JP2012165172A - Communication system, communication device and supervision control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the conventional problem of a redundant path and an increased delay when live migration is performed between virtual machines disposed in distribution.SOLUTION: A communication system for providing a live migration function between a plurality of virtual machines to a third communication device comprises: a plurality of first communication devices disposed for each of the plurality of virtual machines for IDC; a plurality of second communication devices constituting a carrier network; the third communication device for connecting a subscriber-side terminal; a NW supervision control device for managing the second communication devices; and an IDC supervision control device for managing the virtual machines and the first communication devices. The communication system further comprises a NMS cooperation line for interconnecting the NW supervision control device with the IDC supervision control device. From the IDC supervision control device, the NW supervision control device acquires information related to the modification of a connection destination of a virtual machine through the NMS cooperation line, so as to modify routing tables in the second communication devices.

Description

  The present invention relates to a communication system, a communication apparatus, and a monitoring control apparatus that provide a live migration function to a communication apparatus used in a cloud environment.

  2. Description of the Related Art In recent years, cloud services that place servers, storage, and software on a network and use information and software necessary for subscriber devices on the network have attracted attention. For example, a technique for virtually unifying and providing distributed information (for example, see Patent Document 1) has been considered.

  In general, IDCs (Internet data centers) that provide information in a cloud environment are often installed in the same building or on the same site. However, in order to respond to regional disasters such as earthquakes, storms and floods, and terrorism, it is necessary to disperse and install IDCs at remote locations. In particular, when providing information services between remote locations, it supports live migration technology that moves the connection destination virtual machine (information provision server) to another virtual machine without bringing down the application on the subscriber side that accesses the information. It is demanded. For this reason, for example, a method of establishing a private line between virtual machines between remote locations and connecting them can be considered, but a communication facility for the dedicated line must be provided, such as the equipment fee of the communication device and the line usage fee, etc. High cost is required. Therefore, a method of using a carrier side communication network instead of a dedicated line is conceivable. However, the information providing service in the current cloud environment depends only on the IDC side and the subscriber side, and the IDC side and the subscriber side Control related to IDC connection is not performed on the carrier side connecting the network, various problems in realizing the live migration function, such as increase in line delay, increase in traffic volume, and need to change the route at least twice There is.

  An example of a conventional communication system 900 is shown in FIG. FIG. 17 is a diagram showing an outline of a conventional communication system 900. A terminal TE # 91 on a subscriber side 901 is connected to a network 904 on a carrier side 902 of a carrier company that provides an Internet connection environment via a communication device CE # 94. It is connected and uses a cloud service provided by the IDC side 903. On the other hand, the IDC side 903 on the side of the IDC service company that provides the IDC service installs two IDCs, IDC # 91 and IDC # 92, at remote locations (for example, Tokyo and Yokohama). In IDC # 91, a virtual machine VM # 91 is installed as a server for providing a cloud service, and is connected to the network 904 on the carrier side 902 via the communication device CE # 91. Similarly, a virtual machine VM # 92 is installed in the IDC # 92 as a server that provides a cloud service, and is connected to the network 904 on the carrier side 902 via the communication device CE # 95. The virtual machine VM # 91, the virtual machine VM # 92, the communication device CE # 91, and the communication device CE # 95 are connected to the monitoring control device NMS # 91 via the monitoring network. The monitoring network is a small-capacity line that only transmits and receives monitoring control commands and the like.

  The carrier side 902 provides a network 904 used by various users. In the example of FIG. 17, the communication device CE # 94 on the subscriber side 901, the communication device CE # 91 on the IDC side 903, and the communication as users. The device CE # 95 is connected, and information transmitted / received between the terminal TE # 91 on the subscriber side 901 and the virtual machine VM # 91 or the virtual machine VM # 92 on the IDC side 903 is sent to the network 904 according to the destination. Transfer within. In the example of FIG. 17, the network 904 is connected to the communication device CE # 93 for connection to the subscriber side 901, the communication device CE # 92 for connection to the virtual machine VM # 91, and the virtual machine VM # 92. And a switch SW # 91 for switching to another destination is disposed between the communication device CE # 92 and the communication device CE # 93. Similarly, a switch SW # 92 is disposed between the communication device CE # 92 and the communication device CE # 96, and a switch SW # 93 is disposed between the communication device CE # 93 and the communication device CE # 96. . The communication device CE # 92, the communication device CE # 93, and the communication device CE # 96 are connected to the monitoring control device NMS # 92 via the monitoring network. The monitoring network is a small-capacity line that only transmits and receives monitoring control commands and the like.

  Thus, in the conventional communication system 900, the carrier side 902 and the IDC side 903 are operated independently, and the monitoring control device NMS # 91 on the IDC side 903 and the monitoring control device NMS # 92 on the carrier side 902 are completely different. It was not linked.

  Here, for example, a case where trouble occurs in the virtual machine VM # 91 and live migration is performed on the virtual machine VM # 92 will be described with reference to FIGS. 18, 19, and 20. FIG. 18, 19, and 20 are diagrams specifically showing connection ports of the communication devices of the communication system 900 shown in FIG. 17. In FIG. 18, the terminal TE # 91 on the subscriber side 901 is connected to the port p1 of the communication device CE # 94. Port p2 and port p3 of communication device CE # 94 are connected to port p1 and port p4 of communication device CE # 93 on carrier side 902, respectively.

