JP7593501B2 - COMMUNICATION SYSTEM, COMMUNICATION METHOD, AND COMMUNICATION PROGRAM - Google Patents

Description

The present invention relates to a communication system, a communication method and a communication program.

When integrating subnetworks at multiple bases into a single IP subnet via an ISP's (Internet Service Provider's) IP (Internet Protocol) network, it is necessary to build a VPN (Virtual Private Network). VXLAN (Virtual eXtensible Local Area Network) is attracting attention as a technology for integrating subnets within such bases.

VXLAN is a technology that sets up a path of UDP (User Datagram Protocol), an L4 protocol of TCP (Transmission Control Protocol)/IP, between a base station and a gateway that serves as the interface between the IP network, and transfers traffic passing between subnets by encapsulating it in UDP.

IP packets forwarded from one subnet to another are encapsulated into UDP packets at the gateway of the originating subnet. The UDP packets are forwarded to the destination gateway by IP forwarding. The destination gateway receives the packets, deletes the UDP header, and then forwards them to the destination according to the IP address attached to the original IP packet. A tunnel that connects such UDP traffic between two locations is called a UDP tunnel.

When traffic congestion occurs in an ISP network, it leads to increased packet loss and transmission delays. Therefore, multipath transmission technology has been proposed as a technology to avoid traffic congestion.

Multipath forwarding technology is a technique that distributes the destination of packets when there are multiple routes connecting two specific routers in a network so that equal traffic flows between routes with similar priority.

When a packet forwarding device such as a router or switch finds routes with similar priority using a routing protocol that finds a forwarding route for traffic, it selects the route to which the packet should be distributed based on information in the packet header, such as the source address, destination address, and port. The most commonly used method for this selection is to calculate a hash value from the header information and then determine the distribution destination based on the calculated hash value.

D. Dutt, “EVPN in the Data Center”, O’reilly media, pp.11-12

Multipath forwarding can also be performed using a similar method for UDP tunnels in VXLAN. Figure 16 is a diagram explaining multipath forwarding in a UDP tunnel in VXLAN. Figure 17 is a diagram explaining the forwarding packet format.

As shown in Figures 16 and 17, when a UDP tunnel connecting two bases has multiple routes, the router calculates a hash value from information in the UDP packet header and the IP packet header carrying UDP to determine each route. Then, based on the calculated hash value, the router and switches on the route determine the destination route. For example, the hash function used is Hash Function = UDP Src Port%2. % calculates the remainder.

The router outputs the packet to an output port (for example, output Port 0, output Port 1) that is assigned the calculated hash number. When using 5-tuple information to calculate the hash value, the hash value changes by changing the UDP source port ID in the UDP header. This is because the IP addresses of the gateways that are the start and end points of the tunnel are the same regardless of the route, and the UDP destination port ID indicates the protocol number, which is the signature for removing the UDP header at the end gateway.

The ideal load balancing state is one in which the traffic flow between two locations is at its maximum and the bandwidth utilization rate between the links is equal. Figures 18 and 19 are diagrams explaining the load balancing state. Since the traffic flow rate decreases when there is other traffic (Figure 18 (1)), in order to achieve ideal load balancing, it is desirable to distribute the traffic between paths that pass through routes where no link sharing occurs (Figure 18 (2)) and adjust the traffic flow rate between the paths at the shared links of the paths.

If multiple routes to which traffic is to be distributed do not share the same link, multipath forwarding technology can properly distribute the load. With conventional technology, it is possible to investigate the routers that a route passes through by using tools such as traceroute for each route.

However, with conventional technology, it was not possible to detect the same link through which multiple routes pass, or to count the number of routes that pass through the same link. As a result, with conventional technology, it was not possible to guarantee that load balancing was actually achieved across multiple routes.

Specifically, because routers determine the output destination using their own policies, even if there is no link sharing at the beginning of the route ((1) in Figure 19), the route may pass a link through which large volume traffic flows ((2) in Figure 19) or link sharing through the same link may occur ((3) in Figure 19). In particular, in overlay networks, even if routes are assigned to paths in the overlay network and load balancing is performed, or routes are assigned to each path in multiple overlay networks, in the underlay network, routes may pass a link through which large volume traffic flows or link sharing may occur.

