JP2010103633A - Communication control method, communication device, and communication system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve a network having a high value-added function. <P>SOLUTION: In a virtual network which is constructed by software on an underlay network and in which communication is performed via at least one node, a counter table 32 having a counter with a preliminarily determined threshold is provided to each of preliminarily determined ports P for connection to other nodes, of each port. When a preliminarily determined trigger occurs, the counter of a pertinent port is incremented or decremented by a preliminarily determined value. In each node, a preliminarily determined action is executed when the value of the counter reaches the threshold as the results of increment or decrement of the counter. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a communication control method, a communication apparatus, and a communication system applied to a network in which a client communicates via a relay node.

  A virtual Ethernet based on Ethernet (registered trademark) is known as a main overlay network (hereinafter also referred to as “virtual network”) constructed on the underlay network. The virtual Ethernet is formed as follows, for example.

    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” or “relay device”).

    A computer that wishes to participate in virtual Ethernet connects to a virtual hub via a virtual network interface (hereinafter referred to as a “virtual network interface”) constructed by software (hereinafter, the virtual network interface is operated). The computer to be called “client”).

    For example, currently widely used for a connection between a virtual hub and a virtual network interface (ie, a connection between a server and a client) or a connection between a virtual hub and a virtual hub (ie, a connection between a server and a server). An underlay network such as TCP / IP is used.

    Switching processing (transmission port determination processing) at the virtual hub is performed based on the destination MAC address of the received Ethernet frame.

  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 constituting the 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.
JP 2008-042665 A

  The above-described conventional network, for example, virtual Ethernet, is obtained by replacing physical Ethernet infrastructure with software emulation, and has the following advantages in the flexibility of Ethernet segment construction.

-Not subject to physical or geographical restrictions-Easy to construct segments by purpose and organization-Ethernet segments can be constructed and discarded on demand However, virtual Ethernet is an Ethernet emulation and is efficient It does not have higher value-added functions such as CDN (Content Delivery Network), resource search on the network, load distribution, and high reliability through redundant routes.

  The realization of a network having a higher value-added function as described above is desired in addition to the virtual Ethernet.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a communication control method, a communication apparatus, and a communication system that realize a network having a high value-added function.

  A communication control method according to an aspect of the present invention is a communication control method applied to a network that performs communication via at least one node, and is a predetermined port for connecting each node to another node. Each is provided with a counter having a predetermined threshold, and when a predetermined trigger occurs, the counter of the corresponding port is incremented or decremented by a predetermined value. As a result of incrementing or decrementing, a predetermined action is executed when the value of the counter reaches the threshold value.

  According to the present invention, a network having a high value-added function can be realized.

  Embodiments of the present invention will be described below with reference to the drawings. In each embodiment, virtual Ethernet is taken as an example.

<First Embodiment>
First, a first embodiment of the present invention will be described. In the first embodiment, a basic concept common to second to sixth embodiments described later will be described.

  FIG. 1 is a diagram illustrating an example of a basic configuration of a virtual network node according to the first embodiment of the present invention. This FIG. 1 is also applied to second to sixth embodiments described later.

  FIG. 1 assumes a virtual hub in virtual Ethernet, but when the virtual Ethernet switching module in the figure is considered as an upper protocol, for example, an Ethernet layer or a TCP / IP layer, it is assumed that a virtual network interface is assumed. Therefore, the present embodiment can be applied to both a virtual hub and a virtual network interface in virtual Ethernet.

  Hereinafter, when the virtual hub and the virtual network interface are not particularly distinguished, they are simply referred to as a virtual network node.

  The virtual hub H shown in FIG. 1 includes a policy control module 11, a policy table 12, a node control module 13, a MAC address table 20, a virtual Ethernet switching module 21, and a plurality of commands received from an administrator or a signaling protocol 10. It has a virtual port P. Each virtual port includes a virtual port control module 31, a counter table 32, a filter module 33, a filter table 34, and an underlay network 35.

  Hereinafter, these various module groups and table groups will be described.

Policy control module 11
The policy control module 11 performs maintenance of the policy table 12 and execution control of actions to be taken by the node, and specifically performs the following processing.

-Setting of the policy table 12 according to an instruction from the administrator or the signaling protocol 10 and transmission of a counter table entry setting instruction to the virtual port control module 31
-Action execution instruction transmission to the virtual port control module 31 or the node control module 13 triggered by the event notification from the virtual port control module 31-Policy table 12
The format of the policy table 12 is shown in FIG. As shown in FIG. 2, the policy table 12 has the following items.

(1) Virtual port number A virtual port P is bound to an entry in this table.

(2) Event number This is for binding the entry of the counter table 32 held by the virtual port control module 31 of the virtual port P designated by the above number and the entry of this table.

(3) Action The contents are instructed to the virtual port control module 31. For example, opening / closing of ports, discarding of transmission / reception data, and the like.

Node control module 13
The node control module 13 controls the movement, division, integration, etc. of its own node according to instructions from the policy control module 11. A detailed description is omitted here. Details are described in detail as a virtual hub control module in, for example, Japanese Patent Application Laid-Open No. 2008-042665.

・ Virtual Ethernet switching module 21
The virtual Ethernet switching module 21 is a part that emulates an Ethernet switching process, and performs switching of data (Ethernet frame) between virtual ports based on the MAC address table 20. In addition, since the process here is a general thing, the detailed description is omitted here.

・ Virtual port P
The virtual port P is an end point of a virtual communication link with another node. The virtual port P has an interface with the virtual Ethernet switching module 21 in the case of a virtual hub and an upper protocol in the case of a virtual network interface. Data is sent to other nodes using the underlay network 35, and data received from the other nodes using the underlay network 35 is passed to them.

    The virtual port P has the following submodules and tables.

-Virtual port control module 31
The virtual port control module 31 performs the following processing.

* Setting of the counter table 32 according to an instruction from the policy control module 11 and transmission of a filter table entry setting instruction to the filter module 33
* Increase / decrease of counter triggered by notification of filter hit from filter module 33, and notification of event occurrence to policy control module 11 when counter value reaches threshold
* Execution of action to virtual port P by instruction from policy control module 11
-Counter table 32
The format of the counter table 32 is shown in FIG. As shown in FIG. 3, the counter table 32 has the following items.