  On the other hand, the port p3 of the communication device CE # 93 on the carrier side 902 is connected to the port p1 of the communication device CE # 92 via the switch SW # 91, and the port p2 of the communication device CE # 93 is connected to the switch SW # 93. To the port p2 of the communication device CE # 96. The port p2 of the communication device CE # 92 is connected to the port p1 of the communication device CE # 96 via the switch SW # 92. The port p3 and the port p4 of the communication device CE # 92 are connected to the port p1 and the port p2 of the communication device CE # 91 on the IDC side 903, respectively. Similarly, the port p3 and the port p4 of the communication device CE # 96 are connected to the port p1 and the port p2 of the communication device CE # 95 on the IDC side 903, respectively.

  On the IDC side 903, the port p3 and the port p4 of the communication device CE # 91 are connected to the virtual machine VM # 91, and the port p3 and the port p4 of the communication device CE # 95 are connected to the virtual machine VM # 92. .

  Now, it is assumed that the terminal TE # 91 on the subscriber side 901 is connected to the network 904 on the carrier side 902 via the communication device CE # 94 and receives an information providing service in the cloud environment. In this case, the subscriber side 901 does not need to be aware of which virtual machine on the IDC side 903 is accessing, but the actual access path is the virtual machine VM # on the path A indicated by the dotted arrow in FIG. 91 is accessed. The specific path A in this case is first connected from the port p2 of the communication device CE # 94 to the port p1 of the communication device CE # 93, and then from the port p3 of the communication device CE # 93 via the switch SW # 91. Connected to port p1 of CE # 92. The port p3 of the communication device CE # 92 is connected to the port p1 of the communication device CE # 91 on the IDC side 903, and the port p3 of the communication device CE # 91 is connected to the virtual machine VM # 91.

  Here, when performing live migration from the virtual machine VM # 91 to the virtual machine VM # 92, first, as shown in FIG. 19, the monitoring controller NMS # 91 transfers from the virtual machine VM # 91 to the virtual machine VM # 92. The migration data to be transferred from the virtual machine VM # 91 to the communication device CE # 91, the communication device CE # 92 on the carrier side 902, the switch SW # 92 and the communication device CE # 96 from the communication device CE # 95 on the IDC side 903 The virtual machine VM # 92 is transferred to the virtual machine VM # 92, and the virtual machine VM # 91 and the virtual machine VM # 92 are maintained in the same state with respect to the terminal TE # 91 on the subscriber side 901.

  Then, as shown in FIG. 20, the monitoring control device NMS # 91 returns from the port p1 to the port p2 with respect to the communication device CE # 91 connected to the terminal TE # 91 on the subscriber side 901 through the path A. Thus, the port p2 of the communication device CE # 91 is connected to the port p4 of the communication device CE # 92. Then, the port p2 of the communication device CE # 92 is connected to the port p1 of the communication device CE # 96 via the switch SW # 92, and the port p3 of the communication device CE # 96 is further connected to the communication device CE # 95 on the IDC side 903. Connect to the port p1 and connect to the virtual machine VM # 92 from the port p3 of the communication device CE # 95.

  Thus, conventionally, a path B from the communication device CE # 91 to the virtual machine VM # 92 is constructed, and is combined with the route A from the terminal TE # 91 on the subscriber side 901 to the communication device CE # 91. A path from the terminal TE # 91 on the subscriber side 901 to the virtual machine VM # 92 on the IDC side 903 was constructed, and live migration was performed to provide an information providing service in the cloud environment in the same manner as the virtual machine VM # 91. . In the above description, transmission data and reception data are not particularly mentioned, and each communication apparatus is connected between one port. Actually, each port for uplink and downlink is used. Are connected in the same manner, and construct route A and route B, respectively.

JP 2006-172217 A

  However, as shown in FIG. 20, since the connection path between the subscriber side 901 and the virtual machine VM # 92 is turned back at the communication device CE # 91, communication between the communication device CE # 91 on the IDC side 903 and the carrier side 902 is performed. Since data reciprocates with the device CE # 92, there arises a problem that a data delay amount and a traffic amount increase.

  Furthermore, when the connection destination is returned from the virtual machine VM # 92 to the virtual machine VM # 91, the line remains blocked by the path B as shown in FIG. Migration data cannot be transferred from the virtual machine to the virtual machine VM # 91.

  In order to solve this problem, after migration from the virtual machine VM # 91 to the virtual machine VM # 92, the route setting of the communication device CE # 94 on the subscriber side 901 is performed according to the convenience of the subscriber side 901. And a line for transferring migration data from the virtual machine VM # 92 to the virtual machine VM # 91 is secured. For example, as shown in FIG. 21, the subscriber side 901 is requested to change the route of the communication device CE # 94. In the case of the route A in FIG. 20, the port p1 of the communication device CE # 94 is connected to the port p1 of the communication device CE # 93 on the carrier side 902 via the port p2, but after the route change in FIG. Port p1 of device CE # 94 is connected to port p4 of communication device CE # 93 on carrier side 902 via port p3. The port p2 of the communication device CE # 93 is connected to the port p2 of the communication device CE # 96 via the switch SW # 93, and the port p4 of the communication device CE # 96 to the port of the communication device CE # 95 on the IDC side 903. connected to p2. And finally, it is connected from the port p4 of the communication device CE # 95 to the virtual machine VM # 92. In this manner, conventionally, the connection path from the terminal TE # 91 to the virtual machine VM # 92 along the path A and path B in FIG. 20 is changed to the path C in FIG. In order to avoid the increasing problem and to secure a line for transferring migration data from the virtual machine VM # 92 to the virtual machine VM # 91.