The present invention has been made in consideration of the above, and aims to provide a communication system, a communication device, and a communication program capable of detecting routes in an underlay network for multiple distributed candidate paths in an overlay network.

In order to solve the above-mentioned problems and achieve the object, the communication system according to the present invention is a network that integrates multiple networks and forwards packets according to a single address system, and has a network that incorporates an overlay network consisting of overlay paths set up on an underlay network of an address system different from that of the network in question, and when multiple routes are registered as output routes to the same boundary packet forwarding device located on the boundary between the underlay network and another network, the packet forwarding device of the underlay network determines the output route of the packet based on forwarding information held by the packet passing through the overlay path. The system is characterized by having a selection unit that determines the search path to be searched, writes an identifier of its own device in a search packet that passes through the search path to be searched, selects the search path to be searched from among the distributed candidate paths that are overlay paths of distributed candidates in the overlay network, and transmits the search packet that passes through the selected search path; a search unit that searches for devices and links in the underlay network of the search path based on the identifier written in the returned search packet; and a detection unit that detects overlay paths that have the same starting and ending points and that pass through different packet forwarding devices and links in some or all of the routes based on the search results of the search unit for multiple search packets.

According to the present invention, it is possible to detect routes in an underlay network for multiple distributed candidate paths in an overlay network.

FIG. 1 is a diagram illustrating an example of a configuration of a communication system according to an embodiment. FIG. 2 is a diagram for explaining a transfer packet format. FIG. 3 is a diagram illustrating an example of the configuration of the router illustrated in FIG. FIG. 4 is a diagram for explaining the selection of a search path. FIG. 5 is a diagram illustrating the VXLAN format. FIG. 6 is a diagram for explaining a route search of a search path. FIG. 7 is a diagram for explaining route detection. FIG. 8 is a diagram for explaining the calculation of the traffic flow rate. FIG. 9 is a diagram for explaining flow allocation. FIG. 10 is a sequence diagram illustrating a processing procedure of the communication processing according to the embodiment. FIG. 11 is a diagram illustrating another example of the configuration of the communication system according to the embodiment. In FIG. FIG. 12 is a diagram illustrating the second modification of the embodiment. FIG. 13 is a diagram illustrating the second modification of the embodiment. FIG. 14 is a diagram illustrating a third modified example of the embodiment. FIG. 15 is a diagram illustrating an example of a computer that realizes the learning device and the estimation device by executing a program. FIG. 16 is a diagram illustrating multipath forwarding in a UDP tunnel in a VXLAN. FIG. 17 is a diagram for explaining a transfer packet format. FIG. 18 is a diagram illustrating the load distribution state. FIG. 19 is a diagram illustrating the load distribution state.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to this embodiment. In addition, in the description of the drawings, the same parts are indicated by the same reference numerals.

[Embodiment]
In this embodiment, a packet forwarding device (e.g., a router) located at the boundary between the underlay network and other networks sends a search packet to a candidate distribution path in the overlay network, and routers in the underlay network that pass along the way include their own identifiers (e.g., IP addresses) in the search packet.

In the embodiment, based on this search packet, it is possible to detect the path route, such as which router in the underlay network each distribution candidate path passes through, i.e., the routes in the underlay network of multiple distribution candidate paths in the overlay network. In the embodiment, it is possible to detect the same link in the underlay network through which multiple paths pass, and to count the number of paths that pass through the same link in the underlay network among multiple paths.

[Communication system]
Next, a communication system according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the configuration of a communication system according to an embodiment.

As shown in Figure 1, the communication system 100 according to the embodiment is a network that integrates multiple networks and forwards packets according to a single address system, and has a network that incorporates an overlay network consisting of overlay paths set up on an underlay network with an address system different from that of the network in question. A virtually constructed overlay network (subnet 1) is set up on the underlay network.

The underlay network is composed of a network that integrates multiple networks. In the communication system 100, the underlay network is a network in which multiple routers 30 are provided, and which forwards packets according to a single address format. An overlay path in the integrated network is set in a part of the underlay network so as to incorporate an overlay section different from the underlay network, and a flow is assigned to the overlay path.

