JP2008054214A - Network virtual system and network virtual program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To optimize traffic on a virtual network. <P>SOLUTION: A network virtual system comprises multiple clients 20 connected to one of underlay networks 40 (#1 to #6) constructed in advance and a network virtual unit 155 for constructing virtual networks on the underlay networks through the use of software. Furthermore, the network virtual unit has a virtual hub 10 for communication relay between clients in the constructed virtual networks d1 to d6. Considering a predetermined connection standard, the virtual hub judges whether a client pair of multiple clients requiring communication relay via the virtual hub should communicate directly with each other without relay via the hub, and if it is determined that direct communication should take place, directs clients of this pair to communicate with each other directly via peer-to-peer links S1 to S5. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a network virtualization system for constructing an overlay network such as virtual Ethernet (registered trademark) on an underlay network such as TCP / IP, and a program applied to the network virtualization system. In particular, the present invention relates to a network virtualization system and a network virtualization program for optimizing overlay network traffic.

  For example, a virtual network constructed by software on an existing physical network (hereinafter referred to as “underlay network”) such as a TCP / IP network is referred to as an overlay network.

  As main overlay networks (hereinafter also referred to as “virtual networks”), what is called a peer-to-peer (P2P) network (for example, Non-Patent Documents 1 to 7) and virtual local area networks (hereinafter “ Referred to as “virtual LAN”).

  Peer-to-peer networks are currently widely used for file exchange (Winny, BitTorrent, etc.), voice communication (Skype, etc.) and the like. Depending on the method, the following two major steps are taken: (1) Host and content search depend on special servers, but final data transmission / reception is performed between peer-to-peer hosts, and (2) data In addition to transmission / reception, the host and content search itself can be divided into those that do not require a special server and are distributed and coordinated among hosts in completely equal positions. In some cases, the above two intermediate methods using a host group that plays a special role are used for searching hosts and contents, although they are not specific servers.

  On the other hand, a typical virtual LAN includes Softether (Packetix) (for example, Non-Patent Documents 8 to 9) manufactured by SoftEther Corporation. Softeter (Packetix) is a software group for constructing a virtual Ethernet which is an example of a virtual network. For example, a virtual Ethernet construction method using Softeter (Packetix) or similar software is as follows.

  That is, first, software that emulates an Ethernet switching hub (hereinafter referred to as “virtual hub”) is operated on a certain computer (hereinafter, the computer that operates the virtual hub is referred to as “server”).

  Next, a computer wishing to participate in virtual Ethernet connects to a virtual hub (hereinafter referred to as a virtual network interface) via a virtual network interface (hereinafter referred to as “virtual network interface”) constructed by software. The computer that operates is called “client”).

  A general configuration of such a virtual Ethernet is shown in FIG.

  For example, the underlay network 150 that is a TCP / IP network is a physical network that serves as a basis for constructing the virtual Ethernet 154 that is an overlay network. The underlay network 150 includes a firewall 156, a router 157, and a link layer communication device (a hub or a switching hub in the case of Ethernet). The router 157 is a network layer device that constitutes a TCP / IP network, and is a general device, and therefore, detailed description thereof is omitted here. The firewall 156 is a security device installed at the boundary between the internal network 160 and the external network 161, and since it is a general device, its detailed description is omitted here.

The client 153 is a computer that is connected to the virtual hub 151 that constitutes the virtual Ethernet 154, and is connected to the virtual hub 151 via the virtual network interface 152. In this way, the virtual hub 151 functions as a relay node that holds connections with a plurality of clients 153. The server 155 is a network virtualization apparatus in which such a virtual hub 151 is arranged. In this specification, the server 155 that is such a network virtualization apparatus and the client 153 that is a terminal controlled by the network virtualization apparatus are collectively referred to as a network virtualization system. The virtual Ethernet 154 is constructed by connecting a plurality of clients 153 each having a virtual network interface 152 to one or a plurality of virtual hubs 151.
The Institute of Electronics, Information and Communication Engineers, September 2004, “P2P General I: Challenge of Brokerless Model” The Institute of Electronics, Information and Communication Engineers, October 2004, "P2P General Remarks II P2P Technology" IEICE Journal, December 2004, "P2P General Review III P2P Service and Business" IEICE Journal, January 2005, "P2P General Remarks IV Latest Trends and Future Prospects" ASCII, Inc., UNIX (registered trademark) MAGAZINE, September 2005, "Basic knowledge of P2P technology [1]" ASCII Co., Ltd., UNIX MAGAZINE, October 2005, “Basic knowledge of P2P technology [2]” ASCII Co., Ltd., UNIX MAGAZINE, November 2005, "Basic knowledge of P2P technology [3]" ASCII Corporation, Official SoftEther Usage Guide http: // www. softther. com, SoftEther Corporation website

