WO2012063306A1 - Virtual machine control device, virtual machine control system, virtual machine control method of virtual machine control device, and virtual machine control program - Google Patents

Abstract

The objective of the present invention is to prevent transmitted data from being dropped and to enable synchronous saving or recovery of computers, even in an environment having a short inter-system process timeout. The saving/recovery instruction unit (111) of a management computer (110) transmits a VM (220) (virtual machine) save command to each execution computer (200). The saving/recovery request unit (210) of each execution computer (200) receives the save command, halts the VM (220), converts the VM (220) to data, and saves the VM (220). After an inter-system process timeout time has elapsed, a device operation control unit (211) halts a vNICb (228) (virtual communication device), and a buffer saving/recovery unit (212) saves the communication buffer. When the VM (220) is resumed, the saving/recovery request unit (210) recovers the VM (220) that has been converted to data, and the buffer saving/recovery unit (212) recovers the communication buffer. As a result, it is possible to synchronously save or recover VMs (220).

Description

Virtual computer control device, virtual computer control system, virtual computer control method for virtual computer control device, and virtual computer control program

The present invention relates to, for example, a virtual machine control device, a virtual machine control system, a virtual machine control method for a virtual machine control device, and a virtual machine control program for stopping or operating a plurality of virtual machines in synchronization.

There is a virtualization technology that can convert a computer into data and save it as a file.
By using this technique, the user can repeatedly restore a computer once saved in a file, and reproduce or construct a system development environment or test environment.

However, in an environment where inter-system processing is performed between computers to monitor each other's computers, it is necessary to synchronize and restore the computers.
This is because if the computers are individually saved or restored, inconsistency occurs in the inter-system processing, and an abnormality occurs in the application program operating in the computer.

JP 2009-80692 A

In the prior art, loss of communication data is prevented in order to prevent inconsistencies in inter-system processing by synchronizing or saving or restoring computers. Specifically, the computer is saved when transmission / reception of communication data between computers is completed. Then, after all the computers were restored and all the computers were operating, transmission / reception was resumed.
However, in the prior art, since the communication suspension period between computers is long, inconsistency may occur in an environment where the timeout of inter-system processing is short.

An object of the present invention is to prevent communication data from being lost and to save or restore a computer in synchronization, for example, even in an environment where the timeout of intersystem processing is short.

The virtual machine control device of the present invention is
A virtual communication device that performs communication processing and a virtual computer that communicates data via the virtual communication device are incorporated,
A command receiving unit that receives a stop command for commanding stop of the virtual machine from a virtual machine management device that manages a plurality of virtual machines including the virtual machine;
A computer control unit for stopping the virtual machine when the command receiving unit receives the virtual machine stop command;
When the command receiving unit receives a stop command for the virtual machine, a predetermined number for receiving data transmitted by another virtual machine before the other virtual machine is stopped by a stop command for another virtual machine. A communication device control unit that stores data received in the reception waiting time by the virtual communication device in a predetermined storage area after the reception waiting time elapses;

According to the present invention, for example, even in an environment where the timeout of inter-system processing is short, communication data can be prevented from being lost and the computers can be stored synchronously.

1 is a configuration diagram of a computer control system 100 according to Embodiment 1. FIG. 3 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the first embodiment. 3 is a flowchart showing a computer control method (at the time of restoration) of the computer control system 100 according to the first embodiment. 2 is a diagram illustrating an example of hardware resources of a management computer 110 and an execution computer 200 in Embodiment 1. FIG. FIG. 2 is a schematic diagram illustrating an example of a method for using the computer control system 100 according to the first embodiment. FIG. 3 is a schematic diagram illustrating an example of a problem to be solved by the computer control system 100 according to the first embodiment. FIG. 3 is a configuration diagram of a computer control system 100 according to a second embodiment. 9 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the second embodiment. 10 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the third embodiment. 10 is a flowchart showing a computer control method (at the time of restoration) of the computer control system 100 according to the third embodiment. 10 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the fourth embodiment. 10 is a flowchart showing a computer control method (at the time of restoration) of the computer control system 100 according to the fourth embodiment.

Embodiment 1 FIG.
A description will be given of a mode in which when a computer is stopped, a communication device is stopped after a predetermined time in order to receive data transmitted from another computer.

FIG. 1 is a configuration diagram of a computer control system 100 according to the first embodiment.
The configuration of the computer control system 100 in the first embodiment will be described with reference to FIG.

The computer control system 100 (an example of a virtual computer control system) includes a plurality of physical computers. The physical computer is a computer (computer) configured as hardware.
Hereinafter, one physical computer is referred to as “management computer 110”, and the remaining physical computers are referred to as “execution computer 200”.
However, one physical computer may be used as the management computer 110 and the execution computer 200. Furthermore, the computer control system 100 may be configured with one physical computer.

The management computer 110 and the execution computer 200 communicate via the network 101.

The execution computers 200A and 200B (an example of a virtual computer control device) are physical computers in which one or a plurality of virtual computers are incorporated.
The management computer 110 (an example of a virtual computer management apparatus) is a physical computer that manages a plurality of virtual computers.
The virtual computer is an execution unit that executes software (for example, an operating system) that imitates a physical computer.
Hereinafter, the virtual machine is referred to as “VM220 (Virtual Machine)”.

