user can control the processing function of the router. It takes time until a router having an active net function (hereinafter referred to as an active router) is widely introduced and deployed in the network, and the active net application technology assumes a mixture of existing routers and active routers over a long period of time. is important.
[0005]
The active net is not exclusive to the current Internet protocol. A typical implementation is a common protocol called ANEP (Active Network Encapsulation Protocol: Internet RFC DRAFT, July 1997) in the upper layer of the UDP / IP protocol. This is realized using a header. Therefore, in the inter-terminal communication using the active net, packets of the active net are transferred by UDP communication between the terminal and the active router and between the intermediate active routers.
[0006]
In the host terminal (client, server, etc.) and the active router, routing information regarding the position and route of the adjacent active router is required. Therefore, for example, a management node is provided in the network for managing the introduction status of the active router to the network and managing route (route) information and distributing the information to the host terminal and the active router, and the management node is active. With the addition and deletion of routers, the location information and route information of the active router are changed and the changed information is distributed.
[0007]
[Problems to be solved by the invention]
However, with the above-described technology, an active router cannot be introduced and deployed until a management node is prepared in the network. In addition, every time it is added to or deleted from the network, the route information must be updated, and the processing at the management node for managing the information becomes complicated.
[0008]
Accordingly, a first problem of the present invention is to enable the introduction of an active router without providing a management node in the network for managing and distributing the position information and route information of the active router as described above. Is to provide a route setting method.
[0009]
A second problem is to provide a route management method for managing a communication path (packet transfer route) set according to such a route setting method.
[0010]
A third problem is to provide an active router included in a network in which a communication path is set according to such a route setting method.
[0011]
[Means for Solving the Problems]
In order to solve the first problem, the present invention has a structure in which a plurality of nodes having a packet transfer function are connected by a route, and a part of the plurality of nodes. Is composed of an active router having a predetermined processing function together with a packet transfer function, and the first terminal and the second terminal in a network in which a transfer rule for relaying a packet node in communication between the terminals is determined in advance. In a route setting method for setting a route for transferring a packet while performing processing at the active router with a terminal of a first terminal, the first type of packet is transferred from the first terminal to the predetermined transfer rule. The order of the active routers that forward to the second terminal via each node in the network and relay the first type packet in the process Based on this, the active router to which the packet is to be transferred next is determined from each active router when the packet is transferred from the second terminal to the first terminal, and the packet is transferred from the second terminal to the first terminal. Configured to set the route.
[0012]
In such a route setting method, in a situation where it is not clear where the active router exists in the network, the first type packet is sent from the first terminal in the network according to a predetermined transfer rule. The data is transferred to the second terminal via each node. In this process, several active routers relay the first type of packet. Based on the order of the active router that relays the first type of packet, from the second terminal to the first terminal that is opposite to the direction from the first terminal to the second terminal. The active router to which the packet is to be transferred next is determined from each active router in transferring the packet. As a result, the connection of the active routers determined as described above becomes a packet transfer route from the second terminal to the first terminal.
[0013]
In addition, from the viewpoint of setting a transfer route from the first terminal to the second terminal, the present invention provides the route setting method according to claim 2, wherein the second terminal is connected to the second terminal. The order of the active router that relays the packet of the type according to the forwarding route set as described above, relays the active router in the network to the first terminal, and relays the packet of the second type in the process. Based on the above, the active router to be transferred next is determined from each active router when the packet is transferred from the first terminal to the second terminal, and the packet transfer route from the first terminal to the second terminal is determined. Configured to set.
[0014]
In such a route setting method, the second type packet is relayed from the second terminal to the active router in the network according to the transfer route from the second terminal to the first terminal set as described above. Transfer to the first terminal. Based on the order of the active router that relays the second type of packet in the process, the direction from the first terminal to the second terminal is opposite to the direction from the second terminal to the first terminal. Next, the active router to be transferred next is determined from each active router in transferring the packet toward. As a result, the connection of the active routers determined as described above becomes a packet transfer route from the first terminal to the second terminal.
[0015]
From the viewpoint of performing processing in each active router in a distributed manner, the present invention provides each route setting method when each of the first packets is transferred from the first terminal to the second terminal. The active router may be configured to determine the previous active router in the transfer of the first packet as the active router to which the packet is to be transferred next in the packet transfer route from the second terminal to the first terminal. it can.
[0016]
Further, from the same point of view, the present invention provides that, when the second type packet is transferred from the second terminal to the first terminal in the route setting method, each active router receives the second packet. In the transfer route of the packet from the first terminal to the second terminal, the active router immediately before the transfer of the packet can be determined as the active router to which the packet is to be transferred next.
[0017]
In order to solve the second problem, the present invention provides, as described in claim 5,
In the transfer route management method set by any one of the above route setting methods, each active router is designated as an active router to which a packet is to be transferred next when the packet is transferred from the second terminal to the first terminal. The determined active router is managed in correspondence with the first terminal that is the final transfer destination of the packet.
[0018]
In a situation where a route is set from a second terminal to a plurality of first terminals, there may be a plurality of active routers to be determined as active routers to which a packet is to be transferred next from each active router. In such a situation, each active router, when receiving a packet, can determine the active router to be transferred next based on the first terminal as the final transfer destination.
[0019]
Similarly, in order to solve the second problem, the present invention provides a transfer route management method set by any one of the route setting methods described above, wherein each active router When transferring a packet from one terminal to the second terminal, the active router determined as the active router to which the packet is to be transferred next is managed in correspondence with the second terminal that is the final transfer destination of the packet. Configured.
[0020]
In order to solve the second problem, as described in claim 7, the present invention provides a forwarding route management method set by any one of the route setting methods described above, wherein each active router includes: When the packet is transferred from the second terminal to the first terminal, the active router determined as the active router to which the packet is to be transferred next is changed to the type of service distributed from the second terminal to the first terminal. It is configured to be managed in correspondence.
[0021]
When performing the same information distribution service (multicast service) from a second terminal to a plurality of first terminals, in the management method as described above (Claim 5), each active router has an active router to be transferred next. Even if they are the same, if the first terminal as the final transfer destination is different, the same active router must be managed separately. According to the route management method as described above, each active router manages the active router that is the next transfer destination of the packet for each type of service, so the management complexity in the above situation is reduced. Is prevented.
[0022]
In order to solve the third problem, the present invention provides, as described in claim 8, an active router provided in a network in which a transfer route is set by any one of the route setting methods described above, Management that manages the active router determined as the next active router to which the packet is to be transferred next when the packet is transferred from the first terminal to the first terminal in association with the first terminal that is the final transfer destination of the packet Configured to have means.
[0023]
According to the present invention, an active router provided in a network in which a transfer route is set by any one of the route setting methods as described in claim 9, the first terminal is changed to the second terminal. It is configured to have route management means for managing the active router determined as the active router to which the packet is to be transferred next when the packet is transferred in correspondence with the second terminal that is the final transfer destination of the packet .