  However, in the conventional method described above, as shown in FIG. 20, the route change for returning the connection at the communication device CE # 91 on the IDC side 903 and the route change at the communication device CE # 94 on the subscriber side 901 are performed. Two route changes had to be performed. Moreover, live migration usually needs to be dealt with immediately, and the subscriber side 901 is not notified in advance of the migration timing and destination, and further, the setting of the communication device CE # 94 on the subscriber side 901 is changed. Requesting the user is also troublesome for the subscriber side 901 and is not desired to be performed by the IDC side 903 providing the cloud service.

  In view of the above problems, the object of the present invention is to reduce the redundancy of the route and the accompanying increase in delay and reduce the number of route changes when performing live migration between virtual machines distributed between remote locations. To provide a communication system, a communication apparatus, and a monitoring control apparatus that can realize migration.

  A communication system according to the present invention includes a plurality of first communication devices arranged for each of a plurality of IDC virtual machines, a plurality of second communication devices having a plurality of ports constituting a carrier network, and a subscriber. A third communication device for connecting a terminal on the side to the carrier network, an NW monitoring and control device for managing a plurality of second communication devices constituting the carrier network, the plurality of virtual machines distributed, and the An IDC monitoring and control device that manages a plurality of first communication devices, and a communication system that provides a live migration function between the plurality of distributed virtual machines for the third communication device, An NMS link for connecting the NW monitoring controller and the IDC monitoring controller to each other is provided, and the NW monitoring controller controls the IDC monitoring control. Information about the connection destination change of the virtual machine from the device acquired via the NMS linkage line, and changes the routing table of the second communication device.

  In particular, the second communication device holds two sets of routing tables before and after migration switching, and switches the two sets of routing tables according to a command from the NW monitoring and control device. In accordance with the migration information acquired from the IDC monitoring and control device via the NMS link line, the second communication device is notified of a table number specifying one of the two sets of routing tables. .

  Alternatively, the second communication device holds a rewritable routing table from the NW monitoring and control device, and the NW monitoring and control device holds two sets of routing tables before and after migration switching, and connects the NMS link line. The routing table of the second communication device is updated in accordance with the migration information acquired from the IDC monitoring and control device.

  Further, migration information transmitted / received between the NW monitoring control device and the IDC monitoring control device via the NMS link is transmitted / received using a predetermined protocol.

  In particular, the protocol is transmitted / received using any one of the SNMP protocols MIB, Telnet, FTP, and HTTP.

  A communication apparatus according to the present invention includes a first routing table used before migration, a second routing table used after migration, and a control unit that selects the first routing table or the second routing table. The control unit selects the first routing table or the second routing table according to the information notified from the NW monitoring and control device.

  The NW monitoring and control apparatus according to the present invention obtains migration-related information from the IDC monitoring and controlling apparatus via the NMS link line and accommodates the third communication apparatus on the subscriber side. A monitoring control unit for changing the routing table is provided.

  The communication system, communication apparatus, and monitoring control apparatus according to the present invention do not cause path redundancy and associated delay increase when live migration is performed between virtual machines distributed and distributed between remote locations. A live migration function with a low frequency can be realized.

It is a figure which shows the system image of the communication system 100 which concerns on this embodiment. It is a figure which shows the example of a connection path | route in the communication system 100 concerning this embodiment. It is a figure which shows the structural example of communication apparatus CE # 1. It is a figure which shows the structural example of the monitoring control apparatus NMS # 1. It is a figure which shows the structural example of the monitoring control apparatus NMS # 2. It is a figure which shows the example of a connection by a NMS cooperation line. It is a figure which shows the example of a path | route of the cloud service in the communication system 100 which concerns on this embodiment. It is a figure which shows the example of a connection path | route after the live migration in the communication system 100 which concerns on this embodiment. It is a flowchart which shows the flow of the route change process (method 1) in the communication system 100 which concerns on this embodiment. It is a figure which shows the image of the route change (method 1) in the communication system 100 which concerns on this embodiment. It is a figure which shows the example of a routing table (method 1) in the monitoring control apparatus NMS # 1 and the monitoring control apparatus NMS # 2 of the communication system 100 which concerns on this embodiment. It is a figure which shows the example of a routing table (method 1) in communication apparatus CE # 3 of the communication system 100 concerning this embodiment. It is a flowchart which shows the flow of the path | route change process (method 2) in the communication system 100 concerning this embodiment. It is a figure which shows the image of the route change (method 2) in the communication system 100 which concerns on this embodiment. It is a figure which shows the example (routing 2) of a routing table in the monitoring control apparatus NMS # 1 and the monitoring control apparatus NMS # 2 of the communication system 100 concerning this embodiment. It is a figure which shows the example of a routing table (method 2) in communication apparatus CE # 3 of the communication system 100 concerning this embodiment. It is a figure which shows the system image of the conventional communication system 900. FIG. It is a figure which shows the example of a connection path | route in the conventional communication system 900. FIG. It is a figure which shows the example of a path | route of the migration data transfer in the conventional communication system 900. It is a figure which shows the example of a connection path | route after the live migration in the conventional communication system 900. FIG. It is a figure which shows the example of a path | route change in the conventional communication system 900.

  Hereinafter, embodiments of a communication system, a communication apparatus, and a monitoring control apparatus according to the present invention will be described in detail with reference to the drawings.

[Overview of the entire system]
FIG. 1 is a diagram illustrating an outline of a communication system 100 according to the present embodiment, and corresponds to the communication system 900 of FIG. 17 described in the related art. In FIG. 1, a terminal TE # 1 on the subscriber side 101 is connected to the network 104 on the carrier side 102 via the communication device CE # 4 and uses a cloud service provided by the IDC side 103. On the other hand, an IDC service company (IDC side 103) that provides an IDC service different from the carrier company that manages the network 104 on the carrier side 102 transfers two IDCs, IDC # 1 and IDC # 2, to remote locations (for example, Tokyo and Yokohama). Etc.).