The management computer 110 includes a save / restore instruction unit 111, a physical N / W device 118 (N / W: Network), and a physical NIC 119 (NIC: Network Interface Card).

The save / restore instruction unit 111 transmits a save instruction and a restore instruction of the VM 220 to the execution computers 200A and 200B.

The physical N / W device 118 and the physical NIC 119 are hardware that performs communication via the network 101.

The execution computers 200A and 200B include a VM 220, a vNICb 228 (vNIC: virtual NIC), a VMM 229 (Virtual Machine Monitor), a management domain unit 201, and a physical NIC 206.

The vNICb 228 (an example of a virtual communication device) executes communication processing by executing software that imitates a network interface card incorporated in the VM 220.

The VM 220 includes a program unit 221 that communicates data via the vNICb 228.
The program unit 221 executes a program representing a procedure of predetermined business processing (for example, online processing).

For example, the VM 220 (the program unit 221 thereof) and the vNICb 228 operate as follows.
(1) The vNICb 228 stores data received from another VM 220 in a reception buffer (a vNICb area described later).
The VM 220 processes the data stored in the reception buffer of the vNICb 228.
(2) When data indicating that another VM 220 is operating is transmitted by another VM 220, the vNICb 228 receives the data as life / death monitoring data.
When the alive monitoring data is not received within a predetermined timeout time by the vNICb 228, the VM 220 determines that another VM 220 is not operating.

The VMM 229 is a VM management unit that manages the VM 220.

The management domain unit 201 is a VMM management unit that manages the VMM 229.

The management domain unit 201 includes a program unit 202, a vNICa 203, a virtual bridge 204, and a physical N / W device 205.
The program unit 202 executes a program that causes the execution computer 200 to function.
The vNICa 203 executes communication processing by executing software that imitates a network interface card incorporated in the management domain unit 201.
The virtual bridge 204 executes communication processing by executing software that imitates a bridge (relay device) incorporated in the management domain unit 201.

The physical N / W device 205 and the physical NIC 206 are hardware that performs communication processing.

The management domain unit 201 further includes a save / restore request unit 210, a device operation control unit 211, and a buffer save / restore unit 212.

The save / restore request unit 210 (an example of a command receiving unit and a computer control unit) receives a VM 220 stop command (a save command to be described later) from the management computer 110.
The save / restore request unit 210 stops the VM 220 when receiving a stop command.
The save / restore request unit 210 receives an activation command (restoration command to be described later) of the VM 220.
When receiving the activation command, the save / restore request unit 210 activates the VM 220 after the received data saved when the vNICb 228 stops is stored in the reception buffer of the vNICb 228.

When the save / restoration request unit 210 receives a stop command, the device operation control unit 211 (an example of a communication device control unit) stops the vNICb 228 after a predetermined reception wait time has elapsed.
The predetermined reception waiting time is a time for receiving data transmitted by another VM 220 before the other VM 220 is stopped.
The timeout time for determining whether or not another VM 220 is operating is an example of a predetermined reception waiting time.

Further, the device operation control unit 211 activates the vNICb 228 when the save / restore request unit 210 receives the activation command.

When the device operation control unit 211 receives a stop command, the buffer storage / restoration unit 212 (an example of a communication device control unit) stores data received by the vNICb 228 during a reception waiting time in a predetermined storage area (a VM storage medium described later) ).
When the save / restore request unit 210 receives an activation command, the buffer save / restore unit 212 stores the saved data of the vNICb 228 in the receive buffer of the vNICb 228.

Next, a computer control method of the computer control system 100 will be described.
The VM 220, vNICb 228, and “˜ unit” of the computer control system 100 operate using a CPU (Central Processing Unit).
The save / restore instruction unit 111 of the management computer 110 communicates via the physical N / W device 118 and the physical NIC 119.
Further, the save / restore request unit 210 of the execution computer 200 communicates via the vNICa 203, the virtual bridge 204, the physical N / W device 205, and the physical NIC 206.
These descriptions are omitted in the following description.

Further, it is assumed that the main computer (normal system) VM 220 is operating in the execution computer 200A, and the secondary (standby system) VM 220 is operating in the execution computer 200B.

The main VM 220 executes predetermined business processing (for example, online processing) at normal times, and the subordinate VM 220 is stopped due to a failure or the like (when the main VM 220 is down). In addition, the business process is executed instead of the main VM 220.
The main VM 220 and the subordinate VM 220 periodically transmit and receive data (signals) to check whether the partner VM 220 is in a down state. Hereinafter, the data for confirming the state of the partner VM 220 is referred to as “life monitoring data”.
The VM 220 cannot receive the life / death monitoring data of the other party even after a predetermined time-out period has elapsed since the transmission of the life / death monitoring data, or the predetermined time-out time has elapsed since the previous reception of the life / death monitoring data. If new life and death monitoring data cannot be received, it is determined that the partner VM 220 is down.

FIG. 2 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the first embodiment.
The flow of the computer control method when saving (stopping) the VM 220 will be described with reference to FIG.

It is assumed that data (control information, flag, register value, etc.) representing the VM 220 is set in a predetermined storage area of the execution computer 200 and the VM 220 is operating.
Hereinafter, data representing the VM 220 is referred to as “VM data (or VM file)”, and a storage area in which the VM data is set is referred to as “VM area”.

