JP2011209811A - Virtual machine system and virtual machine arrangement method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To rearrange a virtual machine inside a virtual machine system in another virtual machine system according to a disaster.SOLUTION: This virtual machine system including a virtual machine server 12 wherein virtual machines 10, 11 are constructed, and connected to other virtual machine systems 3, 4 via a network 2 includes: a disaster detection part 13 detecting the prescribed disaster from disaster information and notifying information about the detected disaster; a storage part 14 storing processing setting information registered with coping information corresponding to the kind of the disaster and system information in the virtual machine systems 3, 4 of the virtual machine server that is a candidate of a rearrangement destination of the virtual machine; a coping determination part 15 determining the virtual machine 10 to be rearranged and a virtual machine server 31 of the rearrangement destination based on the notified information about the disaster, the processing setting information, and the system information; and a server manager part 16 making the virtual machine server 12 rearrange the virtual machine 10 in the virtual machine server 31 of the rearrangement destination.

Description

  The present invention relates to a virtual machine system and a virtual machine arrangement method for rearranging virtual machines.

  There are cases where a company distributes a large number of servers used for business in a plurality of geographically separated data centers. By distributing various servers in each data center, even if some data centers cannot be used due to problems, etc., operations can be continued using other data centers. Thus, the availability of the entire system can be increased.

  However, when some data centers cannot be used, the load on the data center used instead increases and the processing capacity and redundancy of the server decrease, which is not preferable for system operation. . In addition, when a natural disaster such as an earthquake or a volcanic eruption that may have a long-term effect occurs, the reliability of the server in the data center in the disaster occurrence area is significantly reduced. Therefore, when it is considered that there will be a problem in the continuity of the latest system operation in response to a report or prediction of a disaster or the like, it is desirable to quickly relocate the virtual machine to another server or data center. In other words, it is desirable to perform virtual machine migration and the like to perform virtual machine relocation in response to a disaster report or prediction.

  Patent Document 1 describes a virtual machine management apparatus that performs migration control considering not only a migration policy for optimizing the entire system but also individual application software. In the virtual machine management apparatus described in Patent Document 1, a user can define a migration policy that depends on application software. Therefore, migration of a virtual machine can be realized more flexibly than before.

JP 2009-116852 A (paragraphs 0014-0017)

  However, in order to avoid the influence of the system in response to a report or prediction of a disaster or the like, it is necessary to determine which virtual machine is to be migrated, but virtual machine management described in Patent Document 1 The apparatus determines a migration destination of a virtual machine in migration, and does not determine a virtual machine to perform migration.

  Accordingly, an object of the present invention is to provide a virtual machine system and a virtual machine arrangement method that can relocate a virtual machine in a virtual machine system to another virtual machine system according to a report or prediction of a disaster or the like. And

  A virtual machine system according to the present invention includes a virtual machine server on which a virtual machine is constructed and is connected to another virtual machine system via a network, and detects a predetermined disaster from disaster information. In the virtual machine system including the disaster detection unit for notifying the detected disaster information, the processing setting information in which the processing corresponding to the type of disaster is registered, and the virtual machine server that is a candidate for the relocation destination of the virtual machine A virtual unit that performs relocation based on a storage unit that stores system information related to hardware and software, disaster information notified by the disaster detection unit, and processing setting information and system information stored in the storage unit A virtual machine is constructed, and a correspondence determination unit that determines the machine and the virtual machine server to which the virtual machine is to be relocated That the virtual machine server, as determined by the relationship determining unit, characterized by comprising a server manager unit for relocating a virtual machine server relocation destination virtual machine.

  A virtual machine arrangement method according to the present invention is a virtual machine arrangement method in a virtual machine system that includes a virtual machine server in which a virtual machine is constructed and is connected to another virtual machine system via a network, A predetermined disaster is detected, information on the detected disaster is notified, the notified disaster information, processing setting information in which a corresponding process corresponding to the type of disaster stored in the storage unit is registered, virtual Based on system information on hardware and software in a virtual machine system including a virtual machine server that is a candidate for a machine relocation destination, a virtual machine to be relocated and a virtual machine server to which the virtual machine is relocated Determine and relocate the virtual machine to the virtual machine server where the virtual machine is built according to the decision. And characterized in that repositioned over server.