  The IDC # 1 on the IDC side 103 is provided with a virtual machine VM # 1 as a server that provides a cloud service, and is connected to the network 104 on the carrier side 102 via the communication device CE # 1. Similarly, a virtual machine VM # 2 is installed in IDC # 2 as a server that provides a cloud service, and is connected to the network 104 on the carrier side 102 via the communication device CE # 5. Further, the virtual machine VM # 1, the virtual machine VM # 2, the communication device CE # 1, and the communication device CE # 5 are connected to the monitoring control device NMS # 1 via the monitoring network. The monitoring network is a small-capacity line that only transmits and receives monitoring control commands and the like.

  Further, the carrier side 102 provides a network 104 to various users (subscribers). In the example of FIG. 1, the communication device CE # 4 on the subscriber side 101 and the communication device CE # 1 on the IDC side 103 are provided. And the communication device CE # 5 are connected via the network 104, and are transmitted and received between the terminal TE # 1 on the subscriber side 101 and the virtual machine VM # 1 or the virtual machine VM # 2 on the IDC side 103 Is transferred via a route formed in the network 104 according to each destination. In the example of FIG. 1, the network 104 includes a communication device CE # 3 for connecting the subscriber side 101, a communication device CE # 2 for connecting the virtual machine VM # 1, and a virtual machine VM # 2. A communication device CE # 6 for connection is arranged. Further, a switch SW # 1 for switching to another destination is provided between the communication device CE # 2 and the communication device CE # 3, and a switch SW # is provided between the communication device CE # 2 and the communication device CE # 6. 2. A switch SW # 3 is disposed between the communication device CE # 3 and the communication device CE # 6. Further, the communication device CE # 2, the communication device CE # 3, and the communication device CE # 6 are connected to the monitoring control device NMS # 2 via the monitoring network. The monitoring network is a small-capacity line that only transmits and receives monitoring control commands and the like.

  Furthermore, in the communication system 100 according to the present embodiment, a communication network (NMS link line 105) that connects the monitoring control device NMS # 2 on the carrier side 102 and the monitoring control device NMS # 1 on the IDC side 103 is constructed. Yes. This part is greatly different from the communication system 900 of FIG. 17 described in the prior art. Then, the supervisory control device NMS # 2 on the carrier side 102 and the supervisory control device NMS # 1 on the IDC side 103 cooperate with each other by transmitting and receiving information related to migration on the IDC side 103 via the NMS linkage line 105. Be able to work. On the other hand, in the conventional communication system 100 described with reference to FIG. 17, the carrier side 902 and the IDC side 903 with different operating companies are operated independently, and the monitoring control device NMS # 91 on the IDC side 903 and the carrier side The monitoring control device NMS # 92 of 902 was not linked at all.

  On the other hand, in the communication system 100 according to the present embodiment, the monitoring control device NMS # 1 on the IDC side 103 and the monitoring control device NMS # 2 on the carrier side 102 are linked by an NMS link line 105, for example, SNMP. Using the MIB (Management Information Base) of (Simple Network Management Protocol), it is possible to transmit the migration related information of the monitoring control device NMS # 1 on the IDC side 103 from the monitoring control device NMS # 1 to the monitoring control device NMS # 2. it can. Here, the migration-related information includes, for example, migration source and migration destination virtual machine information and port setting information. In the present embodiment, an example of using the MIB of SNMP has been shown. However, a standard protocol other than SNMP may be used, Telnet (Telecommunication network), FTP (File Transfer Protocol), HTTP (Hyper Text Transfer Protocol). An existing protocol such as may be used. Alternatively, it may be transmitted by a predetermined original protocol or a simple text file. In any case, if the migration related information of the supervisory control device NMS # 1 on the IDC side 103 can be transmitted from the supervisory control device NMS # 1 to the supervisory control device NMS # 2 via the NMS link line 105, the same effect as this embodiment is achieved. Is obtained.

  In this way, when performing live migration from the virtual machine VM # 1 to the virtual machine VM # 2, it is possible to change the path of each communication device managed by the monitoring control device NMS # 2 on the carrier side 102, The migration path construction can be appropriately performed as described later. As a result, it is possible to provide a live migration function in which route redundancy and associated delay increase do not occur, and the number of route changes is small.

[Port connection example of communication system 100]
FIG. 2 is a diagram illustrating a port connection example of the communication system 100 according to the present embodiment illustrated in FIG. 1, and corresponds to the communication system 900 illustrated in FIG. 18 described in the related art. 1 denote the same components as those in FIG.

  On the subscriber side 101, the terminal TE # 1 is connected to the port p1 of the communication device CE # 4, and the port p2 and the port p3 of the communication device CE # 4 are the port p1 and the port of the communication device CE # 3 on the carrier side 102 Each is connected to p4.

  On the carrier side 102, the port p3 of the communication device CE # 3 is connected to the port p1 of the communication device CE # 2 via the switch SW # 1, and the port p2 of the communication device CE # 3 is connected via the switch SW # 3. To the port p2 of the communication device CE # 6. The port p2 of the communication device CE # 2 is connected to the port p1 of the communication device CE # 6 via the switch SW # 2. The port p3 and the port p4 of the communication device CE # 2 are connected to the port p1 and the port p2 of the communication device CE # 1 on the IDC side 103, respectively. Similarly, the port p3 and the port p4 of the communication device CE # 6 are connected to the port p1 and the port p2 of the communication device CE # 5 on the IDC side 103, respectively.