(1) Event number This is for binding the entry of the policy table 12 held by the policy control module 11 and the entry of this table.

        The policy control module 11 is assigned and notified to the virtual port control module 31 when the counter table is set.

(2) Filter number This is for binding an entry in the filter table 34 held by the filter module 33 and an entry in this table.

        The virtual port control module 31 assigns and notifies the filter module 33 when a filter entry is created.

(3) Counter Increase / decrease when the transmitted / received data hits the filter entry held by the filter module 33.

        An initial value is set by an instruction from the policy control module 11.

(4) Counter threshold value An upper threshold value or a lower threshold value of the counter, and an event occurs when the counter value reaches these values.

        An initial value is set by an instruction from the policy control module 11.

(5) Counter increment / decrement value This is the increment / decrement value when the counter is incremented / decremented.

        An initial value is set by an instruction from the policy control module 11.

-Filter module 33
The filter module 33 creates and holds a filter entry in the filter table 34 according to an instruction from the virtual port control module 31. When the transmission / reception data hits the filter entry, the virtual port control module 31 is notified of this.

-Filter table 34
The format of the filter table 34 is shown in FIG. As shown in FIG. 4, the filter table 34 has the following items.

(1) Filter number This is for binding the entry of the counter table 32 held by the virtual port control module 31 and the entry of this table.

        The virtual port control module 31 is assigned and notified when a filter entry is created.

(2) Filter pattern The filter pattern of the filter entry is assumed to be the same as the general one used in network processing, so the details are omitted here. This will be described in detail in the form.

(3) Direction Indicates transmission or reception.

(4) Action The content is instructed from the policy control module 11 via the virtual port control module 31. For example, opening / closing of ports, discarding of transmission / reception data, and the like.

-Underlay Network 35
This is a protocol for actually communicating with other nodes.

  FIG. 5 shows a general processing flow by the various module groups shown in FIG.

  Here, as an example, after receiving an Ethernet frame in which a source MAC address is MAC-A and a destination MAC address is a broadcast address at a virtual port 2 of a plurality of virtual ports P, The processing details of the module group (between) will be described below, with the processing that the Ethernet frame with the destination MAC address MAC-A is transmitted only once.

  The initial setting of the policy table 12, and the counter table 32 and the filter table 34 associated therewith is triggered by an instruction from the administrator or the signaling protocol 10. These processes are the main points of this embodiment. Because it is not, details are omitted.

  1. When the policy control module 11 of the virtual network node receives the above instruction from the administrator or the signaling protocol 10 (step S1), it performs the following processing.

    (a) The policy table 12 is set as shown in FIG. 6 (step S2).

    (b) A counter table setting request is transmitted to the virtual port control module 31 of the virtual port 2 (step S3). The specific contents are as follows.

・ Counter type: Increase / decrease stop type (If the upper threshold value is reached, the subsequent increase is stopped until it falls below the threshold value again, or if the lower threshold value is reached, the subsequent decrease is made until it exceeds the threshold value again. Stop type counter)
・ Counter initial value: 1
Event number: 100, upper threshold: 1, increment value: 1, filter pattern: source MAC address MAC-A, destination MAC address broadcast, direction: reception event number: 101, lower threshold: 0, decrease value: 1 , Filter pattern: destination MAC address MAC-A, direction: transmission When receiving the counter table setting request from the policy control module 11, the virtual port control module 31 of the virtual port 2 performs the following processing.

    (a) The counter table 32 is set as shown in FIG. 7 (step S4).

    (b) A filter setting request is transmitted to the filter module 33 (step S5). The specific contents are as follows.

Filter number: 10, filter pattern: transmission source MAC address MAC-A, destination MAC address broadcast, direction: reception, action: NULL
Filter number: 11, filter pattern: destination MAC address MAC-A, direction: transmission, action: NULL
The action “NULL” represents the default operation of the virtual port. Usually, it means “PASS” for transmission / reception data, that is, transmission / reception (hereinafter the same).

  3. When the filter module 33 of the virtual port 2 receives the filter setting request from the virtual port control module 31, the filter module 33 performs the following processing.

    (a) The filter table 34 is set as shown in FIG. 8 (step S6).

  4). In the above state, it is assumed that the corresponding virtual network node receives an Ethernet frame from the virtual port 2 whose source MAC address is MAC-A and whose destination MAC address is a broadcast address (step S7).

  5. When receiving the Ethernet frame, the filter module 33 of the virtual port 2 performs the following processing.

    (a) Since the filter entry with the filter number 10 is hit, a filter hit notification is transmitted to the virtual port control module 31 (step S8). The specific contents are as follows.

-Filter number: 10
6). When receiving the filter hit notification from the filter module 33, the virtual port control module 31 of the virtual port 2 performs the following processing.

    (a) The counter table 32 is searched and there is an entry with the filter number 10, but since it is an increase / decrease stop type counter, the counter is not increased (step S9).

    (b) As a result, there is no change in the counter value, so no event occurs.

  7). Next, it is assumed that the corresponding virtual network node transmits an Ethernet frame whose destination MAC address is MAC-A to the virtual port 2.

  8). When the filter module 33 of the virtual port 2 transmits the Ethernet frame, the filter module 33 performs the following processing.

    (a) Since the filter entry with the filter number 11 is hit, a filter hit notification is transmitted to the virtual port control module 31 (step S8). The specific contents are as follows.

-Filter number: 11
9. When receiving the filter hit notification from the filter module 33, the virtual port control module 31 of the virtual port 2 performs the following processing.

    (a) The counter table 32 is searched, and an entry with the filter number 11 exists, so the counter is decreased by 1 (step S9).

    (b) As a result, since the counter value has reached the lower threshold value, an event notification is transmitted to the policy control module 11 (step S10). The specific contents are as follows.

-Virtual port number: 2
・ Event number: 101
10. When the policy control module 11 receives the event notification from the virtual port control module 31 of the virtual port 2, the policy control module 11 performs the following processing.

    (a) The policy table 12 is searched (step S11), and since there are entries of the virtual port number 2 and the event number 101, an action request is transmitted to the virtual port control module 31 of the virtual port 2 (step S12). The specific contents are as follows.