  According to the present invention, a virtual machine in a virtual machine system can be relocated to another virtual machine system in response to a report or prediction of a disaster or the like.

It is a block diagram which shows the structure of the communication system containing the data center which is 1st Embodiment of the virtual machine system by this invention. It is a block diagram which shows the structure of the data center shown in FIG. It is a block diagram which shows the structure of the virtual machine management system shown in FIG. It is explanatory drawing which shows an example of virtual machine priority information. It is explanatory drawing which shows an example of data center information. It is explanatory drawing which shows an example of the information of the virtual machine server registered into data center information. It is explanatory drawing which shows an example of the disaster registered into disaster response information. It is explanatory drawing which shows an example of the response | compatibility with respect to the disaster registered into disaster response information. It is a sequence diagram which shows the operation example of the data center shown in FIG. It is a block diagram which shows the principal part of the virtual machine system by this invention.

Embodiment 1. FIG.
FIG. 1 is a block diagram showing a configuration of a communication system including a data center which is a first embodiment (embodiment 1) of a virtual machine system according to the present invention. With reference to FIG. 1, a configuration of a communication system including a data center 100, which is a first embodiment of a virtual machine system according to the present invention, will be described.

  As shown in FIG. 1, a plurality of data centers 100, 200, 300, and 400 are connected to a network 500 via respective gateways (GWs) 110, 210, 310, and 410. Each data center is geographically distributed and includes one or more virtual machine servers. The data centers 100, 200, 300, and 400 are connected to each other so as to communicate with each other via the respective GWs and the network 500. The network 500 is, for example, a wide area network (WAN).

  FIG. 2 is a block diagram showing a configuration of the data center shown in FIG. The configuration of the data center 100 will be described with reference to FIG. The configuration of the data centers 200, 300, and 400 shown in FIG. 1 is the same as that of the data center 100 shown in FIG.

  The data center 100 includes a GW 110, a virtual machine management system 120, virtual machine servers 130 and 140, a disaster detection device 150, a storage device 160, a local area network 170, and a storage device network 180. Note that a part of the virtual machine management system 120 and the disaster detection apparatus 150 is realized by, for example, a computer that operates according to control of a predetermined program.

  The GW 110 relays a local area network 170 which is a communication network in the data center 100 and a network 500 used between the data centers.

  The virtual machine management system 120, the virtual machine servers 130 and 140, and the disaster detection device 150 are connected via a local area network 170 so that they can communicate with each other. Further, the virtual machine management system 120, the virtual machine servers 130 and 140, and the disaster detection device 150 communicate with other data centers 200, 300, and 400 via the GW 110 and the network 500 from the local area network 170. Can do.

  The virtual machine management system 120 issues an instruction to the virtual machine server based on the disaster information notified from the disaster detection apparatus 150 and rearranges the virtual machine servers 130 and 140 in a data center other than the data center 100. Specifically, for example, the virtual machine management system 120 according to the first embodiment issues an instruction to the virtual machine server agent 131 of the virtual machine server 130 and uses the existing technology such as live migration to set the virtual machine 132. The virtual machine 132 is rearranged by being moved onto a virtual machine server in a data center other than the data center 100. The detailed configuration of the virtual machine management system 120 will be described later with reference to FIG.

  The virtual machine server 130 includes a virtual machine server agent 131 and virtual machines 132 and 133. The virtual machines 132 and 133 are virtual machines constructed on the virtual machine server 130 and are controlled by the virtual machine server agent 131. For example, the virtual machine server agent 131 can perform control of moving the virtual machines 132 and 133 onto the virtual machine servers of other data centers 200, 300, and 400 connected via the network 500. That is, it has a function of moving a virtual machine.

  Note that the virtual machine server agent 131 may control the virtual machines 132 and 133 on the virtual machine server 130 in accordance with an instruction from the virtual machine management system 120.