The communication system 100 uses VXLAN to establish a VPN between two subnets 1, which are overlay networks. Figure 2 is a diagram explaining the forwarding packet format.

Router 20I (IP address: IP _I ) (border packet forwarding device) of subnet 1 located at the boundary between the underlay network and the overlay network encapsulates traffic passing between subnets 1 (see FIG. 2 ) and forwards the packets to router 20E (IP address: IP _E ) (border packet forwarding device). Specifically, router 20I encapsulates packets sent from server 10-1 (IP address: IPsrc1) into UDP packets. The UDP packets are forwarded to router 20E. Router 20E deletes the UDP header from the received packet and then forwards it to the destination server 11-1 (IP address: IPdst1) according to the IP address attached to the original packet.

In the communication system 100, router 20I detects the route of each overlay path to be distributed before actually assigning traffic flows to the overlay paths to be distributed in the overlay network.

Router 20I selects a search path to be searched from among the distribution candidate paths, which are distribution candidates in the overlay network, and transmits a search packet via the selected search path. Based on the multiple search packets returned, router 20I then detects the path route, such as which router in the underlay network the distribution candidate path passes through, i.e., the routes in the underlay network of the multiple distribution candidate paths in the overlay network. Router 20I then calculates the traffic flow rate of the distribution candidate paths and assigns subnet flows to the distribution candidate paths.

In addition, router 30 (packet forwarding device) of the underlay network that passes through on the way has the function of forwarding the packet as well as the function of including its own IP address in the search packet. If multiple output routes to the same boundary packet forwarding device (e.g., router 20E) are registered in router 30, router 30 determines the output route of the packet based on the forwarding information contained in the packet passing through the overlay path.

[Router configuration]
Next, the router 20I will be described. Fig. 3 is a diagram showing a schematic example of the configuration of the router 20I shown in Fig. 1. The router 20E has a similar configuration to the router 20I.

The router 20I shown in Fig. 3 is realized, for example, by loading a specific program into a computer including a ROM (Read Only Memory), a RAM (Random Access Memory), a CPU (Central Processing Unit), etc., and having the CPU execute the specific program. The router 20I has a communication unit 21, a path selection unit 22 (selection unit), a search unit 23, a route detection unit 24 (detection unit), a traffic flow calculation unit 25 (calculation unit), and a flow allocation unit 26 (allocation unit).

The communication unit 21 communicates with other devices wirelessly or via wired connection. The communication unit 51 is a communication interface that transmits and receives various information with other devices connected via a network or the like.

To search for a path in an overlay network, the path selection unit 22 selects a search path to be searched for a route from among multiple distributed candidate paths in the overlay network, and transmits a search packet via the selected search path. The path selection unit 22 selects a search path to be searched for a route from among multiple overlay paths that can be set between a start-end pair of the overlay section based on the packet header of the flow within the integrated network.

Figure 4 is a diagram explaining the selection of a search path. Figure 5 is a diagram explaining the VXLAN format. Figure 6 is a diagram explaining route search for a search path. In the case of the network configuration shown in Figure 4, there are paths P1, P2, and P3 as candidate paths from router 20I to router 20E. Among paths P1, P2, and P3, path selection unit 22 first selects path P1 as a search path, which is a path to be searched. When selecting a search path, path selection unit 22 may select a search path by exhaustive selection, or may randomly select a UDP source port ID number a predetermined number of times. UDP source port ID can take a value from 0 to 2^32 (=65536).

The path selection unit 22 inputs the same 5-tuple as that of path P1 into the search packet that passes through the search path. The path selection unit 22 encapsulates the search packet for the search path for a UDP tunnel. At this time, the path selection unit 22 changes the UDP source port ID in the header. Specifically, the path selection unit 22 puts the path identifier into the UDP source port ID of the UDP header ((1) in Figure 5).

For example, the route of the path of UDP source port ID=1 (path of Port1) is the route of routers 20I, 30a, 30b, and 20E. When searching for this path P1, the path selection unit 22 changes the UDP source port ID in the header to "1" and sets the path as the search packet. Note that the path selection unit 22 does not change the IP address (src/dst), UDP destination port ID, or protocol ID.