In S001, the save / restore instruction unit 111 of the management computer 110 generates a save command at a specific timing, and transmits the generated save command to each execution computer 200.
The specific timing is when the administrator inputs a save (stop) command of the VM 220 to the management computer 110, when a series of processing ends, when a predetermined time comes, or the like.
The save command is a command for requesting that the VM 220 is stopped and the VM data representing the state of the VM 220 at the time of the stop is stored in a predetermined recording medium (magnetic tape, magnetic disk, semiconductor memory, etc.). Hereinafter, a recording medium that stores VM data is referred to as a “VM storage medium”.
In the save command, information such as a VM name for identifying the VM 220 to be saved and a file name set in the saved VM data are designated.

For example, the save / restore instruction unit 111 transmits a save command designating the main VM 220 to the execution computer 200A, and sends a save command designating the subordinate VM 220 of the execution computer 200B.
Further, the save / restore instruction unit 111 may transmit a save command designating the main VM 220 and the subordinate VM 220 to the execution computer 200A and the execution computer 200B.

After S001, the process proceeds to S002.

The processing of S002 to S009 described below is executed for each execution computer 200 to which the save command is transmitted.

In S002, the save / restore request unit 210 receives the save command transmitted from the save / restore instruction unit 111 of the management computer 110.
It progresses to S003 after S002.

In the description after S003, “VM220” means the VM 220 identified by the VM name specified in the save command, that is, the VM 220 to be saved.

In S003, the save / restore request unit 210 calls the device operation control unit 211, and the device operation control unit 211 stops input / output (I / O: Input / Output) between the VM 220 and the vNICb 228.
Thereafter, the VM 220 cannot process the received data received by the vNICb 228 and stored in the communication (reception) buffer. Further, the VM 220 cannot store the transmission data in the communication (transmission) buffer in order to transmit the data.
It progresses to S004 after S003.

In S004, the save / restore request unit 210 stops the VM 220 and saves the VM data of the stopped VM 220 in the VM storage medium. The VM data is saved with the file name specified by the save command. vNICb228 has not stopped yet.
S004 may be executed in the reverse order of S003 or simultaneously with S003.
It progresses to S005 after S004.

In S005, the save / restore request unit 210 waits without stopping the vNICb 228 until a predetermined reception waiting time elapses.
The predetermined reception waiting time is necessary for the data transmitted by the other VM 220 to reach the vNICb 228 of the destination VM 220 before the other VM 220 is stopped by the save command, and to receive and process the data that the vNICb 228 has reached. It's time.
For example, the saving / restoring request unit 210 waits until a timeout period (an example of a reception waiting time) for alive monitoring elapses from when a save command is received or when alive monitoring data is transmitted / received last time. Thereby, the vNICb 228 of the VM 220 can receive the alive monitoring data transmitted from the partner VM 220.
It progresses to S006 after S005.

In S006, the save / restore request unit 210 calls the device operation control unit 211, and the device operation control unit 211 stops the vNICb 228.
It progresses to S007 after S006.

In S007, the save / restore request unit 210 calls the buffer save / restore unit 212, and the buffer save / restore unit 212 saves the data (communication buffer data) stored in the communication buffer of the vNICb 228 in the VM storage medium. The communication buffer data is assumed to be stored in association with the VM data.
For example, the communication buffer data includes alive monitoring data received from the partner VM 220 after the VM 220 is stopped.
Hereinafter, the storage area of the execution computer 200 secured as a communication buffer for the vNICb 228 is referred to as a “vNICb area”.
It progresses to S008 after S007.

In S008, the save / restore request unit 210 discards the VM 220 by releasing the VM area and the vNICb area.
It progresses to S009 after S008.

In step S <b> 009, the save / restore request unit 210 generates a save completion notification that notifies that the saving of the VM 220 is completed, and transmits the generated save completion notification to the save / restore instruction unit 111 of the management computer 110. In the save completion notification, the VM name of the VM 220 that has been saved is set.
It progresses to S010 after S009.

In step S010, the save / restore instruction unit 111 of the management computer 110 receives a save completion notification from the save / restore request unit 210 of each execution computer 200.
When the save / restore instruction unit 111 receives the save completion notification of all the VMs 220 for the plurality of VMs 220 to be saved, the computer control method (when saving) ends.

FIG. 3 is a flowchart showing a computer control method (at the time of restoration) of the computer control system 100 according to the first embodiment.
The flow of the computer control method when restoring (starting up) the VM 220 will be described with reference to FIG.

In S101, the save / restore instruction unit 111 of the management computer 110 generates a restore command at a specific timing, and transmits the generated restore command to each execution computer 200.
The specific timing is when the administrator inputs a restore (start) command of the VM 220 to the management computer 110 or when a predetermined time comes.
The restoration command is a command for restoring the VM data saved in the VM storage medium to the VM area and starting the VM 220.
In the restoration command, information such as a plurality of VM names for identifying a plurality of VMs 220 to be restored and a file name of VM data used for restoration of the VM 220 are designated.

For example, the save / restore instruction unit 111 transmits a restore command designating the main VM 220 to the execution computer 200A, and sends a restore command designating the subordinate VM 220 of the execution computer 200B.
Further, the save / restore instruction unit 111 may transmit a restore command designating the primary VM 220 and the secondary VM 220 to the execution computer 200A and the execution computer 200B.