[0024]
Furthermore, according to the present invention, an active router provided in a network in which a transfer route is set by any one of the route setting methods as described in claim 10, the second terminal to the first terminal Route management means for managing the active router determined as the active router to which the packet is to be transferred next when the packet is transferred to the first terminal in accordance with the type of service distributed from the second terminal to the first terminal. Configured to have.
[0025]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0026]
Assume that the network in which the route is set according to the route setting method according to the embodiment of the present invention is configured as shown in FIG. In this example, an active net is configured in an IP (internet protocol) network.
[0027]
In FIG. 1, client terminals C1, C2, C3 and servers (terminals) S1, S2 are connected by a network. This network is composed of the router R1, the active router A1, the router R2, the active router A2, the router R3, the active router A3, the router R4, and the active router A4, and paths connecting the nodes. Each router R1, R2, R3, R4 is a router provided in a known IP network (hereinafter referred to as a normal router). The client terminal C1 is connected to the normal router R1 via the path c11, the paths a11 and a12 are connected to the active router A1, the paths a21, a22, and a23 are connected to the active router A2, and the path a31 is connected to the active router A3. , A32 are connected.
[0028]
The server S1 distributes the service specified by the service identifier Sid = 10, and the server S2 distributes the service specified by the service identifier Sid = 30. The client terminals C1 and C2 request a service with the service identifier Sid = 10 from the server C1, and the client terminal C3 requests each service with the service identifiers Sid = 10 and Sid = 30.
[0029]
Each node (normal router, active router, client terminal, server) of the network as described above has a functional configuration as shown in FIG.
[0030]
The client terminal C (C1, C2, C3 in FIG. 1) has processing functions (modules) in each of the link layer, the IP layer, the UDP (user datagram protocol) layer, the active net (AN) layer, and the application layer. is doing. The normal router R (R1, R2, R3, R4 in FIG. 1) is a router provided in a known IP network as described above, and has each processing function in the link layer and the IP layer. The active router A (A1, A2, A3, A4 in FIG. 1) has a processing function in each of the link layer, the IP layer, the UDP layer, and the active net (AN) layer. Therefore, unlike each normal router R, the active router A performs various processes by the processing function in the UDP layer and the active net (AN) layer in addition to the packet transfer performed by the processing function in the link layer and the IP layer. Do.
[0031]
With such a functional configuration, each node (client terminal C, normal router R, active router A, server S) can perform IP communication, and between the client terminal C and active router A, active router A Communication between the active router A and the server S is possible using UDP (user datagram protocol).
[0032]
A packet transferred in the network as described above according to the function of the active net (hereinafter referred to as AN packet) has, for example, a structure shown in FIG.
[0033]
That is, each AN packet has a header area of an IP header, a UDP header, and an AN header, and a data area serving as an area for describing original data. In the IP header area,
Destination IP address (IP.Dst),
Protocol type (PID),
Header checksum result, etc.
Is described.
[0034]
The destination IP address (IP.Dst) represents the IP address of the destination node. The protocol type (PID) represents the type of processing protocol (for example, UDP or the like) adopted by the destination node. The result of the header checksum represents a checksum value calculated in order to ensure the reliability of the header.
[0035]
In the UDP header area,
Destination UDP port number, etc.
Is described.
[0036]
The destination UDP port number represents a port number that identifies the UDP processing module at the destination node.
[0037]
In the area of the AN header,
AN packet type,
Destination AN address (AN.Dst),
Source AN address (AN.Src),
Neighboring AN address (AN.Pre),
AN service identifier (AN.Sid), etc.
Is described.
[0038]
The type of AN packet represents the type of packet to be transferred on the active net. The types of AN packets include a route setting request packet, a route setting response packet, a service participation request packet, a service participation response packet, and an AN packet for normal data transmission, which will be described later. The destination AN address (AN.Dst) represents an address in the active net of the destination node. The source AN address (AN.Src) represents an address in the active net of the packet source node. The adjacent AN address (AN.Pre) represents an address in the active net of its own node (represents the address of a node adjacent to the destination node of the packet before this. Therefore, the active router that has received the AN packet You can see the address of the immediately adjacent active router). The AN service identifier (AN.Sid) represents an identifier for specifying a service provided in the active net. In FIG. 1 described above, for simplicity, the AN service identifier is simply represented as “Sid”.
[0039]
The IP address and AN address may be defined in the same system or in different systems.
[0040]
In the network in the above state, the active router A1 creates an AN routing table as shown in FIG. 4A and the active router A2 as shown in FIG. An AN routing table is created, and the active router A3 creates an AN routing table as shown in FIG. This AN routing table includes a node having an active net function as a destination of an AN packet (destination AN), a node having an active net function to which the packet is to be transferred next (neighboring AN), and a route to which the packet is to be transmitted ( Each of the outgoing routes is represented in the up and down directions.
[0041]
By forming such an AN routing table in each active router, it becomes possible to transfer packets in the active net. For example, in upstream communication from the client terminal C1 to the server S1, when the client C1 issues an AN packet from the path c11 to the active router A1, this packet is transferred to the active router A1 via the normal router R1. The active router A1 performs predetermined processing on the received AN packet, and then refers to the uplink of the AN routing table (see FIG. 4A) to send the processed AN packet to the route a12 (outbound route). ) To the active router A2 of the adjacent AN. Then, the AN packet is transferred to the active router A2 via the normal router R2. When the active router A2 receives the AN packet, the active router A2 performs a predetermined process on the AN packet, and then refers to the uplink of the AN routing table (see FIG. 4B) to send the processed packet to the route a22. From (departure route) to the active router A3 of the adjacent AN. Then, this AN packet is transferred to the active router A3 via the normal router R3. The active router A3 that has received the AN packet performs predetermined processing on the received AN packet, and then refers to the uplink of the AN routing table (see FIG. 4C) to route the processed AN packet to the route. Issued to a server S1 of the adjacent AN from a32 (departure route). Then, the AN packet is transferred to the server S1 via the normal router R4.
[0042]
Each time an AN packet issued from the client terminal C1 passes through each of the active routers A1, A2, A3 as described above, the AN packet is processed and transferred to the server S1, which is the final destination. The server S1 processes the data contained in the AN packet received from the client terminal C1 (application layer processing, etc.), and then issues an AN packet related to information to be distributed to the client terminal C1 from the path s11 to the active router A3 To do. Thereafter, each time an AN packet is received by each active router, as in uplink communication, the AN packet is transferred to an active router that is a neighboring AN with reference to the downlink of the AN routing table. When the active router A1 receives the AN packet, the active router A1 performs a predetermined process on the received AN packet, and then refers to the downstream of the AN routing table (see FIG. 4A) to process the processed AN. A packet is issued from the route a11 to the client terminal C1 of the adjacent AN. Then, the AN packet is transferred to the client terminal C1 via the normal router R1.