The path selection unit 22 forwards the search packet. For example, the route through which the search packet travels is path P1', which passes through routers 20I, 30a, 30b, and 20E and returns to router 20I via the same route. When routers 30a and 30b of the underlay network along the route forward the search packet, they write their own IP addresses in the search packet.

The path selection unit 22 receives the returned search packet and writes the route that the search packet passed through in a path table (not shown) based on the IP address written in the search packet. The path selection unit 22 also transmits and receives the search packet and writes the route in the path table for other distribution candidate paths, for example, path P2 that passes through routers 20I, 30c, 30d, and 20E.

The search unit 23 searches for devices and links in the underlay network of the search path based on the IP address described in the returned search packet. The search unit 23 refers to the path table and searches for transit devices and links in the underlay network of the distribution candidate path for each distribution candidate path to be searched.

The route detection unit 24 detects the path routes, such as which routers in the underlay network each of the distributed candidate paths passes through, based on the search results for the multiple search paths by the search unit 23, i.e., the routes of multiple distributed candidate paths in the overlay network. The route detection unit 24 detects overlay paths that have the same start and end points and that pass through different packet forwarding devices and links in some or all of the routes. Overlay paths that have the same start and end points are subject to packet forwarding by distributing them to multiple routes.

In the communication system 100, the overlay path is VXLAN, and the path to be detected when detecting a group of overlay paths that have different packet forwarding devices and links is a path that is generated when the UDP source port ID of the UDP header of the overlay path is changed. The route detection unit 24 analyzes the path that is generated when the UDP source port ID of the UDP header of the overlay path is changed as the path to be detected when detecting a group of overlay paths that have different packet forwarding devices and links.

As shown in FIG. 7, for example, when the route detection unit 24 finds a path P3 (route passing through routers 20I, 30a, 30b, and 20E) (UDP source port ID 3) that passes through the same route as path P1 (UDP source port ID 1), the route detection unit 24 may suspend the registration of path P3 or exclude path P3 from the candidate distribution paths.

The route detection unit 24 detects the routes of paths, such as which routers in the underlay network each of the distributed candidate paths passes through, i.e., the routes of multiple distributed candidate paths in the overlay network, based on the search results by the search unit 23. The route detection unit 24 calculates the routes that each of the multiple routes passes through, and by comparing each route, detects the same link that multiple routes pass through and counts the number of routes that pass through the same link among the multiple routes.

The traffic flow calculation unit 25 calculates the traffic flow of each distributed candidate path based on the detection results for the route of the distributed candidate path. The traffic flow calculation unit 25 aggregates the search results for passing devices and links, and for the overlay paths detected by the route detection unit 24, counts up, for each passing link, the traffic flow of links in the underlay network when the traffic flow passing through each overlay path is changed. The traffic flow calculation unit 25 calculates the passing traffic flow of each link using the upper limit of the link bandwidth as a constraint.

Figure 8 is a diagram explaining the calculation of traffic flow. For the network shown in Figure 4, the traffic flow calculation unit 25 calculates the passing traffic volume of each link, for example, using linear programming (see Figure 8). The traffic flow calculation unit 25 calculates that the traffic flow of the path with UDP source port ID 1 is 5 Gbps, for example.

The flow allocation unit 26 allocates subnet flows to distribution candidate paths. The flow allocation unit 26 calculates the link bandwidth that can be allocated to an overlay path based on the background traffic volume passing through the link, and allocates flows within the integrated network to be passed through the overlay path so that the calculated bandwidth matches the calculation result. The flow allocation unit 26 uses a predetermined rule to allocate intra-subnet flows to candidate paths so that the total traffic volume of the allocated flows is lower than the traffic flow rate of the candidate path to which the flows are allocated.

Figure 9 is a diagram explaining flow allocation. For example, paths P2 and P4 share some links, so the flow allocation unit 26 allocates traffic volume equally (Figure 9 (1)). Also, path P1 has a large background traffic, so the flow allocation unit 26 allocates a smaller traffic volume to it than to other paths (Figure 9 (2)). Also, in order to minimize the load on the network, the flow allocation unit 26 may allocate intra-subnet traffic flows using an algorithm that solves combinatorial problems such as the knapsack problem or the bin-packing problem.