-Virtual port number: 2
・ Event number: 101
-Filter number: 11
Action: DROP (Ethernet frame discard)
11. When receiving the action request from the policy control module 11, the virtual port control module 31 of the virtual port 2 performs the following processing.

    (a) The counter table 32 is searched, and entries for event number 101 and filter number 11 exist, so a filter setting request is transmitted to the filter module 33. The specific contents are as follows.

Filter number: 11, filter pattern: no change, direction: no change, action: DROP (Ethernet frame discard)
12 Thereafter, the virtual port 2 does not transmit an Ethernet frame with a destination MAC address of MAC-A unless it again receives an Ethernet frame with a source MAC address of MAC-A and a destination MAC address of broadcast.

  13. After that, when an Ethernet frame whose source MAC address is MAC-A and destination MAC address is broadcast is received again at virtual port 2, the corresponding counter in the counter table 32 is incremented by 1, and as a result, an event with event number 100 is generated. In addition, as a result, a corresponding action (filter 11: PASS (pass through Ethernet frame)) is executed for the virtual port 2.

  14 As a result, the virtual port 2 is in the same state as when the processing 6 is completed, that is, a state where the Ethernet frame whose destination MAC address is MAC-A can be transmitted only once.

  According to the first embodiment, a high-value-added network can be realized with a simple mechanism by using a counter that performs not only an increase in value but also a decrease or a stop to increase or decrease by a threshold. It becomes possible.

  In the following embodiments, more specific usage scenes using a plurality of nodes will be described.

<Second Embodiment>
Next, a second embodiment of the present invention will be described. In the second embodiment, an example in which the basic concept shown in the first embodiment is applied to a scene for performing anycast emulation will be described.

  Anycast is a system in which multiple servers that provide the same service have a single address, and a client can enjoy a desired service simply by accessing that address, while accessing which server is actually accessed. You don't need to know if you're doing it.

  In the IP network world, anycast is used for a plurality of DNS server arrangements. In this case, an IP routing frame is used so that a client belonging to a specific routing domain accesses a specific DNS server. The work controls the mapping between the anycast address and the server that is actually accessed by the client.

  On the other hand, in the present embodiment, the following framework is provided in order to realize anycast.

    A service request (packet addressed to an anycast address) from a client is transmitted to all servers.

    -All servers that have received the service request send a service response to the corresponding client.

    Each virtual hub configuring the network operates a counter of the virtual port that transmits and receives the request and response, thereby suppressing transmission of a response to the same request for the virtual port for which the response has been transmitted once.

  By using the above framework, avoiding that all responses from the server group are received by the client and forwarded by the node group leading to it without using a complicated mechanism such as IP routing. Thus, anycast can be realized without wasting network resources.

  FIG. 9 shows an example of the configuration of the communication system according to the present embodiment.

  It is assumed that the communication system according to the present embodiment constitutes a network in which one client and a plurality of servers 0 to 5 are connected via virtual hubs 0 to 7. Each of the virtual hubs 0 to 7 is a node having the same function as the virtual hub H described in the first embodiment. Since the details of the processing of each node are as described in the first embodiment, detailed description thereof is omitted. Here, the explanation will focus on the behavior of the entire system.

  Furthermore, in this embodiment, the IP anycast address is stored in an Ethernet frame having a broadcast address or a multicast address as a destination MAC address, and is transmitted from the client. That is, static binding is used for address resolution from an IP address to a MAC address (the case where ARP address resolution is used will be described later).

  Initially, with reference to FIG. 10, operation | movement in case the structure of this embodiment is not introduced is demonstrated.

  1. Assume that the client has an IP address IP-C, and the servers 0 to 5 have an IP anycast address IP-AC-A.

  2. The IP client transmits an IP packet addressed to IP-AC-A.

  3. Since the destination MAC address of the Ethernet frame storing the IP packet is a broadcast address or a multicast address, it is transmitted to all nodes on the network.

  4). Since the destination of the IP packet stored in the received Ethernet frame is its own address IP-AC-A, the servers 0 to 5 perform appropriate processing on the destination and return a response to the client (destination) IP address IP-C).

  5. The response transmitted by the server 0 is received by the client via the virtual hubs 2, 1, 0.

  6). The response transmitted by the server 1 is received by the client via the virtual hubs 3, 1, 0.

  7). The responses transmitted by the servers 2 and 3 are received by the client via the virtual hubs 5, 4, 1, and 0.

  8). The response transmitted by the server 4 is received by the client via the virtual hubs 6, 4, 1, 0.

  9. The response transmitted by the server 5 is received by the client via the virtual hubs 7, 4, 1, 0.

  FIG. 10 shows the transfer state of the response packet between the nodes in this case. As shown in FIG. 10, the same response packet passes through a plurality of virtual hubs a plurality of times, which is very inefficient. Furthermore, the client receives responses for the number of servers, which is also inefficient.

  Next, with reference to FIGS. 11 to 20, an operation when the mechanism of the present embodiment is introduced will be described.

  1. Assume that the client has an IP address IP-C and a MAC address MAC-C, and the servers 0 to 5 have an IP anycast address IP-AC-A. In addition, the client stores IP-AC-A in an Ethernet frame having a destination MAC address MAC-MC-A and transmits the IP-AC-A.

  2. It is assumed that all the virtual ports 0 to 7 have a policy table, a counter table, and a filter table having the contents shown in FIGS. 12, 13, and 14 (this setting can be made by the administrator or the administrator). It is assumed that the signaling protocol 10 is used, but details are omitted). The wild card in the figure indicates an arbitrary MAC address. In the figure, only virtual port 0 is shown, but the other virtual ports have the same contents.

  3. The IP client transmits an IP packet addressed to IP-AC-A.

  4). Since the destination MAC address of the Ethernet frame storing the IP packet is a multicast address, it is transmitted to all nodes on the network.

  5. When the virtual port on the client side of each virtual hub is the upstream port, and the virtual port on the server side is the downstream port, the policy table, counter table, and filter table for the upstream port of each virtual hub are sent and received by the corresponding Ethernet frame. These are shown in FIGS. 15, 16, and 17, respectively.

  At this time, a new filter entry in which the wild card portion in the filter pattern is embodied is created. Thereafter, in the filter matching process, these embodied filter entries have priority over the wild card entry.