  On the IDC side 103, the port p3 and the port p4 of the communication device CE # 1 are connected to the virtual machine VM # 1, and the port p3 and the port p4 of the communication device CE # 5 are connected to the virtual machine VM # 2.

  In the communication system 100 of FIG. 2, unlike the communication system 900 of FIG. 18, the monitoring control device NMS # 1 on the IDC side 103 and the monitoring control device NMS # 2 on the carrier side 102 are connected via the NMS link line 105. Connected. As a result, the migration related information can be shared between the IDC side 103 and the carrier side 102, and the terminal TE # 1 on the subscriber side 101 connects to the IDC side 103 in accordance with live migration performed by the IDC side 103. The route in side 102 can be changed automatically.

[Configuration of communication device]
Next, the configuration of the communication device will be described. Here, the configuration of the communication device CE # 3 will be described, but other communication devices are configured in the same manner as the communication device CE # 3. FIG. 3 is a block diagram illustrating a configuration example of the communication device CE # 3. In FIG. 3, the communication device CE # 3 includes a CPU board 201, a switch board 202, and an IF board 203.

  The CPU board 201 is a package for controlling the overall operation of the communication device CE # 3 and controls the switch board 202 and the IF board 203. The switch board 202 has a switch section 271 that is controlled by the control section 252 of the CPU board 201 and switches connection paths between the ports of the IF board 203. The IF panel 203 has four physical ports, for example, port p1, port p2, port p3, and port p4, as communication interfaces, and the connection path between the ports is changed by the switch panel 202. In FIG. 3, only four ports are configured for easy understanding, but an actual communication apparatus has more ports. Also, the ports p1 to p4 in FIG. 3 correspond to the ports p1 to p4 in FIG.

  The CPU board 201 is connected to the monitoring control device NMS # 2 via a monitoring network, and transmits and receives monitoring control signals to and from the monitoring control device NMS # 2. The CPU board 201 includes, for example, a monitoring interface unit 251, a control unit (CPU) 252, and a DB 253 that stores a routing table and the like. The monitoring interface unit 251 is an interface for connecting to the monitoring control device NMS # 2 via the monitoring network. The control unit 252 is composed of a CPU, and operates according to a program stored in advance inside. For example, in the communication system 100 according to the present embodiment, the monitoring control device NMS # 2 includes the SNMP manager 261, and the control unit 252 of the communication device CE # 3 includes the SNMP agent 262. Then, for example, using the SNMP MIB, the monitoring control device NMS # 2 rewrites the routing table stored in the DB 253 of the communication device CE # 3.

[Configuration of monitoring and control device]
Next, the configuration of the monitoring control device will be described. FIG. 4 is a block diagram illustrating a configuration example of the monitoring control device NMS # 1. 4, the monitoring control device NMS # 1 includes, for example, a monitoring control unit 301, a monitoring interface unit 302, and a DB 303. The monitoring control unit 301 is constituted by a CPU and operates according to a program stored therein in advance. For example, the monitoring control device NMS # 1 has an SNMP manager 351 for monitoring and controlling each of the subordinate devices via the monitoring network on the IDC side 103 connected by the monitoring interface unit 302, and the communication described in FIG. Similarly to the device CE # 3, a monitoring control signal is transmitted and received between the communication device CE # 1, the communication device CE # 5, the virtual machine VM # 1, and the SNMP agent 262 on the virtual machine VM # 2 side. Then, for example, using the SNMP MIB, the monitoring control device NMS # 1 rewrites the routing table stored in the DB 253 of the communication device CE # 1 or the communication device CE # 5. The operation of the monitoring control device NMS # 1 is performed by the operator through the operation terminal 304. Further, the communication device CE # 1 and the communication device CE # 5 are configured similarly to the communication device CE # 3 described in FIG. In addition, the monitoring control device NMS # 1 performs monitoring control on the virtual machine VM # 1 and the virtual machine VM # 2 using the SNMP protocol similarly to the communication device CE # 1 and the communication device CE # 5. In particular, in the present embodiment, the DB 303 manages migration-related information when performing live migration between the virtual machine VM # 1 and the virtual machine VM # 2, and the monitoring control device on the carrier side 102 via the NMS link 105 Migration related information is transmitted to NMS # 2.

  FIG. 5 is a block diagram showing a configuration example of the monitoring control device NMS # 2. The monitoring control device NMS # 2 has basically the same configuration as the monitoring control device NMS # 1 in FIG. In FIG. 5, the monitoring control device NMS # 2 includes, for example, a monitoring control unit 401, a monitoring interface unit 402, and a DB 403. The monitoring control unit 401 is composed of a CPU and operates according to a program stored in advance inside. For example, the monitoring control device NMS # 2 has an SNMP manager 451 for monitoring and controlling each of the devices under control via the monitoring network on the carrier side 102 connected by the monitoring interface unit 402, and the communication described in FIG. Similarly to the device CE # 3, a monitoring control signal is transmitted to and received from the SNMP agent 262 of the control unit 252 such as the communication device CE # 2, the communication device CE # 3, and the communication device CE # 6 in the carrier side 102. Then, for example, using the SNMP MIB, the monitoring control device NMS # 2 rewrites the routing table stored in the DB 253 of each communication device such as the communication device CE # 3. The operation of the monitoring control device NMS # 2 is performed by the operator through the operation terminal 404. Further, particularly in the present embodiment, the migration related information for performing live migration from the monitoring control device NMS # 1 on the IDC side 103 via the NMS link line 105 is managed by the DB 403, and the received migration related information is included in the received migration related information. In response, the monitoring control unit 401 performs processing for changing the routing table of each communication device under its control.