[0043]
Through the procedure described above, communication via the active net is performed between the application layer of the client terminal C1 and the application layer of the server S1. As a result, the client terminal C1 can receive distribution of information related to the service Sid = 10 from the server S1.
[0044]
In the network (IP network) shown in FIG. 1, packets can be transferred according to IP (Internet protocol) by processing modules in the link layer and the IP layer in each node.
[0045]
Next, a procedure for creating the above-described AN routing table, that is, a route setting method will be described.
[0046]
For simplification, the network configuration between the client terminal C1 and the server S1 as shown in FIG. 5 is extracted from the network shown in FIG.
[0047]
In the example shown below, the AN address and the IP address are defined in the same system, and the IP address is used as the AN address of the active net. In a router, a plurality of communication lines are usually connected, and an IP address is assigned to each. Therefore, one of them is selected as a representative and used as an AN address.
[0048]
Based on the request from the application, the client terminal C1 that wants to receive the service specified by the AN service identifier Sid = 10 from the server S1 creates a route setting request packet (AN packet). In each header area of this route setting request packet, UDP as “protocol type (PID)”, a specific number predetermined in the network as “destination UDP port number”, and route setting request as “AN packet type” Type information representing a packet is set, and each address is set as follows.
IP. Dst (destination IP address) = S1
AN. Dst (destination AN address) = S1
AN. Src (source AN address) = C1
AN. Pre (adjacent AN address) = C1
Since it is not possible to know where the active router is on the path from the client terminal C1 to the server S1, the route setting request packet issued from the client terminal C1 is sent to the destination IP in accordance with a known routing method in the IP network. The data is sequentially transferred toward the server S1 specified by the address. The transfer process is shown in FIG.
[0049]
In FIG. 5, the route setting request packet issued from the client terminal C1 is transferred to the active router A1 via the normal router R1 ((1)). When receiving the route setting request packet, the active router A1 notifies the upper layer (UDP layer) of the contents of the route setting request packet as a packet addressed to its own router by the IP layer processing module (details will be described later). .
[0050]
When the processing module in the UDP layer detects that a specific number is set as the destination UDP port number, the contents of the route setting request packet are notified to the higher active net (AN) layer, and this active The net (AN) layer processing module registers an entry in the temporary registration table based on the header information (AN.Dst, AN.Src, An.Pre) (details will be described later). Specifically, as shown in FIG. 6 (a), S1 (= AN.Dst) is a destination AN address (hereinafter simply referred to as “destination AN”) and a source AN address (hereinafter simply referred to as “source AN”). ) As C1 (= AN.Src), C1 (= AN.Pre) as an AN address (hereinafter referred to as “downstream adjacent AN”) indicating the node of the immediately preceding active net, and a route setting request as “outbound route” Each route a11 to which the packet has been transferred is registered in the temporary registration table.
[0051]
When the registration of the entry in the temporary registration table is completed in this way, the header information of the route setting request packet is further reset as follows by the processing module of the active net (AN) layer (details will be described later).
IP. Dst (destination IP address) = S1
AN. Dst (destination AN address) = S1
AN. Src (source AN address) = C1
AN. Pre (adjacent AN address) = A1
In the resetting of the header information, the AN. The value of Pre (adjacent AN address) is changed to the AN address of the own node, that is, the active router A1. The change address is underlined as described above (the same applies hereinafter).
[0052]
The route setting request packet in which the header information is reset as described above is issued from the active router A1. The route setting request packet is further transferred to the active router A2 via the normal router R2 ((2)). The active router A2 that has received this route setting request packet registers an entry in the temporary registration table based on the header information of the received route setting request packet, and, similarly to the active router A1, the header of the route setting request packet. Reset the value of the area.
[0053]
As a result, as shown in FIG. 7A, “Destination AN” is S1 (= AN.Dst), “Departing AN” is C1 (= AN.Src), and “Downstream Adjacent AN” is A1 (= AN.Dst). Pre) and a21 are registered in the temporary registration table as “departure route”. Then, the header information of the route setting request packet is reset as follows.
IP. Dst (destination IP address) = S1
AN. Dst (destination AN address) = S1
AN. Src (source AN address) = C1
AN. Pre (adjacent AN address) = A2
The route setting request packet with the header information reset as described above is issued from the active router A2. The route setting request packet is further transferred to the active router A3 via the normal router R3 ((3)). The active router A3 that has received this route setting request packet registers an entry in the temporary registration table based on the header information of the received route setting request packet, as well as each of the active routers A1 and A2. Reset the value of the header area of the packet.
[0054]
As a result, as shown in FIG. 8A, S1 (= AN.Dst) as “destination AN”, C1 (= AN.Src) as “source AN”, and A2 (= AN.D.) As “downstream adjacent AN”. Pre) and a31 are registered in the temporary registration table as “departure route”. Then, the header information of the route setting request packet is reset as follows.
IP. Dst (destination IP address) = S1
AN. Dst (destination AN address) = S1
AN. Src (source AN address) = C1
AN. Pre (adjacent AN address) = A3
The route setting request packet with the header information reset as described above is issued from the active router A3. The route setting request packet is further transferred to the server S1 via the normal router R4 ((4)).
[0055]
As described above, the route setting request packet is sequentially transferred from the client C1 to the server S1 according to a known IP transfer method based on the destination IP address. A path through which each active router A1, A2, A3 receives a packet with the node of the immediately preceding active net in the upstream communication path between the client terminal C1 and the server S1 by a temporary registration table in which entries are registered in the process. (Exit route) can be recognized.
[0056]
The server S1 that has received the route setting request packet transferred as described above sets an entry corresponding to the downstream direction (S1 → C1) as follows in the AN routing table based on the header information.
Destination AN = Route setting request packet AN. Src (= C1)
Adjacent AN = AN. Of route setting request packet. Pre (= A3)
Outgoing route = line (route) that received the route setting request packet (= s11)
Thus, when the setting of the entry corresponding to the downlink direction is completed in the AN routing table, the server S1 creates a route setting response packet (AN packet). In each header area of this route setting response packet, UDP is set as “protocol type (PID)”, a specific number predetermined in the network as “destination UDP port number”, and route setting response as “AN packet type”. Type information representing a packet is set, and each address is set as follows based on the address information (see (4)) of the received route setting request packet.
IP. Dst = A3 (route setting request packet AN.Pre)
AN. Dst = C1 (route setting request packet AN.Src)
AN. Src = S1 (route setting request packet AN.Dst)
AN. Pre = S1 (AN.Dst of route setting request packet)
The server S1 refers to the AN route table and sends the route setting response packet in which each value of the header area is set as described above from the outgoing route s11 to the active router A3 which becomes the node of the adjacent (immediately) active net. To issue. Then, the route setting response packet is transferred to the active router A3 via the normal router R4 ((5)).