  The virtual machine server 140 includes a virtual machine server agent 141 and virtual machines 142 and 143. The virtual machines 142 and 143 are virtual machines constructed on the virtual machine server 140 and are controlled by the virtual machine server agent 141. For example, the virtual machine server agent 141 can perform control of moving the virtual machines 142 and 143 onto the virtual machine servers of other data centers 200, 300, and 400 connected via the network 500.

  Note that the virtual machine server agent 141 may control the virtual machines 142 and 143 on the virtual machine server 140 in accordance with an instruction from the virtual machine management system 120.

  The virtual machine servers 130 and 140 are connected to the storage device 160 via the storage device network 180 and use the storage device 160 as a storage area for the virtual machines 132, 133, 142, and 143.

  The disaster detection apparatus 150 collects disaster information and notifies the virtual machine management system 120 of the detected disaster when a disaster that is estimated to affect the data center 100 from the collected disaster information is detected. The disaster detection device 150 includes, as disaster information, weather information such as weather warnings, weather warnings, earthquake early warnings, and volcano warnings, information on power transmission provided by electric power companies, information on network failures provided by communication companies, etc. Are collected via the network 500, for example.

  FIG. 3 is a block diagram showing a configuration of the virtual machine management system shown in FIG. As shown in FIG. 3, the virtual machine management system 120 includes a virtual machine server manager 121, a control unit 122, a virtual machine priority information DB 123, a data center information DB 124, and a disaster response information DB 125.

  The virtual machine server manager 121 communicates with the virtual machine server agents 131 and 141 included in the virtual machine servers 130 and 140 and manages the operation of the virtual machine server agent. For example, the virtual machine server manager 121 issues an instruction to the virtual machine server agent 131 to control the movement function of the virtual machine included in the virtual machine server agent 131, and the virtual machines 132 and 133 are changed to virtual machines in other data centers. Relocate on the server.

  The control unit 122 controls the virtual machine server manager 121 based on information stored in the virtual machine priority information DB 123, the data center information DB 124, and the disaster response information DB 125. Further, the control unit 122 communicates with a virtual machine server manager (not shown) existing in a virtual machine management system (not shown) of another data center, thereby managing the virtual machines of the other data centers. Information can be obtained, provided, or exchanged with the system.

  The virtual machine priority information DB 123 stores virtual machine priority information indicating the movement priority of the virtual machines 132, 133, 142, and 143 in the data center 100. For the virtual machine priority information, for example, the priority level is set by the user. The priority is expressed by comparable information such as a numerical value.

  FIG. 4 is an explanatory diagram of an example of virtual machine priority information. The virtual machine priority information table shown in FIG. 4 includes virtual machine IDs and their priorities. Specifically, priorities indicated by “high”, “medium”, and “low” are set for the virtual machines 132, 133, 142, and 143 in the data center 100.

  The data center information DB 124 stores data center information indicating system information related to hardware and software of the data centers 100, 200, 300, and 400 that are communicably connected via the network 500. The data center information includes, for example, data center location information, power allowance, performance and number of virtual machine servers held by the data center, network capacity, etc. as static information (system information related to hardware). . The data center information includes, as dynamic information (system information related to software), for example, another data center (for example, the data centers 200, 300, etc.) by the control unit (for example, the control unit 122) of the virtual machine management system. 400), the state of another data center, information on each virtual machine server, and the like acquired from the virtual machine management system 400) are included.

  FIG. 5 is an explanatory diagram showing an example of data center information. The data center information table shown in FIG. 5 includes a data center ID that identifies a data center, a data center position, an allowable power amount, and a data center state. Specifically, information of each item is registered for each of the data centers 100, 200, 300, and 400.

  FIG. 6 is an explanatory diagram illustrating an example of virtual machine server information registered in the data center information. The virtual machine server information table shown in FIG. 6 includes the virtual machine ID that identifies the virtual machine server, the data center ID that identifies the data center that includes the virtual machine server, the CPU of the virtual machine server, the number of CPUs, and the CPU It is configured by the usage rate, memory mounting amount, memory usage amount, and power consumption. Specifically, information is registered for all virtual machine servers in the data centers 100, 200, 300, and 400 that can exchange information via the network 500.