In such a conventional virtual Ethernet 154, all traffic between the clients 153 belonging to the virtual Ethernet 154 passes through the virtual hub 151, that is, the server 155, and between the clients 153 belonging to the virtual Ethernet 154, or between the client 153 and the server. The positional relationship with 155 does not take into account the topology of the underlay network 150. Therefore,
1. It is difficult to ensure the scalability of the virtual hub 151.
2. The load on the underlay network 150 increases, and the capital investment cost increases.
3. Management costs such as installation of a plurality of virtual hubs 151 and switching of clients 153 between the plurality of virtual hubs 151 increase.
The problem arises.

  The present invention has been made in view of such circumstances, and clients that make up a virtual network such as virtual Ethernet consider the positional relationship on the underlay network to which they belong and the traffic volume between clients. Therefore, it is possible to directly communicate between clients without going through a relay node, that is, a server, and when necessary, in addition to that, communication through a relay node is also used to optimize traffic on the virtual network. An object of the present invention is to provide a network virtualization system and a network virtualization program that can be realized.

  In order to achieve the above object, the present invention takes the following measures.

  That is, the invention of claim 1 is a network virtualization apparatus for constructing a virtual network by software on a plurality of terminals connected to any of a plurality of physical networks constructed in advance and a plurality of physical networks. And. The network virtualization apparatus further includes a relay node that relays communication between terminals in the constructed virtual network. Whether the relay node should directly communicate with each other without relaying the relay node with respect to any one pair of terminals relayed by the relay node in consideration of a predetermined connection standard. In the case where it is determined that direct communication is to be performed, the first instruction means for instructing the pair of terminals to directly communicate with each other without relaying the relay node is provided.

  According to a second aspect of the present invention, in the network virtualization system of the first aspect of the invention, when communication is relayed by a relay node, each terminal communicates with another specific terminal in consideration of connection criteria. If it is determined whether to communicate directly with each other without relaying the relay node, and it is determined that direct communication is to be performed, communication with a specific terminal is performed directly without relaying the relay node. First transition means for transitioning to (1) is provided.

  According to a third aspect of the present invention, in the network virtualization system according to the first or second aspect of the present invention, the connection criterion is the amount of traffic relayed through a relay node between a pair of terminals, on the physical network of the pair of terminals It is at least one of localization in the topology, traffic protocol type, and load status of the relay node.

  According to a fourth aspect of the present invention, there is provided the network virtualization system according to any one of the first to third aspects, wherein each terminal is connected so as to be communicable with the relay node; In accordance with an instruction from one instruction means, a link establishment means for establishing a peer-to-peer link with a terminal that is a direct communication partner is provided.

  According to a fifth aspect of the present invention, in the network virtualization system according to any one of the first to fourth aspects, the relay node communicates with each other directly in consideration of a predetermined disconnection criterion. The second communication is instructed to disconnect the direct communication and shift to the relay relayed communication to the pair of terminals when it is determined that the direct communication is to be disconnected. Instruction means are provided.

  A sixth aspect of the present invention is the network virtualization system according to any one of the first to fifth aspects of the present invention, wherein each terminal considers a disconnection criterion when communicating directly with another specific terminal. Determining whether or not to disconnect the direct communication, and when it is determined to disconnect, includes a second transition unit that disconnects the direct communication with the specific terminal and shifts to the relay relay communication. .

  According to a seventh aspect of the present invention, in the network virtualization system according to the fifth or sixth aspect of the present invention, the disconnection criteria are the amount of traffic due to direct communication between a pair of terminals, and the topology on the physical network of the pair of terminals. It is at least one of localization, traffic protocol type, and relay node load status.

  The invention of claim 8 is a program applied to the network virtualization system of the invention of claim 1.