[0057]
When the active router A3 receives this route setting response packet, since the received route setting response packet is a packet addressed to its own node, the content is notified to the upper layer (UDP layer) by the processing module of the IP layer. (Details will be described later). When the processing module of the UDP layer detects that a specific number is set as the destination UDP port number, the contents of the route setting response packet are notified to the higher active net (AN) layer, and this active net Based on the header information (AN.Dst, AN.Src, AN.Pre) of the received route setting response packet and the temporary registration table (see FIG. 8A) by the processing module of the (AN) layer, To register the entry (details will be described later).
[0058]
Specifically, as shown in FIG. 8B, in the upstream direction (C1 → S1), S1 (= AN.Src) as “destination AN” represents the node of the active net to be transferred next. S1 (= AN.Pre) is registered as the “adjacent AN”, and a32 is registered as the “departure route” indicating the route from which the packet should be issued. Further, in the downstream direction (S1 → C1), C1 (= AN.Dst) as “destination AN”, A2 (value of “downstream adjacent AN” in the temporary registration table) as “adjacent AN”, and outgoing route a31 (value of “departure route” in the temporary registration table) is registered (the same as the contents of FIG. 4C).
[0059]
When the registration of the entry to the AN routing table is completed in this manner, the processing module of the active net (AN) layer further determines the header information of the route setting response packet based on the temporary registration table (see FIG. 8A). Reset as follows (details will be described later).
IP. Dst = A2(Value of “Downstream adjacent AN” in the temporary registration table)
AN. Dst = C1
AN. Src = S1
AN. Pre = A3(An own node's AN address)
In resetting this header information, IP. The value of Dst is changed to the value A2 of “downstream adjacent AN” in the temporary registration table, and AN. The value of Pre (adjacent AN address) is changed to the AN address of the own node, that is, the active router A3.
[0060]
The active router A3 refers to the AN routing table (see FIG. 8B) that has completed the entry registration as described above, and sends the route setting response packet with the header information reset as described above to the route a31. To the active router A2. This route setting response packet is further transferred to the active router A2 via the normal router R3 ((6)). The active router A2 that has received this route setting response packet sets the AN routing table based on the header information of the received route setting response packet and the temporary registration table (see FIG. 7A), similarly to the active router A3. To do.
[0061]
Specifically, as shown in FIG. 7B, in the upstream direction (C1 → S1), S1 (= AN.Src) as “destination AN” and A3 (AN.Pre) as “adjacent AN” , A22 is registered as “departure route”. Further, in the downstream direction (S1 → C1), C1 (= AN.Dst) as “destination AN”, A1 (value of “downstream adjacent AN” in the provisional registration table) as “adjacent AN”, and outgoing route a21 (value of “departure route” in the temporary registration table) is registered (the same as the content of FIG. 4B).
[0062]
When the registration of the entry in the AN routing table is completed in this manner, the active router A2 resets the header information of the route setting response packet as follows based on the temporary registration table (see FIG. 7A).
IP. Dst = A1(Value of “Downstream adjacent AN” in the temporary registration table)
AN. Dst = C1
AN. Src = S1
AN. Pre = A2(An own node's AN address)
The active router A2 refers to the AN routing table (see FIG. 7B) that has completed the entry registration as described above, and sends the route setting response packet with the header information reset as described above to the route a21. To the active router A1. The route setting response packet is further transferred to the active router A1 via the normal router R2 (7). The active router A1 that has received the route setting response packet, like the active routers A3 and A2, has an AN routing table based on the header information of the received route setting response packet and the temporary registration table (see FIG. 6A). Set up.
[0063]
Specifically, as shown in FIG. 6B, in the upstream direction (C1 → S1), S1 (= AN.Src) as “destination AN” and A2 (= AN.Pre) as “adjacent AN” However, a12 is registered as the “departure route”. Further, in the downstream direction (S1 → C1), C1 (= AN.Dst) as the “destination AN”, C1 (the value of “downstream adjacent AN” in the temporary registration table) as the “adjacent AN”, and the outgoing route a11 (value of “departure route” in the temporary registration table) is registered (same as the contents of FIG. 4A).
[0064]
When the registration of the entry in the AN routing table is completed in this manner, the active router A1 resets the header information of the route setting response packet as follows based on the temporary registration table (see FIG. 7A).
IP. Dst = C1(Value of “Downstream adjacent AN” in the temporary registration table)
AN. Dst = C1
AN. Src = S1
AN. Pre = A1(An own node's AN address)
The active router A1 refers to the AN routing table (see FIG. 6B) that has completed the entry registration as described above, and sends the route setting response packet with the header information reset as described above to the route a11. To the client terminal C1. This route setting response packet is further transferred to the client terminal C1 via the normal router R1 ((8)).
[0065]
The client terminal C1 that has received this route setting response packet sets an entry corresponding to the upstream direction (C1 → S1) as follows in the AN routing table based on the header information of the route setting response packet.
Destination AN = Route setting response packet AN. Src (= S1)
Neighbor AN = AN. Of route setup response packet Pre (= A1)
Outgoing route = line (route) that received the route setup response packet (= c11)
As described above, the active routers A1, A2, and A3 existing on the path between the client terminal C1 and the server S1 are connected to the AN routing table (FIG. 6B, FIG. 4A, FIG. 7B, and FIG. 4 (b), FIG. 8 (b) and FIG. 4 (c)) are created, and an AN routing table corresponding to the downlink direction is created in the server S1 and an AN routing table corresponding to the uplink direction is created in the client terminal C1. . In this state, communication of AN packets is possible between the client terminal C1 and the server S1 via the active net (active routers A1, A2, A3) as described above.
[0066]
In the network as described above, each time an IP layer processing module in each of the active routers A1, A2, and A3 receives a packet, it performs processing in accordance with the procedure shown in FIG.
[0067]
In FIG. 9, the destination IP address of the received packet is compared with the own IP address to determine whether the received packet is a packet addressed to the own node (S1). If it is determined that the received packet is a packet addressed to its own node (YES in S1), the contents of the packet are notified to the processing module of the upper layer (UDP layer).
[0068]
On the other hand, if the received packet is not a packet addressed to the own node (NO in S1), it is determined whether or not the communication protocol type is UDP based on the value of the protocol type in the header area (S3). If it is determined that the protocol type is UDP (YES in S3), it is determined whether the UDP port number in the header area is the specific number described above (S4). If the received packet is a route setting request packet as described above, it is determined that the UDP port number is a specific number (YES in S4). In this case, as described above, the received packet (route setting request packet) is notified to the upper layer (UDP layer) as a packet addressed to the own node (S5). As a technique for handling a received packet as a packet addressed to its own node, for example, there is a technique of replacing a destination IP address with its own IP address.