  The disaster response information DB 125 stores disaster response information indicating a disaster and a response to the disaster. FIG. 7 is an explanatory diagram illustrating an example of a disaster registered in the disaster response information. The disaster response information table shown in FIG. 7 includes a disaster ID, a disaster content, and a process ID that identifies a corresponding process. The types of disasters roughly classify disasters such as fire, electric power, weather, and earthquake. In the disaster contents, the disasters are classified in more detail according to the disaster level detected and notified by the disaster detection device 150.

  FIG. 8 is an explanatory diagram showing an example of a response to a disaster registered in the disaster response information. The disaster response information table shown in FIG. 8 includes a process ID and a response process indicating a specific response corresponding to the process ID. Specifically, for example, when the process ID is “4”, a corresponding process for “save a virtual machine having a high priority of“ high ”to another data center for 10 minutes” is registered. . Referring to FIG. 4 and FIG. 7, in the case where the disaster detection device 150 detects the occurrence of a fire at a neighboring facility of the data center 100 from the collected disaster information, the control unit 122 performs the virtual machine server manager. 121 indicates that the virtual machines 132 and 142 whose priority is “high” in the virtual machine priority information are moved to another data center.

  In addition, since the response | compatibility which should respond to every disaster changes with the location conditions of a data center, etc., different disaster response information may be registered for every data center.

  FIG. 9 is a sequence diagram showing an operation example of the data center shown in FIG. With reference to FIG. 9, an operation in which the data center 100 moves a virtual machine when a disaster is detected by the disaster detection apparatus 150 will be described.

  First, the disaster detection apparatus 150 collects disaster information and, when a disaster that affects the data center 100 is detected, notifies the detected disaster to the virtual machine management system 120 (step S1).

  The control unit 122 of the virtual machine management system 120 receives a disaster notification from the disaster detection apparatus 150. The control unit 122 refers to the disaster response information DB 125 (step S2), and checks whether or not a response process corresponding to the disaster indicated in the received notification is registered (step S3). When the response process corresponding to the disaster indicated in the received notification is registered, the control unit 122 acquires the response process from the disaster response information DB 125. In the response process, when a condition such as after a certain time has elapsed is specified, the control unit 122 obtains the response process after waiting until the condition is satisfied.

  The control unit 122 refers to the virtual machine priority information DB 124 and determines a virtual machine (for example, the virtual machine 132) that is the target of the response process according to the content of the response process acquired from the disaster response information DB 125 (step virtual machine 132). S4).

  Next, the control unit 122 refers to the data center information DB 124 and determines a data center (for example, the data center 200) to which the virtual machine 132 is moved and a virtual machine server in the data center (step). S5). The destination virtual machine server is determined using an existing technique such as live migration.

  The control unit 122 instructs the virtual machine server manager 121 to move the virtual machine 132 according to the determinations in steps S4 and S5 (step S6).

  The virtual machine server manager 121 that receives the instruction from the control unit 122 communicates with the virtual machine server agent 131 of the virtual machine server 130 including the virtual machine 132 and instructs to move the virtual machine 132. The virtual machine server agent 131 transfers the contents of the memory of the virtual machine 132 designated for movement and the data related to the virtual machine 132 stored in the storage device 160 via the network 500 to the virtual of the data center of the movement destination. It is sent to the machine server (step S7).

  When the data related to the virtual machine 132 is sent to the destination virtual machine server in step S7, the virtual machine server manager 121 communicates with the virtual machine server agent 131 to operate the virtual machine 132 on the virtual machine server 130. Is stopped (step S8).

  Then, the virtual machine moved in step S7 is activated at the destination (step S9). When the virtual machine server manager 121 confirms that the virtual machine is activated at the destination, the virtual machine server manager 121 deletes the virtual machine 132 on the virtual machine server 130 (step S10).

  For example, the contents of processing performed on the migration source virtual machine 132 from when the migration source data is sent (step S7) to when the virtual machine is stopped (step S8) are the migration destination virtual machine. Although this will not be reflected on the machine, at this time, using the existing technology, the contents of the processing performed on the migration source virtual machine 132 during the period (between steps S7 and S8) are transferred to the migration destination. It may be additionally sent to the virtual machine server and reflected in the virtual machine.