  A ninth aspect of the invention is a program applied to the network virtualization system of the third aspect of the invention.

  The invention of claim 10 is a program applied to the network virtualization system of the invention of claim 5.

  The invention of claim 11 is a program applied to the network virtualization system of the invention of claim 7.

  According to the present invention, clients composing a virtual network communicate directly with each other without going through a relay node in consideration of the positional relationship on the underlay network to which the virtual network belongs and the traffic volume between clients. If necessary, it is possible to realize a network virtualization system and a network virtualization program capable of optimizing traffic on the virtual network by using communication via a relay node in addition to that. It becomes possible.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  In addition, the code | symbol in the figure used for description of each following embodiment attaches | subjects and shows the same code | symbol about the same part as FIG.

(First embodiment)
FIG. 1 is a functional block diagram showing a configuration example of a virtual hub 10 provided in a server constituting the network virtualization system according to the present embodiment.

  FIG. 2 is a functional block diagram showing a configuration example of the client 20 configuring the network virtualization system according to the present embodiment.

  FIG. 3 is a flowchart showing the flow of peer-to-peer link generation processing, deletion processing, and Ethernet frame transmission processing.

  FIG. 4 is a network configuration diagram (initial state) showing an example of a virtual Ethernet configured by the network virtualization system according to the present embodiment.

  FIG. 5 is a network configuration diagram (when traffic occurs) showing an example of a virtual Ethernet configured by the network virtualization system according to the present embodiment.

  FIG. 6 is a network configuration diagram (peer-to-peer communication state) showing an example of a virtual Ethernet configured by the network virtualization system according to the present embodiment.

  In the present embodiment, as shown in FIG. 6, each client 20 (# 1 to # 6) is specified separately from the connection link (virtual networks d1 to d6) to the virtual hub 10 provided in the server 30. Direct communication (hereinafter referred to as “peer-to-peer communication”) between specific clients 20 is realized by generating and maintaining separate communication links (hereinafter referred to as “peer-to-peer links”) S1 to S5 between the clients 20 To do.

  Such an embodiment will be described in detail below.

  As shown in FIG. 1, the virtual hub 10 includes an Ethernet switching hub 11, a forwarding table 12, a connection link control unit 13, an underlay network protocol group 14, a physical network interface 15, and a peer-to-peer link control unit 16. It has.

  The Ethernet switching hub 11 performs the same processing as a so-called general Ethernet switching hub. Therefore, further detailed description is omitted here.

  The forwarding table 12 is a table that holds mapping between a destination MAC address of an Ethernet frame and an output destination port. This is generally used for Ethernet frame transfer processing by the Ethernet switching hub 11. Therefore, further detailed explanation is omitted here.

  The connection link control unit 13 is a part that holds and controls the connection (link) with the client 20 or another virtual hub 10 in the underlay network 40, and the Ethernet frame received from the Ethernet switching hub 11 is appropriately underlaid. It is sent to the link of the network 40. Further, the data received from the underlay network 40 is passed to the Ethernet switching hub 11.

  The underlay network protocol group 14 is generally a higher-level protocol that uses them, such as TCP / IP and HTTP.

  The physical network interface 15 is a physical network interface used for actual communication.

  The peer-to-peer link control unit 16 sends data to the client 20 based on the traffic volume between the connection links (between the clients 20), the traffic protocol, the positional relationship information of the client 20 in the underlay network 40, and the like. Instructs peer-to-peer link creation and deletion.

  Specifically, the connection link control unit 24 (see FIG. 2) of the client 20 generates a peer-to-peer link / shifts to peer-to-peer communication, or deletes a peer-to-peer link / shifts to communication via the virtual hub 10. Instruct. In order to determine these transitions, the traffic amount between the clients 20 and the protocol type information are acquired from the Ethernet switching hub 11, and the address (IP address) of the client 20 in the underlay network 40 is acquired from the connection link control unit 13. , Location information) or information on the number of accommodated clients 20, and CPU usage rate and memory usage rate information are obtained from an OS (not shown).

  As shown in FIG. 2, the client 20 includes an application / upper protocol group 21, an underlay network, in addition to a virtual network interface 22 including an Ethernet frame transmission / reception unit 23, a connection link control unit 24, and a transmission destination management table 25. A protocol group 26 and a physical network interface 27 are provided.