[0069]
In the above process, when the communication protocol type is not UDP (NO in S3), or even if the communication protocol type is UDP, the UDP port number is not a specific number (NO in S4), IP The received packet is transferred to the destination IP address according to the routing table (normal packet transfer process) (S6).
[0070]
Through the processing procedure described above, when the processing module in the IP layer of each active router receives the route setting request packet, it reports the contents to the upper layer as a packet addressed to itself (S5). In addition, when receiving the above route setting response packet or AN packet transferred via the active net, these AN packets are always packets addressed to the own node (refer to the AN routing table). Report to the layer (S2). Further, when a packet transferred according to IP is received, the packet is transferred as it is toward the destination IP address (S3, S4).
[0071]
In the upper layer processing module that has received the notification of the contents of the received packet from the IP layer processing module as described above, for example, the processing is executed according to the procedure shown in FIG.
[0072]
In FIG. 10, in the UDP layer processing module (UDP processing), the contents of the packet are notified to the module corresponding to the UDP port number described in the header area of the received packet (S10). As a result, when the UDP port number is the specific number described above, the contents are notified to the processing module of the active net (AN) layer. On the other hand, if the UDP port number is not the specific number, processing is performed in another module corresponding to the port number (S11).
[0073]
In the processing module (AN processing) of the active net (AN) layer that has received the notification of the contents of the received packet as described above, the processing is executed according to the following procedure.
[0074]
Based on the value of the AN packet type described in the header area of the received packet, it is determined whether the received packet is a route setting request packet, a route setting response packet, or another AN packet (S21). . If it is determined that the received packet is a route setting request packet, AN.destination described in the header area of the packet received by “destination AN” in the AN routing table (see FIG. 4) created in its own node. It is determined whether or not an entry (destination AN = AN.Dst) having a value of Dst (destination AN address) exists (S22). This determination is for determining whether or not the route setting toward the node of the destination AN address described in the received packet (route setting request packet) has already been made.
[0075]
The “destination AN” in the AN routing table is the AN. If there is no entry having a value of Dst (NO in S22), the entry is registered in the temporary registration table (see FIGS. 6A, 7A, and 8A) as described above. (S23). Specifically, the AN. The value of Dst is set to “destination AN”, and AN. The value of Src (source AN address) is set to “source AN”, and AN. The value of Pre (adjacent AN address) is registered in “downstream adjacent AN”, and the IP address of the line (route) that received the packet is registered in “outbound route”. By such processing, the entry is registered in the temporary registration table as described above.
[0076]
When the entry is registered in the temporary registration table as described above, the value of the header area of the route setting request packet is reset as described above, and the route setting request packet is transmitted (S24). . By resetting the value of this header area, AN. The value of Pre corresponds to the AN address of the own node, IP. The value of Dst is the AN. It is changed to the value of Dst.
[0077]
Note that the AN. If an entry having the same value as the value of Dst is already in the temporary registration table, the route setting request packet is not transmitted (S24) after the processing in S23 is executed.
[0078]
Further, the “destination AN” of the AN routing table is the AN. If there is an entry that becomes the value of Dst (YES in S22), the AN routing table is set based on the header information of the route setting request packet (S28). Specifically, the following entry is registered in the downward direction of the AN routing table. The AN. If the value of Src is “Destination AN”, AN. The value of Pre is registered in “adjacent AN”, and the value (IP address value) of the line (route) that received the packet is registered in “outbound route”. Then, a route setting response packet is created, and the route setting response packet is transmitted (S29). The header information of the created route setting response packet is set as follows.
[0079]
AN. Of the received route setting request packet. The value of Src is AN. Dst contains the AN. The value of Dst is AN. Src indicates that the AN address of the own node is AN. Pre., The route setting request packet AN. The value of Pre is IP. Dst is set respectively.
[0080]
Furthermore, if it is determined in S21 that the received packet is a route setting response packet, as described above, the AN routing table (FIG. 6) based on the already created temporary registration table and the header information of the received packet. , FIG. 7 and FIG. 8) are set (S25). Specifically, in the temporary registration table, “destination AN” is the AN. All entries having a value of Scr are searched. Then, referring to each entry, the AN. The value of Dst is registered in “destination AN”, the value of “downstream adjacent AN” in the temporary registration table is registered in “adjacent AN”, and the value of “outbound route” in the temporary registration table is registered in “outbound route”. . Further, with reference to the header information of the received route setting response packet, the AN. If the value of Src is “Destination AN”, AN. The value of Pre is registered in “adjacent AN”, and the IP address of the line (route) that has received the route setting response packet is registered in “outbound route”.
[0081]
When registration of the entry in the AN routing table is completed as described above, the header information of the route setting response packet is reset as follows with reference to the temporary registration table. That is, the value of “issue AN” in the temporary registration table is AN. In Dst, the value of “destination AN” in the temporary registration table is AN. In Src, the value of “Downstream adjacent AN” is IP. Dst (destination IP address) contains the AN address of its own node as AN. Set to Pre respectively. Then, the route setting response packet in which the header information is reset in this way is transmitted (S26).
[0082]
Note that the corresponding entry in the temporary registration table is deleted after the transmission of the route setting response packet.
[0083]
Further, when it is determined in the process at S21 that the received packet is another AN packet, a process corresponding to the type of the packet is performed (S27).
[0084]
Each time the active layer A1, A2, A3 processing module of the IP layer and the processing module of the active net (AN) layer receive the AN packet, the processing is performed in the above-described procedure, whereby each active router A1, A2 In A3, an entry related to the route between the client terminal and the server is registered in the AN routing table. That is, a route is set between the client terminal and the server.
[0085]
Next, in the above-described example, when the same information is simultaneously distributed from the server to a plurality of client terminals (in the case of multicast distribution), the route originating from the server is branched as it approaches the client terminal. Therefore, in the active router of the node closer to the client terminal, the entries in the downlink direction of the AN routing table become very large. In order to prevent such a situation, an entry can be registered in the AN routing table as follows.
[0086]
As shown in FIGS. 11 to 13, the value of “downstream adjacent AN” in the temporary registration table is registered in the “destination AN” together with the “adjacent AN” in the downward direction of the AN routing table. That is, the “adjacent AN” and the “destination AN” are not distinguished in the downlink direction.
[0087]
For example, in the case of the example of the network shown in FIG. 1, as shown in FIG. 11B, in the downstream direction of the AN routing table in the active router A1, an AN that represents the client terminal C1 in “adjacent AN” and “destination AN”. The address is registered, and as shown in FIG. 12B, the AN address representing the active router A1 is registered in the “adjacent AN” and “destination AN” in the downstream direction of the AN routing table in the active router A2, As shown in FIG. 13B, the AN address representing the active router A2 is registered in “adjacent AN” and “destination AN” in the downstream direction of the AN routing table in the active router A3.
[0088]
In this way, the number of entries to be registered in the downward direction of the AN routing table is adjacent to the local node, regardless of how far the active router adjacent to the local node (active router) branches. Just the number of active routers.