  For example, the order of the operation shown in step S7 and the operation shown in step S8 may be switched. Further, for example, the operation shown in step S10 may not be executed, that is, 132 on the virtual machine server 130 that is the migration source virtual machine may not be deleted.

  In such a virtual machine system using the data center 100, when the disaster detection device 150 detects a disaster that affects the data center 100, the virtual machine that satisfies the predetermined condition is moved to the virtual machine server of another data center. Since the virtual machine is rearranged, the moved virtual machine can avoid the influence of the disaster. And the fall of the service level of the server provided by a virtual machine can be avoided.

  In addition, in such an image machine system using the data center 100, even if a part of the functions of the data center is lost due to the occurrence of a disaster and the redundancy of the server is temporarily reduced, Since the virtual machine that provides the server is moved and rearranged in the data center, the temporarily reduced server redundancy can be recovered in a short time.

Embodiment 2. FIG.
A second embodiment (embodiment 2) of the virtual machine system according to the present invention will be described. The configuration of the data center of the second embodiment of the virtual machine system according to the present invention is the same as the configuration of the data center shown in FIGS.

  In the virtual machine system according to the second embodiment, the virtual machine management system 120 issues an instruction to the virtual machine server agent of the virtual machine server based on the disaster information notified from the disaster detection apparatus 150 to perform the existing provisioning. Using this technology, the virtual machine server is recreated in another data center. That is, the virtual machine server manager 121 of the virtual machine management system 120 does not migrate the virtual machine as in the first embodiment but moves the virtual machine to a virtual machine server in another data center to which resources have been allocated in advance. The virtual machine is rearranged by re-creating the virtual machine to be dealt with.

  In such a virtual machine system using the data center 100, by using provisioning, a virtual machine (for example, a virtual machine 132) is transferred from the source data center 100 to the destination data center (for example, the data center 200). Since the communication load can be suppressed compared to the case of moving the whole, quicker switching is possible. Therefore, continuation of the service provided by the virtual machine can be expected even for a disaster report or prediction that does not have much time to relocate the virtual machine, such as earthquake early warning.

  FIG. 10 is a block diagram showing the main part of the virtual machine system according to the present invention. As shown in FIG. 10, in the virtual machine system 1 (for example, equivalent to the data center 100 shown in FIG. 1), the virtual machines 10 and 11 (for example, equivalent to the virtual machines 132 and 133 shown in FIG. 2) are constructed. The virtual machine server 12 (for example, equivalent to the virtual machine server 130 shown in FIG. 2) is included, and other virtual machine systems 3 and 4 (for example, FIG. 1) are connected via the network 2 (for example, equivalent to the network 500 shown in FIG. 1). 1 is a virtual machine system connected to the data center 200 or 300 shown in FIG. 1 and detects a predetermined disaster from the disaster information and notifies the detected disaster information (for example, FIG. 2). Corresponding to the disaster detection device 150) and processing setting information in which the corresponding processing according to the type of disaster is registered (for example, the disaster stored in the disaster response information DB) Storage unit 14 (for example, as shown in FIG. 3) that stores hardware and software system information in virtual machine systems 3 and 4 including virtual machine servers that are candidates for virtual machine relocation destinations. The virtual machine priority information DB 123, the data center information DB 124, and the disaster response information DB 125), the disaster information notified by the disaster detection unit 13, the process setting information and the system information stored in the storage unit 14 (for example, , Corresponding to the data center information stored in the data center information DB 124) and the virtual machine server 31 to which the virtual machine 10 is relocated (for example, the data shown in FIG. 1) The correspondence determining unit 15 (for example, FIG. 3) that determines the virtual machine server in the center 200 And a server manager unit that causes the virtual machine server 12 in which the virtual machine 10 is constructed to relocate the virtual machine 10 to the relocation destination virtual machine server 31 according to the determination by the correspondence determining unit 15 16 (e.g., corresponding to the virtual machine server manager 121 shown in FIG. 3).

  In the above embodiments, virtual machine systems as shown in the following (1) to (4) are also disclosed.