  Next, in FIG. 6, the monitoring control device NMS # 1 on the IDC side 103 described in FIG. 4 and the monitoring control device NMS # 2 on the carrier side 102 described in FIG. 5 cooperate with each other via the NMS cooperation line 105. It is the figure which showed the example of a structure in the case. In FIG. 6, the same reference numerals as those in FIGS. 4 and 5 denote the same components.

  In FIG. 6, in the communication system 100 according to the present embodiment, the monitoring control device NMS # 1 and the monitoring control device NMS # 2 are connected via an NMS link line 105. In FIG. 6, the NMS link line 105 is connected between the monitoring interface unit 302 of the monitoring control device NMS # 1 and the monitoring interface unit 402 of the monitoring control device NMS # 2. An interface may be provided to transmit and receive migration related information between the monitoring control unit 301 and the monitoring control unit 401. 4 and 5, the monitoring control unit 301 and the monitoring control unit 401 are provided with only the SNMP manager 351 and the SNMP manager 451. However, the monitoring is performed so that the SNMP protocol can be used via the NMS link 105. The SNMP agent 352 or the SNMP agent 452 is arranged in either the control unit 301 or the monitoring control unit 401 so that communication by the SNMP protocol can be performed between the monitoring control device NMS # 1 and the monitoring control device NMS # 2. ing. Then, the migration related information is transmitted from the monitoring control device NMS # 1 on the IDC side 103 to the monitoring control device NMS # 2 on the carrier side 102 using the SNMP protocol.

  In this way, the monitoring control device NMS # 2 on the carrier side 102 can acquire the migration related information from the monitoring control device NMS # 1 on the IDC side 103. Then, based on the acquired migration related information, for example, the monitoring control device NMS # 2 changes the path in the carrier side 102 when performing live migration from the virtual machine VM # 1 to the virtual machine VM # 2 in FIG. be able to.

[Example of route change during live migration]
Next, an example of route change during live migration will be described with reference to FIGS. 7 and 8, the same reference numerals as those in FIG. 2 denote the same components.

  In FIG. 7, it is assumed that the terminal TE # 1 on the subscriber side 101 is connected to the network 104 on the carrier side 102 via the communication device CE # 4 and receives an information providing service in the cloud environment. As the first access route, a route A indicated by a dotted arrow is formed as in the conventional example of FIG. 18, and the terminal TE # 1 is accessing the virtual machine VM # 1. In FIG. 7, the path A is first connected from the port p2 of the communication device CE # 4 to the port p1 of the communication device CE # 3 on the carrier side 102, and from the port p3 of the communication device CE # 3 via the switch SW # 1. It is connected to port p1 of communication device CE # 2. The port p3 of the communication device CE # 2 is connected to the port p1 of the communication device CE # 1 on the IDC side 103, and further connected to the virtual machine VM # 1 from the port p3 of the communication device CE # 1.

  Here, in the case of performing live migration from the virtual machine VM # 1 to the virtual machine VM # 2, as in the case described in FIG. 19 of the prior art, the monitoring control device NMS # 1 changes from the virtual machine VM # 1 to the virtual machine. The migration data to be transferred to the VM # 2 is communicated from the virtual machine VM # 1 to the communication device CE # 1, the communication device CE # 2 on the carrier side 102, the switch SW # 2, and the communication device CE # 6. The data is transferred from the device CE # 5 to the virtual machine VM # 2, and the virtual machine VM # 1 and the virtual machine VM # 2 are maintained in the same state with respect to the terminal TE # 1 on the subscriber side 101.

  Then, the monitoring control device NMS # 1 transmits the migration related information to the monitoring control device NMS # 2 on the carrier side 102 via the NMS link 105. Then, the monitoring control device NMS # 2 recognizes the change of the connection destination of the terminal TE # 1 from the port p1 of the communication device CE # 1 to the port p2 of the communication device CE # 5 based on the received migration related information. The supervisory control device NMS # 2 connects the port p1 of the communication device CE # 3 connected to the communication device CE # 4 of the subscriber side 101 to the port p3 from the port p1 of the communication device CE # 3 to the port p3. The routing table of the DB 253 of the communication device CE # 3 is changed so that the connection to p2 is changed. As a result, the route formed from the port p1 of the communication device CE # 3 to the port p3 of the communication device CE # 2 is changed to a route from the port p2 of the communication device CE # 3 to the port p4 of the communication device CE # 6. To do.

  In this way, as shown in FIG. 8, a path D from the terminal TE # 1 on the subscriber side 101 to the virtual machine VM # 2 on the IDC side 103 is formed. This eliminates the need for a route change in the communication device CE # 4 corresponding to the communication device CE # 94 on the subscriber side 901 as described with reference to FIG. Cloud services can be used without being aware of migration switching.

[Route change example 1 during live migration]
Next, an example 1 of the route change method at the time of live migration described with reference to FIGS. 7 and 8 will be described with reference to the flowchart of FIG.

  (Step S101) The monitoring control device NMS # 1 on the IDC side 103 determines whether or not it is necessary to migrate from the virtual machine VM # 1 to the virtual machine VM # 2. For example, when a failure occurs in the virtual machine VM # 1, it is determined that migration to the virtual machine VM # 2 needs to be performed. If migration is to be performed, the process proceeds to step S102; otherwise, the process waits in step S101.