[Communication Processing Procedure]
Next, a description will be given of a procedure of communication processing in the communication system 100. Fig. 10 is a sequence diagram showing a procedure of communication processing according to an embodiment.

As shown in Figure 10, in communication system 100, router 20I selects a search path to be searched for from among multiple candidate distributed paths in the overlay network (step S1). Router 20I encapsulates a search packet for the search path in a UDP packet (step S2) and puts a path identifier in the UDP source port ID of the UDP header. Router 20I forwards the search packet (step S3).

When routers 30a and 30b in the underlay network along the route forward the search packet, they write their own IP addresses in the search packet (steps S4 to S7). When router 20E receives the search packet, it sends the search packet to router 20I (step S8).

When routers 30a and 30b forward the search packet, they write their own IP addresses in the search packet (steps S9 to S12). Router 20I receives the returned search packet and writes the route that the search packet passed through in the path table based on the IP address written in the search packet (step S13). Router 20I also encapsulates other search paths in UDP packets, forwards them, receives the returned search paths, and writes the routes in the path table.

Router 20I refers to the path table and searches for transit devices and links in the underlay network of each of the specified distribution candidate paths (step S14). Router 20I then detects routes for multiple distribution candidate paths in the overlay network based on the search results by search unit 23 (step S15).

Next, router 20I calculates the traffic flow rate of each distribution candidate path (step S16) and assigns the subnet flow to the distribution candidate path (step S17).

[Effects of the embodiment]
Thus, in an embodiment, a packet forwarding device (e.g., a router) located at the boundary between the underlay network and other networks sends a search packet to a candidate distribution path in the overlay network, and routers in the underlay network that pass along the way include their own IP addresses in the search packet.

In the embodiment, based on this search packet, it is possible to detect the path route, such as which router in the underlay network each distribution candidate path passes through, i.e., the routes of multiple distribution candidate paths in the overlay network. Therefore, in the embodiment, by comparing the routes of each distribution candidate path in the detected overlay network, it is possible to count the presence and number of identical links that each path passes through.

In the embodiment, based on the routes of multiple distribution candidate paths in the detected overlay network, traffic flows are assigned to each distribution candidate path based on the bandwidth of the links shared by the distribution candidate paths and the background traffic of each link. This makes it possible to assign flows to distribution candidate paths taking into account the state of the links to which they are assigned, enabling efficient load balancing.

[Modification 1]
In the embodiment, the router 20I selects a path for search, and performs route detection, traffic flow calculation, and flow allocation based on transmission and reception of the path for search. However, the present invention is not limited to this. Fig. 11 is a diagram showing a schematic diagram of another example of the configuration of the communication system according to the embodiment.

As shown in the communication system 100A of FIG. 11, the routers 20I' and 20E' may only send and receive search paths, and the controller 40 may select a search path and perform route detection, traffic flow calculation, and flow allocation based on the search path received by the router 20I'.

The controller 40 has a communication unit 41 that communicates with other devices wirelessly or via wired connection, a path selection unit 22, a search unit 23, a route detection unit 24, a traffic flow calculation unit 25 and a flow allocation unit 26.

When the controller 40 selects a search path, it instructs router 20I' to send a search packet for this search path with a path identifier in the UDP source port ID of the UDP header. Router 20I' sends the returned search packet to the controller. Based on the search packet received from router 20I', the controller 40 searches for passing devices and links in the underlay network of the distribution candidate path and detects the routes of multiple distribution candidate paths. Based on the detection results for the routes of the distribution candidate paths, the controller 40 performs traffic flow calculations and flow allocation.

[Modification 2]
12 and 13 are diagrams for explaining a second modification of the embodiment. The embodiment can be applied to an overlay technology other than VXLAN, for example, MPLS (Multi-Protocol Label Switching) that adds a label to a packet.

As shown in Figure 12, in the case of MPLS, router 20I puts a path identifier (initial value) into the label of the MPLS header ((1) in Figure 12). Therefore, in the case of MPLS, when detecting a group of overlay paths that pass through different packet forwarding devices and links, the path to be detected is the path that is created when the value of the label in the MPLS header is changed.