(1) After the rearrangement of the virtual machine is completed, the server manager unit causes the virtual machine server in which the virtual machine has been constructed to delete the original virtual machine (for example, shown in steps S9 to S10 in FIG. 9). Realized by operation.) Virtual machine system.

(2) The server manager unit moves the virtual machine to the relocation destination virtual machine server by live migration to the virtual machine server in which the virtual machine determined to be rearranged by the correspondence determination unit (for example, The virtual machine system is realized by the operation of the virtual machine server manager 121 in the virtual machine system of the first embodiment.

(3) The server manager unit causes the virtual machine server in which the virtual machine determined to be rearranged by the correspondence determination unit is constructed to recreate the virtual machine on the rearrangement destination virtual machine server by provisioning (for example, This is realized by the operation of the virtual machine server manager 121 in the virtual machine system of the second embodiment.) Virtual machine system.

(4) A virtual machine system in which the storage unit stores process setting information including a corresponding process on the condition that a predetermined time has elapsed.

DESCRIPTION OF SYMBOLS 1,3,4 Virtual machine system 2 Network 10,11 Virtual machine 12,31 Virtual machine server 13 Disaster detection part 14 Storage part 15 Correspondence determination part 16 Server manager part 100,200,300,400 Data center 110,210,310 , 410 Gateway 120, 220, 320, 420 Virtual machine management system 121 Virtual machine server manager 122 Control unit 123 Virtual machine priority information DB
124 Data Center Information DB
125 Disaster response information DB
130,140 Virtual machine server 131,141 Virtual machine server agent 132,133,142,143 Virtual machine 150 Disaster detection device 160 Storage device 170 Local area network 180 Storage device network 500 Network

Claims (7)

A virtual machine system including a virtual machine server on which a virtual machine is constructed and connected to another virtual machine system via a network;
A disaster detection unit that detects a predetermined disaster from the disaster information and notifies the information of the detected disaster;
A storage unit for storing process setting information in which a response process according to the type of disaster is registered, and system information on hardware and software in a virtual machine system including a virtual machine server that is a candidate for a virtual machine relocation destination; ,
Based on the disaster information notified by the disaster detection unit and the processing setting information and system information stored in the storage unit, a virtual machine that performs relocation, and a virtual machine that is the relocation destination of the virtual machine A correspondence determination unit for determining a server;
A virtual machine server in which the virtual machine is constructed includes a server manager unit that relocates the virtual machine to the relocation destination virtual machine server according to the determination by the correspondence determining unit. Machine system. The virtual machine system according to claim 1, wherein the server manager unit causes the virtual machine server in which the virtual machine is constructed to delete the original virtual machine after the rearrangement of the virtual machine is completed. The server manager unit moves the virtual machine to a relocation destination virtual machine server by live migration to a virtual machine server in which the virtual machine determined to be relocated by the correspondence determining unit is built. The virtual machine system according to claim 2. The server manager unit causes a virtual machine server in which a virtual machine determined to be rearranged by the correspondence determining unit is constructed to recreate the virtual machine in a rearrangement destination virtual machine server by provisioning. The virtual machine system according to claim 2. The virtual machine system according to any one of claims 1 to 4, wherein the storage unit stores process setting information including a corresponding process with a predetermined time lapse as a condition. A virtual machine arrangement method in a virtual machine system including a virtual machine server in which a virtual machine is constructed and connected to another virtual machine system via a network,
Detect a specific disaster from disaster information,
Notifying the detected disaster information,
A virtual machine including the notified disaster information, processing setting information in which a response process corresponding to the type of disaster stored in the storage unit is registered, and a virtual machine server that is a candidate for a virtual machine relocation destination Based on system information on hardware and software in the system, a virtual machine to be relocated and a virtual machine server to which the virtual machine is relocated are determined,
A virtual machine arrangement method, comprising: causing a virtual machine server in which the virtual machine is constructed to relocate the virtual machine to the relocation destination virtual machine server according to the determination. The virtual machine placement method according to claim 6, wherein after the rearrangement of the virtual machine is completed, the virtual machine server in which the virtual machine has been constructed is deleted.

Images