After S101, the process proceeds to S102.

The processing of S102 to S106 described below is executed for each execution computer 200 to which a restoration command is transmitted.

In S102, the save / restore request unit 210 receives the restore command transmitted from the save / restore instruction unit 111 of the management computer 110.
After S102, the process proceeds to S103.

In the description after S103, “VM220” means the VM 220 identified by the VM name specified in the restoration command (or a resume command described later), that is, the VM 220 to be restored.

In S103, the save / restore request unit 210 loads (restores) the VM data saved with the file name specified in the restore command.
The loading of VM data means reading the VM data from the VM storage medium and writing the read VM data in the VM area of the execution computer 200.
Thereby, the state of the VM 220 when the VM 220 is stopped can be restored.
It progresses to S104 after S103.

In S104, the save / restore request unit 210 calls the buffer save / restore unit 212, and the buffer save / restore unit 212 loads communication buffer data corresponding to the VM data.
The load of the communication buffer data is to read the communication buffer data from the VM storage medium and write the read communication buffer data to the vNICb area of the execution computer 200.
Thereby, the state of the communication buffer when the vNICb 228 is stopped can be restored.
After S104, the process proceeds to S105.

In S105, the save / restore request unit 210 calls the device operation control unit 211, and the device operation control unit 211 activates the vNICb 228.
After S105, the process proceeds to S106.

In S <b> 106, the save / restore request unit 210 generates a restore completion notification notifying that the restoration of the VM 220 is completed, and transmits the generated restore completion notification to the save / restore instruction unit 111 of the management computer 110. In the restoration completion notification, the VM name of the VM 220 that has been restored is set.
It progresses to S107 after S106.

In S <b> 107, the save / restore instruction unit 111 of the management computer 110 receives a restore completion notification from the save / restore request unit 210 of each execution computer 200.
When the storage / restoration instructing unit 111 receives the restoration completion notification of all the VMs 220 for the plurality of VMs 220 to be restored, the process proceeds to S108.

In S <b> 108, the save / restore instruction unit 111 of the management computer 110 generates a restart command and transmits the generated restart command to each execution computer 200.
The resume command is a command for requesting to resume the operation of the restored VM 220.
The VM name designated by the restoration command is designated as the resume command, and the resume command is sent to each execution computer 200 to which the restoration command has been sent.
After S108, the process proceeds to S109.

The processing of S109 to S111 described below is executed for each execution computer 200 to which a restart command is transmitted.

In S109, the save / restore request unit 210 receives the restart command transmitted from the save / restore instruction unit 111 of the management computer 110.
It progresses to S110 after S109.

In S110, the save / restore request unit 210 activates the VM 220 restored in S103.
Thereby, the operation of the VM 220 can be resumed from the state when the VM 220 is stopped.

The resumed VM 220 acquires received data from the communication buffer of the vNICb 228 restored in S104, and processes the acquired received data.
For example, when the alive monitoring data is received by the vNICb 228 after the VM 220 is stopped, the resumed VM 220 acquires the alive monitoring data from the communication buffer of the vNICb 228 and determines that the partner VM 220 is operating. Also, the VM 220 generates life / death monitoring data that responds to the partner VM 220 and stores the generated life / death monitoring data in the communication buffer of the vNICb 228. Then, the vNICb 228 transmits the alive monitoring data registered in the communication buffer to the partner VM 220.

After S110, the process proceeds to S111.

In step S <b> 111, the save / restore request unit 210 generates a restart completion notification that notifies that the resumption of the VM 220 is completed, and transmits the generated restart completion notification to the save / restore instruction unit 111 of the management computer 110. In the restart completion notification, the VM name of the VM 220 that has been restarted is set.
It progresses to S112 after S111.

In S <b> 112, the save / restore instruction unit 111 of the management computer 110 receives a resume completion notification from the save / restore request unit 210 of each execution computer 200.
When the storage / restoration instructing unit 111 receives a resume completion notification of all VMs 220 for a plurality of VMs 220 to be resumed, the computer control method (at the time of restoration) ends.

FIG. 4 is a diagram illustrating an example of hardware resources of the management computer 110 and the execution computer 200 according to the first embodiment.
In FIG. 4, the management computer 110 and the execution computer 200 include a CPU 911. The CPU 911 is connected to the ROM 913, the RAM 914, the communication board 915, the display device 901, the keyboard 902, the mouse 903, the drive device 904, and the magnetic disk device 920 via the bus 912, and controls these hardware devices. The drive device 904 is a device that reads and writes a storage medium such as an FD (Flexible Disk Drive), a CD (Compact Disc), a DVD (Digital Versatile Disc), and an SSD (Solid State Drive).

The communication board 915 is wired or wirelessly connected to a communication network such as a LAN (Local Area Network), the Internet, or a telephone line.

The magnetic disk device 920 stores an OS 921 (operating system), a program group 922, and a file group 923.

The program group 922 includes programs for executing functions described as “˜unit” (or VM 220, vNICb 228) (hereinafter the same) in the embodiment. The program is read and executed by the CPU 911. That is, the program causes the computer to function as “˜part”, and causes the computer to execute the procedures and methods of “˜part”.

The file group 923 includes various data (input, output, determination result, calculation result, processing result, etc.) used in “˜part” described in the embodiment.