[0089]
The specific processing in the processing module of the active net (AN) layer for realizing the entry registration of the AN routing table as described above is as follows in the process of S25 shown in FIG. The value of “downward adjacent AN” in the temporary registration table is registered in both “destination AN” and “adjacent AN”, and the value of “outbound route” in the temporary registration table is registered in “outbound route”. In the process at S28, the AN. The value (IP address) of the line (route) that received the packet is “outbound route” in both “destination AN” and “adjacent AN” in the downstream direction of the AN routing table. Are registered respectively.
[0090]
When the server receiving the route setting request packet generates a route setting response packet, the header information is
Destination AN = Route setting request packet AN. Pre
Adjacent AN = AN. Of route setting request packet. Pre
Outbound route = Line (route) that received the route setup packet
Can be set to be
[0091]
As described above, when the AN address of the active router adjacent to the active router is set for both “adjacent AN” and “destination AN” in the downstream direction of the AN routing table in the active router, Then, the final destination (client terminal) of the packet is unknown. For this reason, each active router sends packets related to various services to adjacent active routers identified by all AN addresses set in both “adjacent AN” and “destination AN” in the downstream direction of the AN routing table. If it is transferred, a packet related to a service that is not requested by the client terminal may be distributed to the client terminal that is the final destination. In this case, it is necessary for the client terminal that has received a packet related to a service that is not requested to perform selection of the received packet, which increases the processing load on the client terminal.
[0092]
In order to prevent such a situation, in the following example, a route setting is described in which a packet transfer destination is clarified for each service distributed in each active router.
[0093]
For example, assume a network as shown in FIG.
[0094]
In FIG. 14, each node (client terminals C1, C2, C3, normal routers R1, R2, R3, R4, active routers A1, A2, A3, A4, and servers S1, S2) and a path configuration connecting these nodes Is the same as the network shown in FIG. The server S1 delivers a service with a service identifier Sid = 10. The server S2 delivers a service with a service identifier Sid = 30. The client terminals C1 and C2 request a service with a service identifier Sid = 10 from the server S1, and the client terminal C3 has a service with a service identifier Sid = 30 from the server S1 together with a service with a service identifier Sid = 10 from the server S1. Request.
[0095]
In each of the active routers A1, A2, A3, and A4, entries are registered in the AN routing table according to the procedure described above. For example, the active router A1 performs entry registration with respect to the AN routing table as shown in FIG. 15A, and the active router A2 performs entry registration with respect to the AN routing table as shown in FIG. 16A.
[0096]
In the situation as described above, for example, the client terminal C1 that wants to receive delivery of a service with Sid = 10 from the server S1 creates a service participation request packet. In the header area of the service participation request packet, UDP is used as the “protocol type (PID)”, the specific number is used as the “destination UDP port number”, and the type information indicating the service participation request packet is used as the “AN packet type”. In addition to being set, header information is set as follows based on the value of each item in the upward direction of the AN routing table.
IP. Dst (destination IP address) = A1
AN. Dst (destination AN address) = S1
AN. Src (source AN address) = C1
AN. Pre (adjacent AN address) = C1
AN. Sid (AN service identifier) = 10
When the service participation request packet in which the header information is set as described above is created, the client terminal C1 issues this service participation request packet from the route a11. Then, this service participation request packet is IP. Based on Dst (= A1), the packet is transferred to the active router A1 via the normal packet R1 ((1)).
[0097]
In the active router A1 that has received this service participation request packet, the processing module of the IP layer notifies the upper layer of the contents because the packet is a packet addressed to the own node. Then, the processing module of the active net (AN) layer registers the entry in the service temporary registration table based on the header information (AN.Dst, AN.Src, AN.Pre, AN.Sid) of the service participation request packet. Do.
[0098]
This temporary service registration table is the same as the service table having the items “destination AN”, “service identifier”, and “adjacent AN (distribution destination)” as shown in FIGS. 15B and 16B. It has a structure. Specifically, in the temporary service registration table, “destination AN” is S1 (= AN.Dst), “service identifier” is 10 (= AN.Sid), and “adjacent AN (distribution destination)”. C1 (= AN.Pre) is registered.
[0099]
When the registration of the entry in the temporary service registration table is completed in this way, the header information of the service participation request packet is as follows based on the value of each item in the uplink direction of the AN routing table (see FIG. 15A). Will be reset to
IP. Dst = A2
AN. Dst = S1
AN. Src = C1
AN. Pre = A1
AN. Sid (AN service identifier) = 10
The active router A1 refers to the AN routing table and issues a service participation request packet with the header information reset as described above from the route a12 (outbound route). This service participation request packet is an IP. Based on Dst (= A2), the packet is transferred to the active router A2 via the normal packet R2 ((2)).
[0100]
In the active router A2 that has received the service participation request packet addressed to its own node, an entry is registered in the service temporary registration table in the same manner as the active router A1. Specifically, S1 (= AN.Dst) is set in “destination AN”, 10 (= AN.Sid) is set in “service identifier”, and A1 (= AN.Pre) is set in “adjacent AN (distribution destination)”. Each is provisionally registered in the service table.
[0101]
Thereafter, the service participation request packet is sequentially transferred from the active router A2 to the active router A3 and the server S1 in the same procedure ((3), (4)). In the process, an entry is registered in the service temporary registration table in the active router A3 as well. As a result, in the active router A3, the service table includes S1 (= AN.Dst) as “Destination AN”, 10 (= AN.Sid) as “Service Identifier”, and “Neighboring AN (Distribution Destination)”. A2 (= AN.Pre) is registered.
[0102]
When the server S1 receives the service participation request packet transferred as described above, the server S1 generates a service participation response packet. In the header area of the service participation response packet, UDP is used as the “protocol type (PID)”, the specific number as the “destination UDP port number”, and the type information indicating the service participation response packet as the “AN packet type”. In addition to being set, header information is set as follows based on the value of each item in the downward direction of the AN routing table.
IP. Dst (destination IP address) = A3
AN. Dst (destination AN address) = A3
AN. Src (source AN address) = S1
AN. Pre (adjacent AN address) = S1
AN. Sid (AN service identifier) = 10
When the service participation response packet in which the header information is set as described above is created, the server S1 issues this service participation response packet from the path S11. Then, this service participation response packet is sent to the IP. Based on Dst (= A3), it is transferred to the active router A3 via the normal router A4 ((5)).
[0103]
The active router A3 that has received the service participation response packet addressed to its own node performs the main registration of the entry in the service table based on the header information (AN.Src, AN.Sid) of the service participation response packet. Specifically, the AN. Src = S1 and AN. The entry temporarily registered as the value of “destination AN” and “service identifier” in Sid = 10 is temporarily registered in the service table as it is. That is, the entries (S1, 10, A2) temporarily registered in the “destination AN”, “service identifier”, and “neighboring AN (distribution destination)” in the service temporary registration table are directly registered in the service table as they are.