As shown in FIG. 13, in the case of MPLS, the label value is rewritten at each router. For example, in the case of MPLS, the label value is rewritten at each router as follows. The initial label value of 34 entered in the label at router 20I is rewritten from 34 to 45 at router 30a, and from 45 to 23 at router 30b, so that the packet arrives at router 20E with a label value of 23. For this reason, when applying MPLS, a field is required to record the initial label value in the overlay tunnel search packet so that router 20I can map the initial label of the path to the passing router.

[Modification 3]
The embodiment can also be applied to GRE (Generic Routing Encapsulation), which is one of the tunneling protocols. Fig. 14 is a diagram for explaining a third modification of the embodiment.

As shown in Figure 14, in the case of GRE, router 20I inserts a path identifier into the Key field of the GRE header ((1) in Figure 14). Note that in the case of GRE, unlike MPLS, the value of the inserted path identifier remains unchanged. For this reason, in the case of GRE, when detecting a group of overlay paths that pass through different packet forwarding devices and links, the paths to be detected are the paths that arise when the path identifier in the Key field of the GRE header is changed in the header of the overlay path.

[System Configuration of the Embodiment]
Each component of the router 20I and the controller 40 is a conceptual representation of its function and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of the functions of the router 20I and the controller 40 is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc.

In addition, each process performed in the router 20I and the controller 40 may be realized, in whole or in part, by a CPU, a GPU (Graphics Processing Unit), and a program analyzed and executed by the CPU and the GPU. In addition, each process performed in the router 20I and the controller 40 may be realized as hardware using wired logic.

Furthermore, among the processes described in the embodiments, all or part of the processes described as being performed automatically can be performed manually. Alternatively, all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, control procedures, specific names, various data, and parameters described above and illustrated in the drawings can be changed as appropriate unless otherwise specified.

[program]
15 is a diagram showing an example of a computer in which a router 20I and a controller 40 are realized by executing a program. The computer 1000 has, for example, a memory 1010 and a CPU 1020. The computer 1000 also has a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

The memory 1010 includes a ROM 1011 and a RAM 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1100. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to a display 1130, for example.

The hard disk drive 1090 stores, for example, an OS (Operating System) 1091, an application program 1092, a program module 1093, and program data 1094. That is, the programs that define the processes of the router 20I and the controller 40 are implemented as program modules 1093 in which code executable by the computer 1000 is written. The program modules 1093 are stored, for example, in the hard disk drive 1090. For example, the program modules 1093 for executing processes similar to the functional configurations in the router 20I and the controller 40 are stored in the hard disk drive 1090. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

In addition, the setting data used in the processing of the above-described embodiment is stored as program data 1094, for example, in memory 1010 or hard disk drive 1090. Then, the CPU 1020 reads out the program module 1093 or program data 1094 stored in memory 1010 or hard disk drive 1090 into RAM 1012 as necessary and executes it.

Note that the program module 1093 and the program data 1094 are not limited to being stored in the hard disk drive 1090, but may be stored in, for example, a removable storage medium and read by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and the program data 1094 may be stored in another computer connected via a network (such as a local area network (LAN) or wide area network (WAN)). The program module 1093 and the program data 1094 may then be read by the CPU 1020 from the other computer via the network interface 1070.

The above describes an embodiment of the invention made by the inventor, but the present invention is not limited to the description and drawings that form part of the disclosure of the present invention according to this embodiment. In other words, other embodiments, examples, operational techniques, etc. made by those skilled in the art based on this embodiment are all included in the scope of the present invention.