  6). Since the destination of the IP packet stored in the received Ethernet frame is its own address IP-AC-A, the servers 0 to 5 perform appropriate processing on the destination and return a response to the client (destination) IP address IP-C, destination MAC address MAC-C), where server 2 is the fastest and performs response processing, and the response from server 2 is received by the client before the other server sends the response. To do. At this time, the policy table, counter table, and filter table of the upstream ports of the virtual hubs 0, 1, 4, and 5 are as shown in FIGS. 18, 19, and 20, respectively.

  On the other hand, there is no change in the same table of the virtual hubs 2, 3, 6, and 7. However, similarly to the above, when these virtual hubs subsequently receive a response from a certain server, the table becomes the same as above. .

  In such a state, a response transmitted by another server is as follows.

  7). The response transmitted by the server 0 is discarded by the virtual hub 1 via the virtual hub 2.

  8). The response transmitted by the server 1 is discarded by the virtual hub 1 via the virtual hub 3.

  9. The response transmitted by the server 3 is discarded at the virtual hub 5.

  10. The response transmitted by the server 4 is discarded by the virtual hub 4 via the virtual hub 6.

  11. The response transmitted by the server 5 is discarded by the virtual hub 4 via the virtual hub 7.

  FIG. 11 shows the transfer state of the response packet between the nodes in this case. As shown in FIG. 11, as a result, duplicate responses addressed to clients from each server are discarded at a virtual hub closer to the server, so that waste of network resources can be suppressed. Incidentally, it can be seen that in this configuration, at least inflow of a response to the virtual hub 0 is suppressed even in the worst case.

  Next, a case where a normal ARP (Address Resolution Protocol) is used for address resolution of the IP anycast address and the destination MAC address of the Ethernet frame storing the IP anycast address will be described.

  In this case, the address resolution to the MAC address is performed before the client transmits the IP packet of the destination IP address IP-AC-A. For the ARP request and ARP response for that purpose, the same processing as described above is performed. Will be performed. In other words, the ARP request is broadcast and received by all the servers, and only the ARP response from the server that performed the response at the fastest is received by the client. Discarded at the upstream port.

  As a result, the IP packet addressed to IP-AC-A after address resolution is transmitted to only one server, and the response to the IP packet is also transmitted from the server to the client (bidirectional at the IP level). Both are unicast communications).

  According to the second embodiment, anycast can be efficiently emulated with a simple mechanism, and system performance can be improved and resources can be optimized.

<Third Embodiment>
Next, a third embodiment of the present invention will be described. In the third embodiment, an example will be described in which the basic concept shown in the first embodiment is applied to a scene that limits the number of clients that can simultaneously access a server.

  FIG. 21 shows an example of the configuration of the communication system according to the present embodiment.

  The communication system according to the present embodiment includes a network in which a plurality of clients 0 to 8 and one server are connected via virtual hubs 0 to 5. Each of the virtual hubs 0 to 5 is a node having the same function as the virtual hub H described in the first embodiment. Here, a system is assumed in which a plurality of clients 0 to 8 issue a request in the server, the server performs an appropriate process for the request, and then returns a response to the client that sent the request.

  Hereinafter, the operation in the present embodiment will be described with reference to FIGS.

  1. Assume that the server has an IP address IP-S.

  2. It is assumed that the virtual port 0 of the virtual hub 3 has a policy table, a counter table, and a filter table having the contents shown in FIGS. 22, 23, and 24 respectively. 10 is assumed, but details are omitted).

  3. When a client sends a request to the server, the virtual hub 3 sends the request to the virtual hub 4 from the virtual port 0. At that time, the corresponding counter is incremented by one. Thereafter, when the virtual hub 3 receives a response to the request from the virtual port 0, the virtual hub 3 decrements the counter by one.

  4). Here, when the access from the client to the server is not parallel, the corresponding counter of the virtual hub 3 only transitions between 0 and 1, and no action regarding the virtual port 0 of the virtual hub 3 is taken. .

  5. On the other hand, when the access from the client to the server is parallel, that is, when another client sends a request before the response to the request from one client is sent from the server, If the frequency of request transmission from the client is higher than the response transmission, the counter will eventually reach the upper threshold value of 5. At this time, an event 300 occurs in the virtual hub 3, and as a result, a DROP (Ethernet frame discard) action for the filter 30 of the virtual port 0 is executed.

  6). Thereafter, while the counter is in the state of 5, all packets whose destination is IP-S are discarded at the virtual port 0 of the virtual hub 3.

  7). Thereafter, the server sequentially transmits responses, and the counter value is decremented. As a result, when the lower threshold value 3 is reached, an event 301 occurs in the virtual hub 3, and as a result, the filter 30 of the virtual port 0 A PASS (Ethernet frame passing) action is executed.

  8). Thereafter, while the counter does not reach 5 again, all packets whose destination is IP-S are transmitted on the virtual port 0 of the virtual hub 3.

  In the above example, the lower threshold is set to 3, but this can of course be set to a value other than 3 depending on the system requirements, the operation policy, and the like.

  According to the third embodiment, it is possible to limit the number of clients that can efficiently and simultaneously access the server with a simple mechanism, and to improve the availability.

<Fourth Embodiment>
Next, a fourth embodiment of the present invention will be described. In the fourth embodiment, an example in which the basic concept shown in the first embodiment is applied to a scene where abnormal traffic (such as DoS) from a certain direction is blocked or the flow rate is limited. explain.

  FIG. 25 shows an example of the configuration of the communication system according to the present embodiment.

  For example, a configuration is assumed in which the virtual port 0 side of the virtual hub H is the own organization and the virtual port 1 side is the other organization. That is, abnormal traffic from other organizations is targeted for action.

  Hereinafter, the operation in the present embodiment will be described with reference to FIGS.

  1. It is assumed that the virtual port 1 of the virtual hub H has a policy table, a counter table, and a filter table having the contents shown in FIGS. 26, 27, and 28. 10 is assumed, but details are omitted).

  2. If there is no significant difference between the traffic from the virtual port 0 to the virtual port 1 and the traffic in the opposite direction from the viewpoint of the virtual hub H, the corresponding counter has a value sufficiently smaller than the upper threshold 1000 (or the lower threshold 200). It will only increase or decrease.