[0104]
Thus, when the main registration of the entry to the service table is completed, the header information of the service participation response packet is reset as follows with reference to the service table.
IP. Dst = A2 (value of adjacent AN (distribution destination))
AN. Dst = A2 (value of adjacent AN (distribution destination))
AN. Src = S1
AN. Pre = A3 (AN address of own node)
AN. Sid = 10
The active router A3 refers to the AN routing table and issues a service participation response packet in which the header information has been reset as described above from the route a31 (the outgoing route). This service participation response packet is an IP. Based on Dst (= A2), the packet is transferred to the active router A2 via the normal packet R3 ((6)).
[0105]
In the active router A2 that has received the service participation response packet addressed to its own node, the main registration of the entry to the service table is performed in the same manner as the active router A3. Specifically, the AN. Src = S1 and AN. The entry (S1, 10, A1) temporarily registered as the value of “destination AN” and “service identifier” in the service temporary registration table is searched for Sid = 10, and the entry is directly registered in the service table as it is. .
[0106]
Thereafter, the service participation response request packet is sequentially transferred from the active router A2 to the active router A1 and the client terminal C1 in the same procedure ((7), (8)). In the process, the active router A1 similarly applies the AN. Src = S1, AN. The entries (S1, 10, C1) temporarily registered as the values of “destination AN” and “service identifier” with Sid = 10 are directly registered in the service table as they are.
[0107]
As described above, while the service participation request packet issued from each of the client terminals C1, C2, and C3 is transferred to the servers S1 and S2, the entry is temporarily registered in the service temporary registration table. Each server that has received the participation request packet from each client terminal issues a service participation response packet to the corresponding client terminal. While the service participation response packet is transferred, the entry temporarily registered in the service temporary registration table is registered in the service table.
[0108]
As a result, for example, the entry shown in FIG. 16B is registered (mainly registered) in the service table of the active router A2, and the service table of the active router A1 is shown in FIG. 15B. The entry is registered (main registration). Each entry in this service table, for example, the entry in the service table of the active router A2 shown in FIG.
“Destination AN” = S1
“Service identifier” = 10
“Neighboring AN (distribution destination)” = A1, A4
Indicates that the AN packet related to the service of Sid = 10 distributed from the server S1 is transferred to the active routers A1 and A1. Also entry
“Destination AN” = S2
“Service identifier” = 30
“Neighboring AN (distribution destination)” = A4
Represents that the AN packet relating to the service of Sid = 30 distributed from the server S2 is transferred to the active router A4.
[0109]
As described above, the transfer table (neighboring AN (distribution destination)) of AN packets related to a service (service identifier) distributed from a certain distribution source (request destination from the client terminal: destination AN) is described in the service table. (See FIG. 16 (b)). Therefore, even if the AN address (adjacent AN) of the transfer destination active router corresponding to various services is registered in both the “destination AN” and “adjacent AN” in the downstream direction of the AN routing table (FIG. 16A). The AN packet can be transferred to an appropriate transfer destination active router according to the type of service.
[0110]
For example, when the active router A2 having the AN routing table as shown in FIG. 16A and the service table as shown in FIG. 16B receives an AN packet, the following processing is performed.
[0111]
If it is determined that the received AN packet is an AN packet related to the service 10 (AN.Sid = 10) from the server S1 (AN.Src = S1) based on the header information of the received AN packet, the AN routing table Based on the service table, the AN packet is transferred from the paths a21 and a23 to both the active routers A1 and A4. If it is determined that the received AN packet is an AN packet related to the service 30 (AN.Sid = 30) from the server S2 (AN.Src = S2), based on the AN routing table and the service table, The AN packet is transferred only from the path a23 to the active router A4.
[0112]
As described above, each active router that receives the service participation request packet and the service participation response packet and performs processing (processing in the processing module of the active (AN) layer) in the above-described procedure, specifically, The processing is performed according to the procedure as shown in FIG. This process is performed as a process (S27) performed when other packets (service participation request packet, service participation response packet, etc.) are received in the procedure shown in FIG.
[0113]
In FIG. 17, it is determined whether the AN packet addressed to the own node is a service participation request packet, a service participation response packet, or another AN packet (S31). If it is determined that the AN packet is a service participation request packet, “destination AN” and “service identifier” are the received AN packets in the service table (see FIGS. 15B and 16B). AN. Described in the header area. It is determined whether there is an entry having the value of Dst (destination AN) and the value of the service identifier (AN.Sid) (S32). This determination is based on whether or not the transfer destination (neighboring AN) for the service specified by the service identifier distributed from the server of the destination AN address described in the received AN packet (service participation request packet) has already been registered. Is determined.
[0114]
If such an entry does not exist in the service table, the entry is temporarily registered in the service temporary registration table as described above (S33). Specifically, the AN. The value of Dst is “Destination AN”, and AN. The value of Sid is “service identifier”, and AN. The value of Pre is registered in “adjacent AN (distribution destination)”. By such processing, the entry is registered in the service temporary registration table as described above.
[0115]
When the entry is temporarily registered in the service temporary registration table as described above, the value of the header area of the service participation request packet is reset as described above, and the service participation request packet is transmitted ( S34). By resetting the value of this header area, AN. The value of Pre corresponds to the AN address of the own node, IP. The value of Dst is “destination AN” = AN. The value is changed to the value of “adjacent AN” of the entry Dst.
[0116]
It should be noted that AN. Of the packet received by “Destination AN” and “Service Identifier”. Dst and AN. If an entry having the same value as the value of Sid is already in the service temporary registration table, the service participation request packet is not transmitted (S34) after the processing in S33 is executed.
[0117]
In the service table, “Destination AN” and “Service Identifier” are described in AN. If an entry having the value of Dst (destination AN) and the value of the service identifier (AN.Sid) already exists (YES in S32), the entry is registered in the service table based on the header information of the service participation request packet. Performed (S38). Specifically, in the service table, “destination AN” and “service identifier” are the AN. Dst value and AN. The entry of the service participation request packet AN. The value of Pre is additionally registered.
[0118]
Then, a service participation request packet is created, and the service participation response packet is transmitted to the transmission source of the service participation request packet (S39). The header information of the created service participation response packet is set as follows.
[0119]
AN. Of the received service participation request packet. The value of Src is AN. Dst contains the AN. The value of Dst is AN. Src indicates that the AN address of the own node is AN. Pre. Receives AN. Of the received service participation request packet. The value of Pre is IP. Dst is set respectively.
[0120]
Further, if it is determined in the processing at S31 that the received packet is a service participation response packet, as described above, each entry has already been temporarily registered in the service temporary registration table and the header information of the received service participation response packet is included in the header information. Based on the service table (see FIG. 15B and FIG. 16B), the main entry registration is performed (S35). Specifically, in the service temporary registration table, the “destination AN” is the AN. Src and the “service identifier” is the AN. An entry having a value of Sid is searched. Each entry obtained by this search is fully registered in the service table as it is.