  The application / upper protocol group 21 is a user application that uses the virtual network interface 22 and upper protocols (for example, TCP / IP, HTTP, etc.).

  The Ethernet frame transmission / reception unit 23 stores the data from the application / upper protocol group 21 in the Ethernet frame and sends it to the connection link control unit 24. In addition, the Ethernet header is removed from the data received from the connection link control unit 24 and passed to the application / upper protocol group 21.

  The connection link control unit 24 is a part that holds and controls the connection (link) with the virtual hub 10 or another client 20 in the underlay network 40, and sends an Ethernet frame to a link of the appropriate underlay network 40. To do. In addition, the data received from the underlay network 40 is transferred to the Ethernet frame transmission / reception unit 23.

  The transmission destination management table 25 is a table that holds a mapping between a peer-to-peer link and the MAC address of the peer of the link. As shown in FIGS. 9 to 14 to be described later, the Ethernet frame addressed to the peer registered in the transmission destination management table 25 is transferred to the corresponding peer-to-peer link, and the other destination Ethernet frame, broadcast frame, and multicast frame are connected to the virtual hub 10. Is output to the link.

  The underlay network protocol group 26 is generally TCP / IP and a higher-level protocol (for example, HTTP) that uses them.

  The physical network interface 27 is a physical network interface used for actual communication.

  Next, a processing flow such as generation and deletion of a peer-to-peer link will be described with reference to FIGS. 3A, 3B and 3C, and FIGS.

  4 to 6 are diagrams illustrating an example of a virtual Ethernet that is a virtual network configured by the network virtualization system according to the present embodiment. The network virtualization system according to the present embodiment includes a server 30 having a virtual hub 10 configured as shown in FIG. 1 and a client 20 configured as shown in FIG.

  In the initial state of the virtual Ethernet configured by such a network virtualization system, as shown in FIG. 4, the clients 20 (# 1 to # 6) are connected to the virtual hub 10 and all the clients 20 (# 1) are connected. To # 6) transmit / receive Ethernet frames via the virtual hub 10. The segments (usually IP network segments) 40 (# 1 to # 6) of the underlay network 40 are normally connected by routers (not shown in FIGS. 4 to 6). The virtual hub 10 belongs to the segment 40 (# 3), the clients 20 (# 1 to # 3) belong to the segment 40 (# 1), and the clients 20 (# 4 to # 6) belong to the segment 40 (# 5). ing.

  Here, for example, the client 20 (# 3) and the client 20 (# 4) belong to the same segment from the viewpoint of virtual Ethernet, but belong to different segments from the viewpoint of the underlay network 40. And it is located remotely.

  Here, as shown in FIG. 5, for example, the amount of traffic between specific clients 20 increases, for example, the client 20 (# 4) and the client 20 (# 5), or the client 20 (# 3) and the client 20 Assume that communication has occurred between clients 20 located remotely in an underlay network, such as 20 (# 6).

  In the former case, when the peer-to-peer link control unit 16 of the virtual hub 10 determines that the traffic volume between the client 20 (# 4) and the client 20 (# 5) exceeds a predefined threshold, the client 20 Instruct (# 4) and client 20 (# 5) to create a peer-to-peer link between them and transition to peer-to-peer communication.

  The connection link control unit 24 of the client 20 (# 4) and the client 20 (# 5) that has received the above instruction, as shown in the procedure a1 of FIG. A communication path (peer-to-peer link) is generated using a connection, an SSL session, etc. (procedure a2), and the client 20 (# 4) uses the virtual Ethernet MAC address of the client 20 (# 5) as the client 20 (# 5) registers the MAC address of the client 20 (# 4) in the transmission destination management table 25 (procedure a3).

  Thereafter, the connection link control unit 24 of the client 20 (# 4) sends an Ethernet frame (having the MAC address of the client 20 (# 5) as the destination MAC address) for the client 20 (# 5) to the peer-to-peer link. Send out (procedures c 1, c 2, c 3 [NO], c 4 [Yes], c 5 in FIG. 3C), and send Ethernet frames having other destination MAC addresses (unicast) to the link with the virtual hub 10 (Procedure c3 [No], procedure c4 [No], procedure c6). Broadcast frames and multicast frames are also sent to the link with the virtual hub (procedures c3 [Yes] and c6).