  3. Here, for example, when the traffic from the virtual port 1 to the virtual port 0 rapidly increases, the counter eventually reaches the upper threshold 1000. At this time, the event 700 occurs in the virtual hub H, and as a result, the reception stop action for the filter 70 of the virtual port 1 is executed.

  4). Thereafter, no reception processing is performed at the virtual port 1 of the virtual hub H.

  5. Thereafter, each time traffic from the virtual port 0 to the virtual port 1 is generated, the counter is decremented, and eventually the counter reaches the lower threshold value 200. As a result, the event 701 occurs in the virtual hub H, and as a result, the reception start action for the filter 70 of the virtual port 1 is executed.

  6). Thereafter, while the counter does not reach 1000 again, reception processing is performed in the virtual port 1 of the virtual hub H.

  It is possible to introduce the concept of a timer in the counter process. In this case, for example, it is as follows.

  Set 60 seconds as a timer for reaching the upper threshold value 1000 for the corresponding counter.

  If the counter reaches the upper threshold within 60 seconds, the same action as in the above example is executed.

  If the timer period expires before the counter reaches the upper threshold 1000, a reset action (reset to the initial value) is executed on the counter and the timer is started again.

  That is, it can be said that the concept of “per unit time” is introduced as a criterion for judging that the traffic is abnormal.

  In the above example, when the corresponding counter reaches the upper threshold value 1000, reception at the corresponding virtual port is stopped, that is, the virtual port is blocked. On the other hand, if the virtual port has functions such as priority queue and CoS processing, the action of changing these parameters is executed to completely stop the traffic in the direction of the problem. Instead, the flow rate can be limited.

  According to the fourth embodiment, it is possible to efficiently block abnormal traffic from a certain direction or restrict the flow rate by a simple mechanism, and to improve availability.

<Fifth Embodiment>
Next, a fifth embodiment of the present invention will be described. In the fifth embodiment, an example in which the basic concept shown in the first embodiment is applied to a scene in which movement (optimum placement) of a virtual hub is performed according to the traffic volume of a certain region. Will be described.

  FIG. 29 shows an example of the configuration of a communication system according to this embodiment.

  For example, the virtual hub H has virtual ports 0 to 3 and performs traffic switching between the respective ports. Here, it is known that the other nodes connected to the virtual ports 0 and 1 and the other nodes connected to the virtual ports 2 and 3 are physically close to each other. To do.

  In this embodiment, the concept of grouping virtual ports is introduced, and a counter is provided for each group (in the above embodiments, all counters belong to virtual ports). Further, the concept of “inter-virtual port traffic” is also introduced as a filter pattern (in the previous embodiments, the filter pattern was based on the attribute of the data itself).

  Hereinafter, the operation in the present embodiment will be described with reference to FIGS.

  1. Virtual ports 0 to 3 of the virtual hub H are defined as a virtual port group.

  2. It is assumed that the virtual port group of the virtual hub H has a policy table, a counter table, and a filter table having the contents shown in FIGS. 30, 31, and 32. It is assumed that protocol 10 is used, but details are omitted.)

  3. From the viewpoint of the virtual hub H, there is no significant difference in traffic between the virtual port 0 and the virtual port 1 and traffic between the virtual port 2 and the virtual port 3, and the corresponding counter will only increase or decrease near the initial value 100. .

  4). Here, for example, when the traffic between the virtual port 0 and the virtual port 1 increases rapidly and clearly becomes larger than the traffic between the virtual port 2 and the virtual port 3, the counter will eventually reach the upper threshold value 200. At this time, an event 600 occurs in the virtual hub H, and as a result, an action (movement to the execution environment on the right side in the figure and reset of the counter to the initial value) is executed for the own node. Note that information indicating the execution environment on the right side in the drawing is stored in advance in a predetermined recording area, and movement to the execution environment is possible by referring to the information.

  5. Thereafter, the virtual hub H operates in a position physically close to other nodes with a larger traffic volume, and becomes more efficient from the viewpoint of consumption of network resources.

  6). Thereafter, for example, when the traffic between the virtual port 2 and the virtual port 3 increases, the counter reaches the lower threshold value 0, for example, contrary to the above. At this time, an event 601 occurs in the virtual hub H, and as a result, an action for the own node (movement to the execution environment on the left side in the figure and resetting the counter to the initial value) is executed. Note that information indicating the execution environment on the left side in the figure is stored in advance in a predetermined recording area, and movement to the execution environment is possible by referring to the information.

  In the above example, the right movement and the left movement are defined as the action. However, it is also possible to define the movement to the center (original place) according to the operating environment and to move the movement in two stages.

  As in the fourth embodiment, the concept of a timer can be introduced into the counter process.

  Furthermore, in the above example, the action is the movement of the own node, but the action of dividing the own node is also assumed. In this case, for example, the processing 4 part in the above example is replaced as follows.

-A new virtual hub H 'is created in the execution environment on the right side of the figure.-The virtual hub H has transferred the other nodes accommodated by the virtual port 0 and virtual port 1 to the newly created virtual hub H'. Change the connection.

  The virtual hub H generates a connection (virtual port) with the newly created virtual hub H ′.

  In the above example, the action is to divide the own node, but conversely, an action to be integrated with other nodes is also assumed. In this case, for example, the divided virtual hub H ′ operating on the right side performs the following processing.

  -If traffic between virtual ports other than the virtual ports connected to the virtual hub H decreases, the policy table, counter table, and filter table settings (details are omitted because they are not significantly different from those already described), An action (integration with another virtual hub) for the own node is executed. Information indicating the integration destination virtual hub is stored in a predetermined recording area in advance, and integration into the execution environment can be performed by referring to the information.

  Change the connection of the other node accommodated in the virtual port other than the virtual port connected to the virtual hub H to the virtual hub H.

  -Disconnect the virtual port connected to the virtual hub H and terminate the local node.

  According to the fifth embodiment, it is possible to move, divide, and integrate (optimum placement) a virtual hub according to the traffic volume of a certain region with a simple mechanism, and improve system performance and resources. Can be optimized.

<Sixth Embodiment>
Next, a sixth embodiment of the present invention will be described. In the sixth embodiment, an example will be described in which the basic concept shown in the first embodiment is applied to a scene where a redundant route is used to construct a highly reliable data distribution network. To do.