[0121]
When the main registration of the entry to the service table is completed as described above, the header information of the service participation response packet is reset as follows. That is, AN. The value of Pre corresponds to the AN address of the own node, IP. Dst value and AN. The value of Dst is changed to the value of “adjacent AN (distribution destination)” in the service table. Then, the service participation response packet in which the header information is reset (changed) in this way is transmitted (S36).
[0122]
Further, when it is determined in the process at S31 that the received packet is another AN packet, a process corresponding to the type of the packet is performed (S37).
[0123]
Each time the active (AN) layer processing module of each active router A1, A2, A3 receives the AN packet, the processing in the procedure as described above is performed, so that each active router A1, A2, A3 stores the service table. An entry relating to a packet distribution route is registered for each service.
[0124]
According to the above-described example, when an active router is configured by introducing and deploying an existing IP network, there is no provision of a management node for comprehensively managing the location information and route information of each active router. However, when communication is performed between the terminal (client terminal) and the terminal (server), a route setting request packet and a route setting response packet between the terminal and the terminal, and further, a service participation request packet and a service participation response By transferring the packet, it becomes possible to set a route between the terminals.
[0125]
【The invention's effect】
As described above, according to the present invention described in claims 1 to 4, the packet is actually transferred in the network, and the transfer route via the active router is set based on the active router that relayed the packet. Therefore, to realize a route setting method that enables the introduction of an active router without providing a management node in the network for managing and distributing the location information and route information of the active router. Can do.
[0126]
Further, according to the present invention of claims 5 to 7, it is possible to provide a route management method for managing a communication path (packet transfer route) set according to such a route setting method.
[0127]
Furthermore, according to the present invention of claims 8 to 10, it is possible to provide an active router included in a network in which a communication path is set according to such a route setting method.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration example of a network in which an active net is configured.
FIG. 2 is a diagram illustrating a function of each node for realizing an active net.
FIG. 3 is a diagram illustrating a configuration example of an AN packet transferred in a network.
FIG. 4 is a diagram illustrating an example of an AN routing table.
FIG. 5 is a diagram illustrating an example of transfer of a route setting request packet and a route setting response packet.
FIG. 6 is a diagram showing an example of a temporary registration table and an AN routing table in the active router A1.
FIG. 7 is a diagram illustrating an example of a temporary registration table and an AN routing table in the active router A2.
FIG. 8 is a diagram showing an example of a temporary registration table and an AN routing table in the active router A3.
FIG. 9 is a flowchart showing an example of a processing procedure in an IP layer processing module in each active router;
FIG. 10 is a flowchart showing an example of a processing procedure in a processing module of a UDP layer and a processing module of an active net (AN) layer in each active router.
FIG. 11 is a diagram showing another example of the temporary registration table and the AN routing table in the active router A1.
FIG. 12 is a diagram showing another example of the temporary registration table and the AN routing table in the active router A2.
FIG. 13 is a diagram showing another example of the temporary registration table and the AN routing table in the active router A3.
FIG. 14 is a diagram illustrating an example of transfer of a service participation request packet and a service participation response packet.
FIG. 15 is a diagram illustrating an example of an AN routing table and a service table in the active router A1.
FIG. 16 is a diagram showing an example of an AN routing table and a service table in the active router A2.
FIG. 17 is a flowchart illustrating an example of a processing procedure related to service table creation in a processing module of an active net (AN) layer in each active filter.

Claims (10)

A plurality of nodes having a packet transfer function are connected by a route, and a part of the plurality of nodes is configured by an active router having a predetermined processing function together with a packet transfer function. Set a route for transferring packets while performing processing at the active router between the first terminal and the second terminal in a network in which transfer rules that relay packet nodes in communication between them are determined in advance In the route setting method when
The first type of packet is transferred from the first terminal to the second terminal via each node in the network according to the predetermined transfer rule,
Based on the order of the active routers that relay the first type of packets in the process, the active packet to be transferred next from each active router when the packet is transferred from the second terminal to the first terminal A route setting method that determines a router and sets a transfer route of a packet from the second terminal to the first terminal. The route setting method according to claim 1,
The second type packet from the second terminal is relayed to the first terminal via the active router in the network according to the transfer route set as described above,
In the process, the active router to be transferred next is determined from each active router when the packet is transferred from the first terminal to the second terminal based on the order of the active router that relays the second type packet. A route setting method in which a transfer route for packets from the first terminal to the second terminal is set. The route setting method according to claim 1 or 2,
When transferring the first packet from the first terminal to the second terminal, each active router changes the active router immediately before the transfer of the first packet from the second terminal to the first terminal. A route setting method in which a packet is determined as an active router to which a packet is to be transferred next in a packet transfer route toward the packet. The route setting method according to claim 2,
When transferring the second type of packet from the second terminal to the first terminal, each active router transfers the previous active router in the transfer of the second packet from the first terminal to the second terminal. A route setting method in which a packet is determined as an active router to which a packet is to be transferred next in a packet transfer route toward a terminal. In the management method of the transfer route set by the route setting method according to any one of claims 1 to 4,
Each active router assigns the active router determined as the next active router to which the packet is to be transferred next when transferring the packet from the second terminal to the first terminal to the first terminal that is the final transfer destination of the packet. Route management method that manages in correspondence. In the management method of the transfer route set by the route setting method according to any one of claims 1 to 4,
Each active router, when transferring a packet from the first terminal to the second terminal, sets the active router determined as the active router to which the packet is to be transferred next to the second terminal that is the final transfer destination of the packet. Route management method that manages in correspondence. In the management method of the transfer route set by the route setting method according to any one of claims 1 to 4,
Each active router distributes, from the second terminal to the first terminal, the active router determined as the active router to which the next packet should be transferred when the packet is transferred from the second terminal to the first terminal. Route management method corresponding to the type of service to be managed. In an active router provided in a network in which a transfer route is set by the route setting method according to claim 1,
When transferring a packet from the second terminal to the first terminal, the active router determined as the active router to which the packet is to be transferred next is managed in correspondence with the first terminal that is the final transfer destination of the packet An active router having route management means. In an active router provided in a network in which a transfer route is set by the route setting method according to claim 1,
When the packet is transferred from the first terminal to the second terminal, the active router determined as the next active router to which the packet is to be transferred is managed in correspondence with the second terminal that is the final transfer destination of the packet. An active router having route management means. In an active router provided in a network in which a transfer route is set by the route setting method according to claim 1,
When the packet is transferred from the second terminal to the first terminal, the active router determined as the active router to which the packet is to be transferred next is changed to the type of service distributed from the second terminal to the first terminal. An active router that is managed in correspondence.

Images