  Similarly, the client 20 (# 5) sends an Ethernet frame for the client 20 (# 4) (having the MAC address of the client 20 (# 4) as the destination MAC address) to the peer-to-peer link, and otherwise The Ethernet frame having the destination MAC address, the broadcast frame, and the multicast frame are transmitted to the link with the virtual hub 10.

  As a result, traffic between the client 20 (# 4) and the client 20 (# 5) does not pass through the virtual hub 10, and the load on the virtual hub 10 and the underlay networks 40 (# 4), 40 The load of (# 5) is reduced.

  As shown in FIG. 5, when the same processing is performed between the client 20 (# 3) and the client 20 (# 6), the client 20 (# 3) and the client 20 (# 6) 10, when the underlay network 40 (# 2), 40 (# 3), 40 (# 4) is used, the underlay network 40 (# 6) is used as shown in FIG. ) Through peer-to-peer communication using only the peer-to-peer link S5, and consequently the load between the virtual hub 10 and the underlay networks 40 (# 2), 40 (# 3), and 40 (# 4) Is reduced.

  Further, when the peer-to-peer link control unit 16 of the virtual hub 10 determines that the clients 20 (# 1) to 20 (# 3) are located in the topological vicinity of the underlay network 40 (# 1), these clients 20 (# 1) # 1, # 2, and # 3) are also instructed to shift to peer-to-peer communication. As a result, peer-to-peer links S1 to S3 are generated between the client 20 (# 1) and the client 20 (# 3), and the communication between these clients 20 (# 1, # 2, # 3) is underlay network 40. Localized at (# 1), the load on the other underlay networks 40 (# 2 to # 6) and the virtual hub 10 is reduced.

The following can be considered as a criterion for determining whether or not a certain group of clients 20 is located in the topological vicinity of the underlay network 40.
The address of the underlay network 40 of the client 20
-Round trip time to client 20.
The number of hops in the underlay network 40 to the client 20
The cost of routing the underlay network 40 to the client 20;

  FIG. 6 shows an example of a state in which peer-to-peer communication between the clients 20 is finally established by the above processing. Solid lines S1 to S5 indicate peer-to-peer communication links, and broken lines d1 to d6 indicate communication links via the virtual hub 10.

  In the above description, the peer-to-peer link generation and the transition to peer-to-peer communication led by the virtual hub 10 have been described. It is also possible to implement a transition to peer-to-peer communication. In this case, the client 20 that has performed the corresponding process can perform subsequent processing by the virtual hub 10 (for example, deletion of a peer-to-peer link). Need to be notified that a peer-to-peer link has been created and peer-to-peer communication has begun.

  Note that the determination that the client 20 independently generates the peer-to-peer link may be the same as the determination criterion of the virtual hub 10. On the other hand, when the traffic between specific clients 20 falls below a predefined threshold, the virtual hub 10 takes the lead, or the client 20 deletes a peer-to-peer link with the specific client 20 via the virtual hub 10. It is also possible to carry out a transition to communication (return to the original state). In this case, the processing on the client 20 side is triggered by peer-to-peer triggered by reception of an instruction from the virtual hub 10 or when its own deletion condition is satisfied, as shown in steps b1 to b3 shown in FIG. The link is deleted, and the MAC address of the client 20 is deleted from the transmission destination management table 25. When the virtual hub 10 performs the initiative, each client 20 needs to regularly notify the virtual hub 10 of information on its own traffic amount.

  As described above, in the network virtualization system according to the embodiment of the present invention, traffic concentration to a specific virtual hub 10 and high load can be avoided by the above-described operation. As described above, scalability of the virtual hub 10 and the entire virtual network can be ensured. Further, since traffic does not pass through the specific virtual hub 10, it is possible to reduce the load on the underlay network 40, thereby suppressing the capital investment of the underlay network 40. Further, direct communication between the clients 20 can improve the communication throughput and response time between the clients 20.

(Second Embodiment)
In the present embodiment, the virtual network interface 22 on the client 20 in the first embodiment has the same function as the virtual hub 10. Therefore, here, differences from the first embodiment will be described.