  An example in which the influence of the use of a wild card in the filter pattern remains only in the filter table 34 will also be described. Incidentally, in the second embodiment, the use example of the wild card in the filter pattern is also shown. However, as described above, the influence extends to the policy table 12 and the counter table 32 as well.

  FIG. 33 shows an example of the configuration of the communication system according to the present embodiment.

  It is assumed that the communication system according to the present embodiment constitutes a network in which one client and one server are connected via virtual hubs 0 to 6. Each of the virtual hubs 0 to 6 is a node having the same function as the virtual hub H described in the first embodiment.

  Further, the client transmits data to the server. For example, since the route to the virtual hub 4 has a problem in communication quality in the WAN, there are two routes (virtual hub 0-virtual hub 1, virtual hub 2). -Assume a system in which the same data is transmitted to the virtual hub 3), and what arrives first in the virtual hub 4 is transmitted ahead, and what arrives later is discarded.

  Hereinafter, the operation in the present embodiment will be described with reference to FIGS.

  1. Assume that the server has an IP address IP-S.

  2. It is assumed that the virtual port 0 of the virtual hub 4 has a policy table, a counter table, and a filter table having the contents shown in FIGS. 34, 35, and 36. 10 is assumed, but details are omitted).

  3. When the client sends data to the server, the virtual hub 4 receives the data from the virtual port 1 or the virtual port 2, but receives the data from either one and sends it to the virtual hub 5 from the virtual port 0 Suppose that At this time, the filter table 34 of the virtual port 0 of the virtual hub 4 is as shown in FIG. 37 (the policy table 12 and the counter table 32 remain as they are).

  4). Here, it is assumed that another data (same data as described above) is received from either the virtual port 1 or the virtual port 2. In this case, the data is discarded according to the contents of the filter 50. Therefore, duplicate data does not flow beyond the virtual hub 5.

  In the above example, the deletion process of the filter entry generated from the wild card is not mentioned, but from the viewpoint of the virtual hub resource, these filter entries should be released in some way. . The following is assumed as an embodiment for that purpose.

-One-shot filter A filter entry has a one-shot attribute. That is, once used, the filter entry is deleted at that time.

-Garbage collection "Recover" filter entries that have not been used for a certain period of time. Note that after the corresponding filter entry is deleted, the entry may be needed again, but as long as the corresponding wildcard entry exists, a similar filter entry should be generated again. There is no particular problem.

  Furthermore, in the above example, it is assumed that the sequence number of TCP or SCTP is used as the wild card part of the filter pattern. On the other hand, if you want to use a higher-level protocol that does not have the concept of sequence number, such as IP or UDP, or that has a sequence number, or if you want to introduce a concept that does not exist in the higher-level protocol in the first place as a filter pattern, It is effective to use the control information.

  In general, in the implementation of a virtual network including virtual Ethernet, a control header is often added before user data (for Ethernet, an Ethernet frame).

  Therefore, a virtual network node such as an application, a virtual network interface, or a virtual hub encodes information having a special meaning in the header part (virtual Ethernet control information storage part) and uses it as a filter pattern. Become.

  According to the sixth embodiment, a redundant path can be efficiently constructed with a simple mechanism, and the reliability of the system can be improved.

<Application>
The present invention can be widely applied to a system that provides a high value-added network using a virtual network. For example, the present invention can be applied to the following.

・ CDN (Content Delivery Network)
-Efficient resource search on the network (anycast function)
・ Highly reliable and highly efficient network (provides functions such as redundant routes and load distribution)
<Summary>
The concept of various functions and information described in the above embodiments is summarized as follows.

[1] (Removed the restriction of “virtual”)
Control of data transmission / reception by nodes (relay nodes, clients), port control, and node execution environment, execution status, execution method, and other processes related to execution in a network that communicates between at least one relay node This is a control method and has the following characteristics.

  Each node has a “counter” having the following characteristics in each “predetermined port” connected to another node.

    -Has a predetermined threshold (integer).

    -When a "predetermined trigger" occurs, the counter of the corresponding port is increased or decreased by a predetermined value (a finite integer of 1 or more).

  When each node reaches the predetermined threshold value within a predetermined time (including infinity) as a result of incrementing or decrementing the counter, the “predetermined action” ”Is executed.

[2] (Basic part)
Control of data (Ethernet frame) transmission / reception by nodes (relay nodes, clients) and ports (virtual ports) in a virtual network (virtual Ethernet) in which clients communicate via at least one relay node (virtual hub) This is a method for controlling the control, the execution environment of the node, the execution state, the execution method, and other processes related to execution, and has the following features.

  Each node has a “counter” having the following characteristics for each “predetermined port” (virtual port) connected to another node.

    -Has a predetermined threshold (integer).

    -When a "predetermined trigger" occurs, the counter of the corresponding port (virtual port) is increased or decreased by a predetermined value (a finite integer of 1 or more).

  When each node reaches the predetermined threshold value within a predetermined time (including infinity) as a result of incrementing or decrementing the counter, the “predetermined action” ”Is executed.

[3] (One of “predetermined triggers”, data transmission / reception)
The “predetermined trigger” in claim 1 has any of the following characteristics.

-When "predetermined data" is transmitted-When predetermined data is received [4] (Specific example of "predetermined data" transmitted and received)
The “predetermined data” in claim 2 has any of the following characteristics.

Data having specific virtual network control information Information for virtual network control added by a client application, a virtual network interface, a virtual hub, or the like.

    For example, in the case of virtual Ethernet, information in the control header added before the Ethernet header.

Data having specific Ethernet frame header information For example, an Ethernet frame exchanged by an end user having a specific source MAC address, destination MAC address, protocol type, etc.

Data having specific Ethernet frame payload information For example, an Ethernet frame exchanged by an end user having data of a specific upper protocol. For example, a source IP address, a destination IP address, a source port number, a destination port number, a sequence number, a stream number, a PPID, a specific content of user data, and the like.

Data [5] transmitted / received between specific ports (“predetermined ports” subject to counter increase / decrease)
The designation method of “predetermined port” in claims 0 and 1 has at least one of the following characteristics.

・ All port settings Assumes applications such as default all ports.

    Of course, all ports may be subject to the following static and dynamic settings.

・ Static setting It is set at initialization by configuration.