In the embodiment, arrows included in the configuration diagrams and flowcharts mainly indicate input and output of data and signals.

In the embodiment, what is described as “to part” may be “to circuit”, “to apparatus”, and “to device”, and “to step”, “to procedure”, and “to processing”. May be. That is, what is described as “˜unit” may be implemented by any of firmware, software, hardware, or a combination thereof.

In the first embodiment, the VM 220 may be a physical computer instead of a virtual computer, and the vNICb 228 may be a physical communication device instead of a virtual communication device.

In Embodiment 1, for example, the following computer control system 100 has been described.

The computer control system 100 includes a save / restore instruction unit 111, a save / restore request unit 210, a device operation control unit 211, and a buffer save / restore unit 212, and saves or restores information on a computer in operation in synchronization.
The save / restore instruction unit 111 instructs a plurality of managed save / restore request units 210 to start synchronous save / restore processing.
The save / restore request unit 210 saves or restores the target computer in accordance with an instruction from the save / restore instruction unit 111.
The device operation control unit 211 operates or stops the network device (vNICb 228) separately from the operation or stop of the computer OS (VM 220) in accordance with the instruction of the save / restore request unit 210.
The buffer saving / restoring unit 212 saves or restores communication data accumulated in the network device in accordance with an instruction from the saving / restoring request unit 210.

The user can save a plurality of computers simultaneously by issuing a save request to the save / restore instruction unit 111.
The save / restore instruction unit 111 transmits the request to the save / restore request unit 210.
Upon receipt of the request, the save / restore request unit 210 temporarily stops the computer OS, saves information, operates only the network device for a predetermined time, and completes transmission / reception of communication data at the device level. Save the buffer.

The user can restore a plurality of computers simultaneously by issuing a restoration request to the save / restore instruction unit 111.
The save / restore instruction unit 111 transmits the request to the save / restore request unit 210.
Upon receiving the request, the save / restore request unit 210 restores the saved computer information, restores the saved network device buffer, and restarts the computer OS.

According to the first embodiment, for example, the following effects can be obtained.

FIG. 5 is a schematic diagram illustrating an example of a method of using the computer control system 100 according to the first embodiment.
The computer control system 100 according to the first embodiment can be used for a computer test as shown in FIG.

(1) The computer 102 constituting the computer system to be tested is converted into data, and the computer 102 (VM 220) converted into data is stored as a computer file 103 (VM data).
(2) The computer 102 is restored using the computer file 103, and (3) a test (for example, a failure test) of the computer 102 is executed.
(4) When it is necessary to newly test the computer 102, the computer 102 is restored again using the computer file 103.

FIG. 6 is a schematic diagram illustrating an example of a problem to be solved by the computer control system 100 according to the first embodiment.
For example, the computer control system 100 can solve the problem of the inter-system processing system as shown in FIG.

The inter-system processing system includes a primary computer (first VM 220) and a secondary computer (second VM 220).
(1) The primary computer and the secondary computer periodically send and receive life / death monitoring data (solid arrows). When the primary computer or the secondary computer cannot receive the life / death monitoring data from the counterpart computer within a predetermined time-out period, it is determined that the counterpart computer has failed. For example, a slave computer that determines that the primary computer has failed operates as a new primary computer.
(2) When stopping the test of the inter-system processing system, after stopping communication, the computer is stopped, and the computer is converted into data and stored.
(3) When the test of the inter-system processing system is resumed, the computer is restored and resumed, and then the communication is resumed (the period during which the communication device is stopped and the computer is operating is indicated by a bold line).
(4) In this case, the life and death monitoring data (dotted arrow) cannot be communicated between the time when communication is stopped and the time when it is restarted. Therefore, the life and death monitoring time-out occurs before the computer stops or when the computer is restarted. To do.
(5) However, since the main computer is operating, there is a contradiction that both computers operate as the main computer.

On the other hand, the computer control system 100 stops the communication after the reception waiting time for stopping the computer and receiving the alive monitoring data has elapsed. For this reason, the above contradiction does not occur.

Embodiment 2. FIG.
A mode in which the vNICb 228 filters unnecessary data during the reception waiting time after the VM 220 is stopped will be described.
Hereinafter, items different from the first embodiment will be mainly described. Matters whose description is omitted are the same as those in the first embodiment.

FIG. 7 is a configuration diagram of the computer control system 100 according to the second embodiment.
The configuration of the computer control system 100 according to the second embodiment will be described with reference to FIG.

The execution computers 200A and 200B include a filtering unit 213 in addition to the configuration of the first embodiment (see FIG. 1).

The filtering unit 213 causes the vNICb 228 to receive only specific data when the save / restore request unit 210 receives a VM 220 stop command (save command).
For example, the filtering unit 213 causes the vNICb 228 to receive only data transmitted from a specific VM 220.

FIG. 8 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the second embodiment.
The flow of the computer control method when saving (stopping) the VM 220 will be described with reference to FIG.

The computer control system 100 executes S201 and S202 in addition to the processing of the first embodiment (see FIG. 2).
Further, S001 is partially different from the first embodiment in processing.
Hereinafter, S001, S201, and S202 will be mainly described, and other processing will be omitted.

In S001, the save / restore instruction unit 111 of the management computer 110 generates a save command and transmits the generated save command to each execution computer 200.
However, a filtering rule is specified in the save command. The filtering rule is information that restricts data to be received by the vNICb 228 in the reception waiting time after the VM 220 is stopped.