  (Step S102) The monitoring control device NMS # 1 on the IDC side 103 transfers the migration data from the virtual machine VM # 1 to the virtual machine VM # 2. For example, the route described in FIG. 19 is used as the transfer route.

  (Step S103) The supervisory control device NMS # 1 on the IDC side 103 monitors and controls the carrier side 102 via the NMS link line 105 with the communication device number whose connection destination is changed and the table number to be selected in accordance with the migration. Notify device NMS # 2.

  Here, it is assumed that two types of routing tables are stored in the DB 253 of the communication device CE # 3. For example, FIG. 10 is a diagram illustrating an example of the routing table of the communication device CE # 3 stored in the DB 253. In the routing table of FIG. 10, ports with the same VLAN number are connected. In the case of Table 1, the port p1 and the port p3 of the communication device CE # 3 are connected with the VLAN number “V1”, and the port p2 and the port p4. Are connected by the VLAN number “V2”. In the case of Table 2, the port p1 and the port p2 of the communication device CE # 3 are connected by the VLAN number “V2”, and the port p3 and the port p4 are connected by the VLAN number “V1”.

  Further, as shown in FIG. 11, the DB 303 of the monitoring control device NMS # 1 and the DB 403 of the monitoring control device NMS # 2 store table numbers to be selected for each communication device before and after the live migration switching. ing. These tables are set in advance by the operator. In the example of FIG. 11, the communication device CE # 1 is the communication device on the IDC side 103, and it can be seen that the table number before switching is Table 1, and the table number after switching is Table 2. Similarly, it can be seen that the communication device CE # 2 and the communication device CE # 3 are communication devices on the carrier side 102, and the table number before switching is Table 1, and the table number after switching is Table 2.

  In this way, the supervisory control device NMS # 1 and the supervisory control device NMS # 2 manage the table numbers of the communication devices under their control, and the subordinates of both supervisory control devices via the NMS link line 105. Share and manage communication device table numbers.

  (Step S104) The monitoring control device NMS # 2 on the carrier side 102 instructs the communication device that needs to change the routing path, via the monitoring network, the table number to be used. Before live migration switching, it is assumed that table 1 is instructed to communication device CE # 3, and port p1 and port p3 are connected as VLAN number “V1” according to the routing table of FIG. 12 (a)). This corresponds to the path A before the live migration switching, as described with reference to FIG. In step S103, the monitoring control device NMS # 2 instructs the communication device CE # 3 to specify the table 2.

  (Step S105) The communication device CE # 3 connects the port p1 and the port p2 as the VLAN number “V2” according to the routing table of FIG. This is shown in FIG. This corresponds to the route D after the live migration switching as described with reference to FIG.

  In this way, a path D is formed from the terminal TE # 1 on the subscriber side 101 to the virtual machine VM # 2 on the IDC side 103, and the communication device CE on the subscriber side 901 as described with reference to FIG. It is not necessary to change the route in the communication device CE # 4 corresponding to # 94, the burden on the subscriber side 101 is reduced, and the terminal TE # 1 can use the cloud service without being aware of switching of live migration. .

[Example 2 of route change during live migration]
Next, an example 2 of the route change method at the time of live migration described with reference to FIGS. 7 and 8 will be described with reference to the flowchart of FIG. In the case of Example 1 described in FIG. 9, two types of routing tables before and after the live migration switching are stored in advance in the DB 253 of the communication device CE # 3, and the monitoring control device NMS # 2 instructs the table switching. However, in Example 2 in FIG. 13, only one routing table is stored in the DB 253 of the communication device CE # 3, and the monitoring control device NMS # 2 is used during live migration. By updating the table in DB 253, the route A in FIG. 7 is switched to the route D in FIG. Hereinafter, it demonstrates in order according to the flowchart of FIG.

  (Step S201) Similar to step S101 in FIG. 9, the monitoring control device NMS # 1 on the IDC side 103 determines whether or not it is necessary to migrate from the virtual machine VM # 1 to the virtual machine VM # 2. If migration is to be performed, the process proceeds to step S202. If not, the process waits in step S201.

  (Step S202) Similar to step S102 of FIG. 9, the monitoring control device NMS # 1 on the IDC side 103 transfers migration data from the virtual machine VM # 1 to the virtual machine VM # 2.

  (Step S <b> 203) The monitoring control device NMS # 1 on the IDC side 103 monitors and controls the carrier side 102 via the NMS link line 105 with the communication device number to which the connection destination is changed and the port number setting information in accordance with the migration. Notify device NMS # 2.

  Here, the port number setting information is information indicating the VLAN number set for each port. In the DB 303 of the monitoring control device NMS # 1 and the DB 403 of the monitoring control device NMS # 2, as shown in FIG. 14, the VLAN numbers set to the respective ports before and after the live migration switching are stored for each communication device, respectively. Has been. These tables are set in advance by the operator. In the example of FIG. 14, the VLAN number before switching of the port p1 of the communication device CE # 3 is “V1”, and the VLAN number after switching is “V2”. Similarly, the port p2 and the port 3 have the same VLAN number before and after switching, and are “V2” and “V1”, respectively.

  On the other hand, the routing table as shown in FIG. 15A is stored in the DB 253 of the communication device CE # 3 in the DB 253 of the communication device CE # 3. For example, the routing table shown in FIG. 15A stores the routing table before the live migration switching described in FIG. 14, and the port p1 and the port p3 of the communication device CE # 3 have the same VLAN number “V1”. As shown in FIG. 16A, for example, information to which the VLAN number “V1” is added is transmitted and received between the port p1 and the port p3. Similarly, when the routing table after the live migration switching described in FIG. 14 is written in the DB 253 of the communication device CE # 3, the port p1 and the port p2 of the communication device CE # 3 are the same VLAN number “V2”. As shown in FIG. 16B, for example, information to which the VLAN number “V2” is added is transmitted and received between the port p1 and the port p2.