-Dynamic setting It is dynamically set by the administrator or the signaling protocol 10 during operation.

[6] ("predetermined action")
The “predetermined action” in claims 0 and 1 has at least one of the following characteristics.

・ Counter control
-Stop / start counter increment / decrease
-Counter value reset ・ Transmission / reception control (transmission / reception control of specific data)
Stop (start) transmission of the corresponding data (data that triggered the counter increase / decrease) at the corresponding port, or stop (start) reception.

・ Port control (transmission / reception control of all data)
-Open / close Stops (starts) transmission at the port or stops (starts) reception.

-Change of parameters related to QoS, CoS, etc.
-Change, add, or delete the connection destination Change, add, or delete the connection destination of the corresponding port.

・ Node control
-Node division, integration, and movement For example, determine the positional relationship between you and the server or client from the port mark status, and migrate to an appropriate location.

[7] (Counter is shared by port group)
The counter according to claims 0 and 1 is shared by the grouped multiple port groups.

[8] (Counter type)
The counter according to claims 0 and 1 has any of the following characteristics.

  As a result of increasing / decreasing the counter by a predetermined value, when the predetermined threshold is reached, the increase / decrease is continued when a further event occurs.

  As a result of increasing the counter by a predetermined value, after reaching a predetermined threshold, the subsequent increase is stopped. After that, when the counter is decreased by a predetermined value and as a result again falls below a predetermined threshold, the subsequent increase is resumed.

  As a result of decreasing the counter by a predetermined value, after reaching a predetermined threshold, the subsequent decrease is stopped. Thereafter, when the counter is incremented by a predetermined value and as a result again exceeds a predetermined threshold, the subsequent decrease is resumed.

[9] (Wild card for “predetermined data” to be transmitted and received)
The “predetermined data” in claim 3 has the following characteristics.

  ・ Specify some wildcards.

  ・ When the transmitted / received data matches a predetermined part, replace the part specified by the wildcard with the value of the corresponding part of the actually transmitted / received data (learning), and keep it as new “predetermined data” Used for subsequent processing.

[10] (corresponding to the anycast emulation of the second embodiment)
A network system comprising at least one relay node having a counter having the characteristics of claims 0 to 5, 7, and 8, a client connected by the relay nodes, and at least one server group, and the following characteristics: have.

・ At least one server group
-Identified by a single identifier.

    -Provide the same service to clients.

    -When a request for a service provided by itself is received from a client specified by the identifier, processing necessary for the service is performed and a response is returned to the client.

·client
-When enjoying the service, send a request addressed to the identifier.

    -Receive a response to the request from any server.

・ Relay node
-All virtual ports have the following counters:

      * Has 1 as the initial value.

      * Has 1 as the upper threshold and 0 as the lower threshold.

      * When a request having the identifier as a destination is received from a client, the counter is incremented by one.

      * Sending a response from the server to the client to the request decrements the counter by one.

      * If the counter reaches the upper threshold, the counter is not incremented thereafter.

      * If the counter has reached the lower threshold, the counter will not be decreased thereafter.

    -When the counter reaches the lower threshold in the counter, the same response is not transmitted (discarded) from the virtual port as long as the counter is at the lower threshold.

  As described above in detail, according to the embodiment of the present invention, virtual Ethernet, more specifically, a virtual port that is an end point of a virtual communication link that connects nodes constituting the virtual Ethernet is software. Focusing on a certain point, high-value-added functions can be realized by introducing and implementing the concept of “marking to a port (virtual port) triggered by data transmission / reception”, which could not be realized with a physical port. It is possible to realize a network having

  In each of the above-described embodiments, the virtual Ethernet has been described as an example. However, the present invention is not limited to this and can be applied to a non-virtual network.

  In addition, the method described in the above embodiment includes a magnetic disk (floppy (registered trademark) disk, hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), a magneto-optical disk (MO) as programs that can be executed by a computer. ), And can be distributed in a storage medium such as a semiconductor memory.

  In addition, as long as the storage medium can store a program and can be read by a computer, the storage format may be any form.

  In addition, an OS (operating system) running on a computer based on an instruction of a program installed in the computer from a storage medium, MW (middleware) such as database management software, network software, and the like realize the above-described embodiment. A part of each process may be executed.

  Furthermore, the storage medium in the present invention is not limited to a medium independent of a computer, but also includes a storage medium in which a program transmitted via a LAN or the Internet is downloaded and stored or temporarily stored.

  Further, the number of storage media is not limited to one, and the case where the processing in the above embodiment is executed from a plurality of media is also included in the storage media of the present invention, and the media configuration may be any configuration.

  The computer according to the present invention executes each process in the above-described embodiment based on a program stored in a storage medium, and a single device such as a personal computer or a server computer or a plurality of devices are connected to a network. Any configuration of the system or the like may be used.

  In addition, the computer in the present invention is not limited to a personal computer or server computer, but includes a processing unit, a microcomputer, and the like included in an information processing device, and is a generic term for devices and devices that can realize the functions of the present invention by a program. is doing.

  The present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Moreover, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine the component covering different embodiment suitably.