For example, the save / restore instruction unit 111 specifies (1) a data transmission source and (2) a data type as a filtering rule as follows. However, both (1) data transmission source and (2) data type may be specified as filtering rules.
(1) The save / restore instruction unit 111 designates data transmitted from the secondary VM 220 (or vNICb 228) as data to be received by the primary vNICb 228, and as data to be received by the secondary vNICb 228. Specifies data transmitted from (or vNICb 228).
(2) The save / restore instruction unit 111 specifies alive monitoring data as data to be received by the vNICb 228.
Further, the save / restore instruction unit 111 may specify data (for example, broadcast data) to be discarded by the vNICb 228 as a filtering rule.

By specifying the filtering rule, it is possible to solve the problem that the communication buffer capacity of the vNICb 228 is insufficient due to unnecessary data, and important data (for example, alive monitoring data) cannot be stored in the communication buffer.

After S001, the process proceeds to S002.

The save / restore request unit 210 receives the save command (S002), and the device operation control unit 211 stops the input / output between the VM 220 and the vNICb 228 (S003).
It progresses to S201 after S003.

In S201, the save / restore request unit 210 calls the filtering unit 213, and the filtering unit 213 sets the filtering rule specified in the save command in the vNICb 228.
Thereafter, the vNICb 228 receives only the data designated as “data to be received” by the filtering rule, and discards the other data. Data received by the vNICb 228 is stored in the communication buffer, and data discarded by the vNICb 228 is not stored in the communication buffer.
It progresses to S004 after S201.

The save / restore request unit 210 stops the VM 220 and saves the VM data (S004), and waits until a predetermined reception waiting time elapses (S005).
After the reception waiting time elapses, the device operation control unit 211 stops the vNICb 228 (S006), and the buffer storage / restoration unit 212 stores the communication buffer data (S007).
It progresses to S202 after S007.

In S202, the save / restore request unit 210 calls the filtering unit 213, and the filtering unit 213 deletes the filtering rule set in the vNICb 228.
It progresses to S008 after S202.

The save / restore request unit 210 discards the VM 220 (S008) and transmits a save completion notification (S009).
The save / restore instruction unit 111 of the management computer 110 receives a save completion notification from the save / restore request unit 210 of each execution computer 200 (S010), and the computer control method (during saving) ends.

The computer control method (at the time of restoration) for restoring the VM 220 is the same as that in the first embodiment (see FIG. 3).

In the second embodiment, for example, the following computer control system 100 has been described.

The computer control system 100 includes a filtering unit 213.
The filtering unit 213 performs filtering of communication data sent to the virtual network device (vNICb 228) in accordance with an instruction from the save / restore request unit 210.
File tally prevents unwanted communication data from accumulating in a buffer in the device. The filtering rules are sent together with the list of computers to be saved by the save / restore instruction unit 111 when the save / restore instruction unit 111 issues a command to the save / restore request unit 210.

Embodiment 3 FIG.
A mode in which a plurality of VMs 220 are stopped or restarted at the same time will be described.
Hereinafter, items different from the first and second embodiments will be mainly described. Matters whose description is omitted are the same as in the first and second embodiments.

The configuration of the computer control system 100 is the same as that of the first embodiment (see FIG. 1).

However, the save / restore request unit 210 receives a stop command (save command) in which a stop time (a start time of a save process described later) is designated.
The save / restore request unit 210 stops the VM 220 at the stop time specified in the stop command.
The buffer storage / restoration unit 212 stores the data received in the reception waiting time by the vNICb 228 in the storage area (VM storage medium) after the reception waiting time has elapsed from the stop time specified in the stop command.

FIG. 9 is a flowchart showing a computer control method (at the time of storage) of the computer control system 100 according to the third embodiment.
The flow of the computer control method when saving (stopping) the VM 220 will be described with reference to FIG.

The computer control system 100 executes S301 in addition to the processing of the first embodiment (see FIG. 2).
Further, S001 is partially different from the first embodiment in processing.
Hereinafter, S001 and S301 will be mainly described, and other processing will be omitted.

In S001, the save / restore instruction unit 111 of the management computer 110 generates a save command and transmits the generated save command to each execution computer 200.
However, in the save command, a start time for starting the save process of the VM 220 is specified.
It progresses to S002 after S001.

In S002, the save / restore request unit 210 receives a save command.
It progresses to S301 after S002.

In S301, the save / restore request unit 210 waits until the start time specified in the save command is reached.
The processes after S003 (see FIG. 2) are started from the start time specified in the save instruction.

FIG. 10 is a flowchart showing a computer control method (at the time of restoration) of the computer control system 100 according to the third embodiment.
The flow of the computer control method when restoring (starting up) the VM 220 will be described with reference to FIG.

The computer control system 100 executes S311 in addition to the processing of the first embodiment (see FIG. 3).
Further, S108 is partially different from the first embodiment in processing.
Hereinafter, S311 and S108 will be mainly described, and other processing will be omitted.

The processing from S101 to S107 is the same as in the first embodiment.

In S <b> 108, the save / restore instruction unit 111 of the management computer 110 generates a restart command and transmits the generated restart command to each execution computer 200.
However, in the restart instruction, a start time for starting the restart processing of the VM 220 is specified.
After S108, the process proceeds to S109.