  In this way, the monitoring control device NMS # 1 and the monitoring control device NMS # 2 manage the VLAN number set for each port number of the communication devices under their control, and both of them via the NMS link line 105. The port number and VLAN number of the communication device under the supervisory control device are shared and managed.

  (Step S204) The monitoring control device NMS # 2 on the carrier side 102 notifies the port number setting information to the communication device that needs to change the routing path via the monitoring network. 14 and FIG. 15, the VLAN number of the port p1 of the communication device CE # 3 is “V1” before the live migration switching, and the VLAN number of the port p1 of the communication device CE # 3 is “V1” after the live migration switching. V2 ".

  (Step S205) The communication device CE # 3 connects the port p1 and the port p2 according to the routing table of FIG. 15B (VLAN number “V2” of the port p1). This is shown in FIG. This corresponds to the route D after the live migration switching as described with reference to FIG.

  In this way, a path D is formed from the terminal TE # 1 on the subscriber side 101 to the virtual machine VM # 2 on the IDC side 103, and the communication device CE on the subscriber side 901 as described with reference to FIG. It is not necessary to change the route in the communication device CE # 4 corresponding to # 94, the burden on the subscriber side 101 is reduced, and the terminal TE # 1 can use the cloud service without being aware of switching of live migration. .

  As described above, the communication system 100, the communication device CE # 3, the monitoring control device NMS # 1, and the monitoring control device NMS # 2 according to the present embodiment are arranged between virtual machines that are distributed between remote locations. When performing live migration, it is possible to realize a live migration function that does not increase the redundancy of the route and the accompanying delay and reduces the number of route changes.

  The communication system, communication apparatus, and monitoring control apparatus according to the present invention have been described with reference to the respective embodiments. However, the communication system, the communication apparatus, and the monitoring control apparatus may be implemented in various other forms without departing from the spirit or main features thereof. it can. For this reason, the above-described embodiment is merely an example in all respects and should not be interpreted in a limited manner. The present invention is defined by the claims, and the present invention is not limited to the text of the specification. Further, all modifications and changes belonging to the equivalent scope of the claims are within the scope of the present invention.

100, 900 ... communication system 101, 901 ... subscriber side 102, 902 ... carrier side 103, 903 ... IDC side 104, 904 ... network 105 ... NMS link 201 ... CPU board 202... Switch board 203 .. IF board 251, 302, 402... Monitoring interface unit 252... Control unit 253, 303, 403.
262, 352, 452 ... SNMP agents 301, 401 ... monitoring control units 304, 404 ... operation terminals 351, 451 ... SNMP manager TE # 1 ... terminals CE # 1, CE # 2, CE # 3, CE # 4, CE # 5, CE # 6 ... Communication devices NMS # 1, NMS # 2 ... Monitoring and control devices VM # 1, VM # 2 ... Virtual machines SW # 1, SW # 2, SW # 3 ... Switch

Claims (7)

A plurality of first communication devices arranged for each of a plurality of virtual machines for IDC;
A plurality of second communication devices having a plurality of ports constituting a carrier network;
A third communication device for connecting a subscriber-side terminal to the carrier network;
An NW monitoring and control device that manages a plurality of second communication devices constituting the carrier network;
An IDC monitoring and control device that manages the plurality of virtual machines and the plurality of first communication devices distributed between the plurality of virtual machines distributed to the third communication device. In a communication system that provides a live migration function,
An NMS link for connecting the NW monitoring control device and the IDC monitoring control device to each other is provided,
The NW monitoring and control device acquires information related to the change of the connection destination of the virtual machine from the IDC monitoring and control device via the NMS link line, and changes the routing table of the second communication device. Communications system. The communication system according to claim 1,
The second communication device holds two sets of routing tables before and after migration switching, and switches between the two sets of routing tables in response to a command from the NW monitoring and control device.
The NW monitoring and control apparatus assigns a table number for designating one of the two sets of routing tables to the second communication apparatus according to the migration information acquired from the IDC monitoring and control apparatus via the NMS link line. A communication system characterized by notifying. The communication system according to claim 1,
The second communication device holds a rewritable routing table from the NW monitoring and control device,
The NW monitoring and control device holds two sets of routing tables before and after migration switching, and according to the migration information acquired from the IDC monitoring and control device via the NMS link line, the routing table of the second communication device The communication system characterized by updating. The communication system according to any one of claims 1 to 3,
Migration information transmitted / received between the NW monitoring control device and the IDC monitoring control device via the NMS link is transmitted / received using a predetermined protocol. The communication system according to claim 4,
The protocol is transmitted / received using any one of the SNMP protocols MIB, Telnet, FTP, and HTTP. In the second communication device used in the communication system according to any one of claims 2 to 5,
A first routing table to use before migration;
A second routing table to use after migration;
A controller for selecting the first routing table or the second routing table;
The control unit selects a first routing table or a second routing table according to information notified from the NW monitoring control device. In the NW monitoring and control device used in the communication system according to any one of claims 2 to 5,
A monitoring control unit that acquires migration-related information from the IDC monitoring control device via the NMS link and changes a routing table of the second communication device that accommodates the third communication device on the subscriber side; An NW monitoring and control device characterized by being provided.

Images