FIG. 3 is a diagram illustrating an example of a basic configuration of a virtual network node according to the first embodiment of the present invention. The figure which shows the format of a policy table. The figure which shows the format of a counter table. The figure which shows the format of a filter table. The figure which shows the general processing flow by various module groups. The figure which shows an example of the policy table in the embodiment. The figure which shows an example of the counter table in the embodiment. The figure which shows an example of the filter table in the embodiment. The figure which shows an example of a structure of the communication system which concerns on the 2nd Embodiment of this invention. The figure for demonstrating operation | movement when the structure of the embodiment is not introduced. The figure for demonstrating operation | movement when the structure of the embodiment is introduced. The figure which shows an example (the 1) of the policy table in the embodiment. The figure which shows an example (the 1) of the counter table in the embodiment. The figure which shows an example (the 1) of the filter table in the embodiment. The figure which shows an example (the 2) of the policy table in the embodiment. The figure which shows an example (the 2) of the counter table in the embodiment. The figure which shows an example (the 2) of the filter table in the embodiment. 6 is a diagram showing an example (part 3) of the policy table in the embodiment. FIG. The figure which shows an example (the 3) of the counter table in the embodiment. The figure which shows an example (the 3) of the filter table in the embodiment. The figure which shows an example of a structure of the communication system which concerns on the 3rd Embodiment of this invention. The figure which shows an example of the policy table in the embodiment. The figure which shows an example of the counter table in the embodiment. The figure which shows an example of the filter table in the embodiment. The figure which shows an example of a structure of the communication system which concerns on the 4th Embodiment of this invention. The figure which shows an example of the policy table in the embodiment. The figure which shows an example of the counter table in the embodiment. The figure which shows an example of the filter table in the embodiment. The figure which shows an example of a structure of the communication system which concerns on the 5th Embodiment of this invention. The figure which shows an example of the policy table in the embodiment. The figure which shows an example of the counter table in the embodiment. The figure which shows an example of the filter table in the embodiment. The figure which shows an example of a structure of the communication system which concerns on the 6th Embodiment of this invention. The figure which shows an example of the policy table in the embodiment. The figure which shows an example of the counter table in the embodiment. The figure which shows an example (the 1) of the filter table in the embodiment. The figure which shows an example (the 2) of the filter table in the embodiment. The figure which shows the general structure of virtual Ethernet.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Administrator or signaling protocol, 11 ... Policy control module, 12 ... Policy table, 13 ... Node control module, 20 ... MAC address table, 21 ... Virtual Ethernet switching module, 31 ... Virtual port control module, 32 ... Counter table, 33 ... Filter module, 34 ... Filter table, 35 ... Underlay network, H ... Virtual hub, P ... Virtual port.

Claims (13)

A communication control method applied to a network that performs communication via at least one node,
A counter having a predetermined threshold is provided in each predetermined port for connecting to another node of each node,
When a predetermined trigger occurs, the counter of the corresponding port is increased or decreased by a predetermined value,
A communication control method, wherein a predetermined action is executed when the value of the counter reaches the threshold value as a result of increasing or decreasing the counter in each node. A communication control method applied to a virtual network configured by software on an underlay network and performing communication via at least one node,
A counter having a predetermined threshold is provided in each predetermined port for connecting to another node of each node,
When a predetermined trigger occurs, the counter of the corresponding port is increased or decreased by a predetermined value,
A communication control method, wherein a predetermined action is executed when the value of the counter reaches the threshold value as a result of increasing or decreasing the counter in each node. In each node, when the counter value reaches the threshold value within a predetermined time as a result of increasing or decreasing the counter, the predetermined action is executed. Communication control method.   The communication control according to claim 1 or 2, wherein the predetermined trigger is generated when predetermined data is transmitted or when predetermined data is received. Method.   The predetermined data includes 1) data having specific virtual network control information for controlling the virtual network, 2) data having specific header information, 3) data having specific payload information, and 4 3. The communication control method according to claim 2, wherein the communication control method corresponds to at least one of data transmitted and received between specific ports.   When specifying the predetermined port, 1) all port settings for specifying all ports by default, 2) static setting for specifying ports at initialization, and 3) during operation by an administrator or a signaling protocol The communication control method according to claim 1, wherein at least one of dynamic setting for designating a port is performed.   The predetermined actions are 1) stop or start of increment or decrease of the counter value, or reset, 2) stop or start of transmission or reception of specific data, or 3) stop or start of transmission or reception at the corresponding port. The communication control according to claim 1, wherein the communication control corresponds to at least one of a parameter relating to start or transmission / reception, a change of a connection destination, and 4) movement, division, and integration of nodes. Method.   The communication control method according to claim 1, wherein the counter is shared by a plurality of grouped port groups.   The counter 1) increases or decreases the counter by a predetermined value, and continues to increase or decrease when a further event occurs after reaching a predetermined threshold. 2) After reaching a predetermined threshold as a result of increasing by a predetermined value, the subsequent increase is stopped, and then the predetermined threshold is again set as a result of decreasing the counter by a predetermined value. If it falls below, the subsequent increase is resumed. 3) After the counter is decreased by a predetermined value, the predetermined threshold is reached and then the subsequent decrease is stopped. 2. The method according to claim 1, wherein when the value is increased by a predetermined value and exceeds a predetermined threshold value again, it corresponds to at least one of those for restarting the subsequent decrease. Or The communication control method according to.   A part of the predetermined data is designated as a wild card, and when the transmitted / received data matches a predetermined part, the designated part of the wild card is replaced with the value of the corresponding part of the actually transmitted / received data. 6. The communication control method according to claim 5, wherein the communication control method is stored as new predetermined data and used for subsequent processing. A communication device as a node for relaying communication in a network,
A counter having a predetermined threshold provided in each predetermined port for connecting to another node;
Means for increasing or decreasing the counter of the corresponding port by a predetermined value when a predetermined trigger occurs;
And a means for executing a predetermined action when the value of the counter reaches the threshold value as a result of increasing or decreasing the counter. A communication device as a node that relays communication by a virtual hub in a virtual network constructed by software on an underlay network,
A counter having a predetermined threshold provided in each predetermined port for connecting to another node;
Means for increasing or decreasing the counter of the corresponding port by a predetermined value when a predetermined trigger occurs;
And a means for executing a predetermined action when the value of the counter reaches the threshold value as a result of increasing or decreasing the counter. A virtual network communication system constructed by software on an underlay network, in which a client communicates with at least one server via at least one relay node,
Each server is identified by a single identifier, provides the same service to the client, and receives a request for the service provided by the client, specified by the identifier, from the client, performs processing necessary for the service. And means for returning a response to the client,
When the client enjoys the service, the client includes a means for transmitting a request addressed to the identifier and receiving a response to the request from any server,
In the relay node, all virtual ports have a counter, and this counter has 1 as an initial value, 1 as an upper threshold, and 0 as a lower threshold. 1) The identifier is received from the client. When a request held at the destination is received, the value of the counter is incremented by 1. 2) When a response from the server to the client is transmitted in response to the request, the value of the counter is decreased by 1. 3) If the value has reached the upper threshold value, the counter value is not increased thereafter. 4) If the counter value has reached the lower threshold value, control is performed so that the counter value is not decreased thereafter. And means for controlling not to transmit the same response from the virtual port as long as the counter reaches the lower threshold when the counter reaches the lower threshold. Communication system to be.

Images