In S109, the save / restore request unit 210 receives a resume command.
It progresses to S311 after S109.

In S311, the save / restore request unit 210 waits until the start time specified in the resume command.
The processes after S110 (see FIG. 3) are started from the start time specified in the resume instruction.

In the third embodiment, for example, the following computer control system 100 has been described.

The computer control system 100 has a process start time synchronization function.
The save / restore instruction unit 111 instructs the save / restore request unit 210 including the time to start the save / restore process.
The save / restore request unit 210 stops the VM 220 at the time designated by the save / restore instruction unit 111 and performs subsequent processing.
After the restoration of the VM 220, the save / restore request unit 210 resumes the operation of the VM 220 at the time designated by the save / restore instruction unit 111.

According to the third embodiment, a plurality of VMs 220 can be stored or restored in synchronization.

The computer control system 100 according to the third embodiment may filter the received data of the vNICb 228 as in the second embodiment.

Embodiment 4 FIG.
A mode in which the internal time of each execution computer 200 is adjusted to save or restore a plurality of VMs 220 in synchronization will be described.
Hereinafter, items different from the first to third embodiments will be mainly described. Matters whose description is omitted are the same as in the first to third embodiments.

The configuration of the computer control system 100 is the same as that of the first embodiment (see FIG. 1).
However, it is assumed that the management computer 110 and the execution computers 200A and 200B have a clock indicating the internal time.

Further, the save / restore request unit 210 (an example of a time adjustment unit) receives time information (time notification described later) indicating the internal time of the management computer 110 from the save / restore instruction unit 111, and based on the received time information. Thus, the time difference between the internal time of the management computer 110 and the internal time of the execution computer 200 is calculated.
The save / restore request unit 210 receives a stop command in which a stop time is specified.
The save / restore request unit 210 determines a stop timing at which the internal time and the stop time of the management computer 110 coincide with each other based on the time difference between the stop time specified in the stop command and the internal time (a save process start timing described later). ) The VM 220 is stopped.
The buffer storage / restoration unit 212 stores the data received in the reception waiting time by the vNICb 228 in the storage area (VM storage medium) after the reception waiting time has elapsed from the stop timing.

FIG. 11 is a flowchart illustrating a computer control method (at the time of storage) of the computer control system 100 according to the fourth embodiment.
The flow of the computer control method when saving (stopping) the VM 220 will be described with reference to FIG.

The computer control system 100 executes S401 to S403 in addition to the processing of the third embodiment (see FIG. 9).
Further, S301 is partially different from the third embodiment in processing.
Hereinafter, S401 to S403 and S301 will be mainly described, and other processing will be omitted.

In step S <b> 401, the save / restore instruction unit 111 of the management computer 110 transmits a time notification in which the current time of the management computer 110 is set to each execution computer 200.
It progresses to S402 after S401.

In S402, the save / restore request unit 210 of each execution computer 200 receives the time notification.
It progresses to S403 after S402.

In S403, the save / restore request unit 210 of each execution computer 200 determines the internal time of the management computer 110 and the internal time of the execution computer 200 based on the current time of the management computer 110 set in the received time notification. Calculate the time difference.
For example, the save / restore request unit 210 calculates the time difference between the current time of the management computer 110 set in the time notification and the internal time of the execution computer 200 when the time notification is received. However, the time difference may be calculated in consideration of the communication time and processing time of time notification.
After S403, the process proceeds to S001.

The save / restore instruction unit 111 of the management computer 110 transmits a save command designating the start time of the save process (S001), and the save / restore request unit 210 of each execution computer 200 receives the save command (S002).
It progresses to S301 after S002.

In S301, the save / restore request unit 210 of each execution computer 200 stores the internal time of the management computer 110 and the save based on the start time of the save process specified in the save command and the time difference between the internal times calculated in S403. The start timing at which the process start time coincides is calculated.
Then, the save / restore request unit 210 waits until the start timing, and the processing after S003 is started from the start timing.

For example, when the internal time of the execution computer 200 is advanced by “1 minute” from the internal time of the management computer 110, the save / restore request unit 210 causes the internal time of the execution computer 200 to be “1 minute” after the start time of the save process. Wait until the time.
When the internal time of the execution computer 200 is “1 minute” behind the internal time of the management computer 110, the save / restore request unit 210 sets the internal time of the execution computer 200 “1 minute” before the start time of the save process. Wait until the time.

Thereby, even if the internal time of the management computer 110 and the internal time of the execution computer 200 are shifted, each VM 220 can be stored at the time specified by the storage / restore instruction unit 111.

FIG. 12 is a flowchart illustrating a computer control method (at the time of restoration) of the computer control system 100 according to the fourth embodiment.
The flow of the computer control method when restoring (starting up) the VM 220 will be described with reference to FIG.

The computer control system 100 executes S411 to S413 in addition to the processing of the third embodiment (see FIG. 10).
In S411 to S413, the save / restore request unit 210 of each execution computer 200 calculates the time difference between the internal times of the management computer 110 and the execution computer 200, similarly to S401 to S403 (see FIG. 11).

In S311, the save / restore request unit 210 of each execution computer 200 performs the restart process until the start timing at which the internal time of the management computer 110 matches the start time of the restart process, as in S301 (see FIG. 11). stand by.