  In the present embodiment, as shown in FIG. 7, the client 20 has the same function as the Ethernet switching hub 11 held by the virtual hub 10 in the virtual network interface 70 used to connect to the virtual hub 10. By providing the Ethernet switching hub 76, direct communication between the specific clients 20 is realized.

  In FIG. 7, the Ethernet switching hub 76 performs an Ethernet frame transfer process based on the destination MAC address in the same manner as a general so-called Ethernet switching hub 11. Here, as a part greatly different from the first embodiment, from the Ethernet switching hub 76, the Ethernet frame transmitting / receiving unit 73, the virtual hub 10, and the other clients 20 are all recognized as “equivalent connection destinations”. It is.

  That is, the Ethernet frame transmitted from the Ethernet frame transmitting / receiving unit 73 is either the connection link A for the virtual hub 10 or the connection link B for the other client 20 by the Ethernet switching hub 76 based on the destination MAC address. Is output. On the other hand, an Ethernet frame addressed to the own station (corresponding client 20) arriving from the outside, such as the virtual hub 10 or another client 20, is transmitted to the Ethernet frame transmission / reception unit 73 by the Ethernet switching hub 76 based on the destination MAC address. It is only transferred to the connection link C.

  Although not useful in the present embodiment, depending on the connection state between the clients 20, an Ethernet frame between the clients 20 other than the client 20 can be obtained by the function of the Ethernet switching hub 76 of the virtual network interface 70 of a certain client 20. It is also functionally possible to transfer. Here, it is assumed that the same traffic state occurs in the same configuration as in the first embodiment, and the peer-to-peer link control unit 16 of the virtual hub 10 issues a similar instruction to each client 20.

  As a result, each client 20 accommodates connections from other clients 20 in the Ethernet switching hub 76 held by its own virtual network interface 70. The overall connection state in that case is shown in FIG. 8, and the Ethernet switching hubs 76 (# 1 to # 6) held by the virtual network interfaces 70 (# 1 to # 6) of the respective clients 20 (# 1 to # 6). The contents of the transfer table 75 (# 1 to # 6) are shown in FIGS.

  In this case, as shown in FIG. 8, since a loop is formed between the virtual network interfaces 70, some method is necessary to avoid a loop of Ethernet frame transfer. Therefore, in this embodiment, a transfer table 75 having a data configuration as shown in FIGS. 9 to 14 is provided, and the transfer loop is avoided by devising the setting of the transfer table 75 and the transfer policy.

  9 shows the client 20 (# 1), FIG. 10 shows the client 20 (# 2), FIG. 11 shows the client 20 (# 3), FIG. 12 shows the client 20 (# 4), and FIG. FIG. 14 is a data configuration diagram showing an example of the transfer table 75 of the Ethernet switching hub held by the client 20 (# 6).

  That is, in the transfer table 75, in addition to the destination MAC address 75b and the destination connection link 75c, the source connection link 75a can be set or learned. Then, the destination MAC address 75b of the Ethernet frame to be transferred hits the address field of the destination MAC address 75b of an entry in the forwarding table 75, and the connection link that received the Ethernet frame is the source connection link 75a of the entry. Only when it matches with what is described in the field, the Ethernet frame is output to the transmission destination connection link 75c of the entry. Otherwise, the Ethernet frame is discarded.

  On the other hand, as another means, operating the spanning tree protocol between the virtual hub 10 and the Ethernet switching hub 76 in the virtual network interface 70 of each client 20 is also an effective loop avoidance means.

  As described above, the virtual network interface 22 on the client 20 in the first embodiment has the same function as the virtual hub 10 as in the network virtualization system according to the embodiment of the present invention. Also, it is possible to achieve the same operational effects as those of the first embodiment.

  In the first and second embodiments described above, as a transition condition from communication via the virtual hub 10 to peer-to-peer communication between specific clients 20, or from peer-to-peer communication between specific clients 20 to a virtual hub. As an example of the transition condition to the communication via the network 10, the example in which the traffic amount between specific clients 20 and the positional relationship of each client 20 on the underlay network 40 are considered has been described. However, the transition conditions are not limited to these, and other than that, for example, as described below, 1) the load status of the virtual hub, 2) the protocol type of communication between clients, and 3) the virtual Possible hub failure.