100, 100A Communication system 10, 10-1, 10-2, 11, 11-1, 11-2 Server 20I, 20I', 20E, 20E', 30, 30a, 30b, 30c, 30d Router 21, 41 Communication unit 22 Path selection unit 23 Search unit 24 Route detection unit 25 Traffic flow calculation unit 26 Flow allocation unit 40 Controller

Claims (7)

A communication system having a network that integrates a plurality of networks and forwards packets according to a single address system, and that incorporates an overlay network consisting of an overlay path set on an underlay network of an address system different from that of the network,
When a plurality of routes are registered as output routes toward the same boundary packet forwarding device located at the boundary between the underlay network and another network, the packet forwarding device of the underlay network determines an output route of the packet based on forwarding information of the packet passing through the overlay path, and writes an identifier of the own device in a search packet that passes through the search path to be searched,
a selection unit that selects a search path to be searched from among distribution candidate paths that are distribution candidate overlay paths of the overlay network, and transmits the search packet via the selected search path;
a search unit that searches for devices and links in an underlay network of the search path based on an identifier described in the returned search packet;
a detection unit that detects, based on a search result for a plurality of search packets by the search unit, a group of overlay paths that have the same start point and end point and that pass through different packet forwarding devices and links in some or all of the routes;
A communication system comprising: a calculation unit that calculates, for each passing link, a traffic flow rate of a link in the underlay network when a traffic flow rate passing through each overlay path is changed for the overlay paths detected by the detection unit;
an allocation unit that calculates a link bandwidth that can be allocated to an overlay path based on a background traffic volume passing through the link, and allocates an intra-network flow to be passed through the overlay path so as to match the calculation result;
2. The communication system of claim 1, further comprising: The overlay path is a Virtual eXtensible Local Area Network (VXLAN),
The communication system according to claim 1 or 2, characterized in that when detecting a group of overlay paths in which the packet forwarding devices and links through which the packets pass are different, the paths to be detected are paths that arise when the UDP source port ID in the UDP header of the headers of the overlay paths is changed. the overlay path is MPLS (Multi-Protocol Label Switching),
The communication system according to claim 1 or 2, characterized in that when detecting a group of overlay paths that pass through different packet forwarding devices and links, the path to be detected is a path that arises when the value of a label in the MPLS header label in the header of the overlay path is changed. the overlay path is a Generic Routing Encapsulation (GRE),
The communication system according to claim 1 or 2, characterized in that when detecting a group of overlay paths through which the packet forwarding devices and links passing through are different, the paths to be detected are paths that arise when a path identifier in a Key field of a GRE header in the header of the overlay path is changed. A communication method executed in a communication system having a network that integrates a plurality of networks and forwards packets according to a single address system, the network incorporating an overlay network consisting of an overlay path set on an underlay network of an address system different from that of the network, comprising:
When a plurality of routes are registered as output routes to the same boundary packet forwarding device located at the boundary between the underlay network and another network, the packet forwarding device of the underlay network determines an output route of the packet based on forwarding information of the packet passing through the overlay path, and writes an identifier of the own device in a search packet that passes through the search path to be searched;
A step in which a device disposed in the communication system selects a search path to be searched from among distribution candidate paths that are distribution candidate overlay paths of the overlay network, and transmits the search packet via the selected search path;
A step of searching for devices and links in an underlay network of the search path by a device located in the communication system based on an identifier described in the returned search packet;
a step of a device disposed in the communication system detecting, based on a search result for the plurality of search packets, a group of overlay paths having the same start point and end point, the group of overlay paths passing through which are different packet forwarding devices and links in a part or all of the routes;
A communication method comprising: A communication program for causing a computer to execute a method,
A communication system having a network that integrates a plurality of networks and forwards packets according to a single address system, and that incorporates an overlay network consisting of an overlay path set on an underlay network of an address system different from that of the network, comprising: a computer as a packet forwarding device of the underlay network;
When a plurality of routes are registered as output routes toward the same boundary packet forwarding device located at the boundary between the underlay network and another network, the device executes a step of determining an output route of a packet based on forwarding information of the packet passing through the overlay path, and writing an identifier of the device itself in a search packet passing through the search path to be searched;
A computer as a device disposed in the communication system,
selecting a search path to be searched from among distribution candidate paths that are distribution candidate overlay paths of the overlay network, and transmitting the search packet via the selected search path;
A step in which a device located in the communication system searches for devices and links in an underlay network of the search path based on an identifier described in the returned search packet;
detecting, based on a search result for the plurality of search packets, a group of overlay paths having the same start point and end point, the group of overlay paths having different packet forwarding devices and links passing through in a part or all of the routes;
A communication program characterized by causing the program to execute the above steps.

Images