1) Virtual hub load status
When the load of the virtual hub 10 exceeds a predefined threshold, the virtual hub 10 selects a specific pair of clients 20 from the group of clients 20 accommodated by the virtual hub 10 based on some condition, and sends them to peer-to-peer communication. Instruct to migrate.

  On the other hand, when the load of the virtual hub 10 falls below a predefined threshold, the virtual hub 10 selects a specific pair of clients 20 from a group of clients 20 accommodated by the virtual hub 10 based on some condition, and selects them. An instruction to shift to communication via the virtual hub 10 is given. The load of the virtual hub 10 is not limited, but may be, for example, the total traffic amount, CPU usage rate, memory usage rate, number of accommodated clients, and the like.

2) Protocol type for communication between clients
The virtual hub 10 instructs the clients 20 to shift to peer-to-peer communication, for example, when communication of a predefined protocol is started between specific clients 20 or continued for a certain period of time. Conversely, when communication with a predefined protocol is terminated with a specific client 20 or when it has been interrupted for a certain time, the client 20 is instructed to shift to communication via the virtual hub 10. In this case, the client 20 needs to periodically notify the virtual hub 10 of information such as its own traffic content. Alternatively, the client 20 can independently perform the same process for a predefined protocol.

3) Virtual hub failure
The virtual hub 10 instructs the client 20 to shift to peer-to-peer communication when a failure occurs in itself and the processing itself stops. On the other hand, when the virtual hub 10 recovers from the failure, the virtual hub 10 instructs the client 20 accommodated before the failure to shift to communication via the virtual hub 10. In addition, when the client 20 detects a failure of the virtual hub 10 that accommodates itself by some means such as keep alive processing (heartbeat processing), the client 20 can transition to peer-to-peer communication by itself. When recovery from a failure of the accommodated virtual hub 10 is detected, it is possible to shift to communication via the virtual hub 10.

  The best mode for carrying out the present invention has been described above with reference to the accompanying drawings, but the present invention is not limited to such a configuration. Within the scope of the invented technical idea of the scope of claims, a person skilled in the art can conceive of various changes and modifications. The technical scope of the present invention is also applicable to these changes and modifications. It is understood that it belongs to.

(Applied technology field)
The present invention can be applied to a wide range including not only a virtual network but also a technology for constructing a virtual network in which traffic is concentrated on a certain node and a technology for constructing an overlay network.

The functional block diagram which shows the structural example of the virtual hub with which the server which comprises the network virtualization system which concerns on 1st Embodiment is equipped. The functional block diagram which shows the structural example of the client which comprises the network virtualization system which concerns on 1st Embodiment. The flowchart which shows the flow of the production | generation process of a peer to peer link. The flowchart which shows the flow of the deletion process of a peer to peer link. The flowchart which shows the flow of an Ethernet frame transmission process. The network block diagram (initial state) which shows an example of the virtual Ethernet comprised by the network virtualization system which concerns on 1st Embodiment. 1 is a network configuration diagram (when traffic occurs) showing an example of a virtual Ethernet configured by a network virtualization system according to a first embodiment. The network block diagram (peer to peer communication state) which shows an example of the virtual Ethernet comprised by the network virtualization system which concerns on 1st Embodiment. The functional block diagram which shows the structural example of the client which comprises the network virtualization system which concerns on 2nd Embodiment. The network block diagram which shows an example of the virtual Ethernet comprised by the network virtualization system which concerns on 2nd Embodiment. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which client 20 (# 1) hold | maintains. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which a client 20 (# 2) hold | maintains. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which a client 20 (# 3) hold | maintains. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which a client 20 (# 4) hold | maintains. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which a client 20 (# 5) hold | maintains. The data block diagram which shows an example of the transfer table of the Ethernet switching hub which a client 20 (# 6) hold | maintains. The figure which shows the general structural example of virtual Ethernet.

Explanation of symbols

  A, B, C: Connection link, d1-d6 ... Virtual network, S1-S5 ... Peer-to-peer link, 10 ... Virtual hub, 11 ... Ethernet switching hub, 12 ... Forwarding table, 13 ... Connection link control unit, 14 ... Underlay Network protocol group, 15 ... Physical network interface, 16 ... Peer to peer link control unit, 20 ... Client, 21 ... Application / upper protocol group, 22 ... Virtual network interface, 23 ... Ethernet frame transmission / reception unit, 24 ... Connection link control unit, 25 ... Destination management table, 26 ... Underlay network protocol group, 27 ... Physical network interface, 30 ... Server, 40 ... Underlay network, 70 ... Virtual network interface, 73 ... Ethernet Frame transmission / reception unit, 75 ... forwarding table, 76 ... Ethernet switching hub, 150 ... Underlay network, 151 ... Virtual hub, 152 ... Virtual network interface, 153 ... Client, 154 ... Virtual Ethernet, 155 ... Server, 155 ... Network virtualization Device, 156 ... Firewall, 157 ... Router, 160 ... Internal network, 161 ... External network

Claims (11)

A plurality of terminals connected to any of a plurality of pre-built physical networks;
A network virtualization apparatus that constructs a virtual network by software on the plurality of physical networks;
The network virtualization apparatus further includes a relay node that relays communication between the terminals in the constructed virtual network,
Whether the relay node should directly communicate with each other without relaying the relay node for any pair of terminals relayed by the relay node in consideration of a predetermined connection criterion A network including first instruction means for instructing the pair of terminals to directly communicate with each other without relaying the relay node. Virtualization system. The network virtualization system according to claim 1,
Whether the terminals should communicate directly with each other without relaying the relay node in consideration of the connection criteria when the communication is relayed by the relay node. And determining whether or not direct communication is to be performed, the network virtual network includes a first transition unit that transitions to communicate directly with the specific terminal without relaying the relay node. System. In the network virtualization system according to claim 1 or 2,
The connection criteria are the amount of traffic relayed through the relay node between a pair of terminals, the localization of the pair of terminals in the topology on the physical network, the protocol type of the traffic, and the load status of the relay node A network virtualization system that is at least one of the above. The network virtualization system according to any one of claims 1 to 3,
Each terminal includes a network interface connected so as to be communicable with the relay node, and a link establishment unit that establishes a peer-to-peer link with a terminal that is a direct communication partner according to an instruction by the first instruction unit. Network virtualization system with The network virtualization system according to any one of claims 1 to 4,
The relay node determines whether or not to disconnect direct communication between a pair of terminals that are in direct communication with each other in consideration of a predetermined disconnection criterion. On the other hand, a network virtualization system comprising second instruction means for instructing to disconnect the direct communication and shift to communication relayed by the relay node. The network virtualization system according to any one of claims 1 to 5,
When each terminal is in direct communication with another specific terminal, it is determined whether or not to disconnect the direct communication in consideration of the disconnection criterion. A network virtualization system comprising second transition means for disconnecting direct communication with the terminal and transitioning to communication relayed by the relay node. In the network virtualization system according to claim 5 or 6,
The disconnection criteria are the amount of traffic due to the direct communication between a pair of terminals, the localization of the pair of terminals in the topology on the physical network, the protocol type of the traffic, and the load status of the relay node A network virtualization system that is at least one of the following. Network virtualization comprising a plurality of terminals connected to any of a plurality of physical networks constructed in advance, and a network virtualization device that constructs a virtual network by a program on the plurality of physical networks The program applied to the system,
A function of relaying communication between the terminals in the constructed virtual network in the network virtualization apparatus;
Whether to communicate directly with each other without relaying in the network virtualization device, regarding any pair of terminals in which the communication is relayed in the network virtualization device in consideration of a predetermined connection standard A program for causing a computer to realize a function of directly communicating with each other without being relayed by the network virtualization apparatus to the pair of terminals when it is determined whether or not to communicate directly .   The connection criteria are the amount of traffic relayed through the network virtualization apparatus between a pair of terminals, the localization of the pair of terminals in the topology on the physical network, the protocol type of the traffic, and the network virtualization The program according to claim 8, wherein the program is at least one of load conditions of the apparatus.   Considering a predetermined disconnection criterion, it is determined whether or not to disconnect direct communication between a pair of terminals that are in direct communication with each other. The program according to claim 8 or 9, further causing the computer to realize a function of disconnecting communication and shifting to communication that relays the network virtualization apparatus.   The disconnection criteria are the amount of traffic due to the direct communication between a pair of terminals, the localization of the pair of terminals in the topology on the physical network, the protocol type of the traffic, and the load status of the network virtualization apparatus The program according to claim 10, which is at least one of the following.

Images