Thereby, even if the internal time of the management computer 110 and the internal time of the execution computer 200 are shifted, each VM 220 can be restarted at the time designated by the save / restore instruction unit 111.

In the fourth embodiment, the save / restore request unit 210 notifies the save / restore instruction unit 111 of the internal time of the execution computer 200, and the save / restore instruction unit 111 sets the time of the internal time between the management computer 110 and the execution computer 200. The difference may be calculated.
In this case, the save / restore instruction unit 111 designates a different start time for each execution computer 200 based on the time difference from each execution computer 200.
For example, the execution computer 200 whose internal time is advanced by “1 minute” from the management computer 110 is assigned a time “1 minute” after the predetermined start time, and the internal time is delayed by “1 minute” from the management computer 110. The execution computer 200 is assigned a time “1 minute” before the predetermined start time.

Further, the save / restore request unit 210 of each execution computer 200 changes the internal time of the execution computer 200 in accordance with the internal time of the management computer 110 notified in S402 or S412. You may correct the time lag.

In the fourth embodiment, for example, the following computer control system 100 has been described.

The computer control system 100 has an internal time synchronization function.
The save / restore instruction unit 111 notifies the save / restore request unit 210 of the internal time of the save / restore instruction unit 111 prior to the start of the save / restore process.
The save / restore request unit 210 adjusts the internal time of the save / restore request unit 210 to match the internal time of the save / restore instruction unit 111 based on the internal time notified from the save / restore instruction unit 111.
Alternatively, the save / restore request unit 210 starts the process at the timing when the internal time of the save / restore instruction unit 111 becomes the time to start the save / restore process.

According to the fourth embodiment, even if the internal time of the management computer 110 is different from the internal time of each execution computer 200, a plurality of VMs 220 can be stored or restored in synchronization.

The computer control system 100 according to the fourth embodiment may filter the received data of the vNICb 228 as in the second embodiment.

100 computer control system, 101 network, 102 computer, 103 computer file, 110 management computer, 111 save / restore instruction unit, 118 physical N / W device, 119 physical NIC, 200 execution computer, 201 management domain unit, 202 program unit, 203 vNICa, 204 virtual bridge, 205 physical N / W device, 206 physical NIC, 210 save / restore request unit, 211 device operation control unit, 212 buffer save / restore unit, 213 filtering unit, 220 VM, 221 program unit, 228 vNICb, 229 VMM, 901 display device, 902 keyboard, 903 mouse, 904 drive device, 911 CPU, 912 bus, 913 ROM, 914 RAM, 91 Communication board, 920 a magnetic disk device, 921 OS, 922 programs, 923 files.

Claims (10)

In a virtual machine control device incorporating a virtual communication device that performs communication processing and a virtual computer that communicates data via the virtual communication device,
A command receiving unit that receives a virtual machine stop command from a virtual machine management device that manages a plurality of virtual machines including the virtual machine;
A computer control unit for stopping the virtual machine when the command receiving unit receives the virtual machine stop command;
When the command receiving unit receives a stop command for the virtual machine, a predetermined number for receiving data transmitted by another virtual machine before the other virtual machine is stopped by a stop command for another virtual machine. A virtual machine control device comprising: a communication device control unit that stores data received during the reception wait time by the virtual communication device in a predetermined storage area after a reception wait time has elapsed. The command receiving unit receives a startup command for the virtual machine from the virtual machine management device,
The communication device control unit stores the data in the storage area in a predetermined reception buffer when the command receiving unit receives a startup command for the virtual machine,
The computer control unit starts the virtual computer after the storage area data is stored in the reception buffer when the command receiving unit receives the virtual computer startup command.
The virtual machine control device according to claim 1, wherein the virtual machine processes data stored in the reception buffer. When the virtual communication device transmits data indicating that another virtual machine is operating by another virtual machine, the data is received as life / death monitoring data,
The virtual machine determines that the other virtual machine is not operating when alive monitoring data is not received within a predetermined timeout time by the virtual communication device,
When the command receiving unit receives a command to stop the virtual machine, the communication device control unit receives data received by the virtual communication device during the reception waiting time after the timeout time has elapsed as the reception waiting time. 3. The virtual computer control apparatus according to claim 1, wherein the virtual computer control apparatus stores the storage area in the storage area. The virtual machine control device further includes:
4. The virtual machine according to claim 1, further comprising: a filtering unit that causes the virtual communication device to receive only specific data when the command receiving unit receives a command to stop the virtual computer. 5. Control device. The virtual machine control device according to claim 4, wherein the filtering unit causes the virtual communication unit to receive only data transmitted from a specific virtual machine. The command receiving unit receives a stop command in which a stop time is designated,
The computer control unit stops the virtual computer at the stop time specified in the stop instruction,
The communication device control unit stores the data received in the reception waiting time by the virtual communication device in the storage area after the reception waiting time has elapsed from the stop time specified in the stop command. The virtual machine control device according to any one of claims 1 to 5. The virtual machine control device further includes:
Time information indicating the internal time of the virtual machine management device is received from the virtual machine management device, and the time difference between the internal time of the virtual machine management device and the internal time of the virtual machine control device based on the received time information A time adjustment unit for calculating
The command receiving unit receives a stop command in which a stop time is designated,
Based on the stop time specified in the stop command and the time difference calculated by the time adjustment unit, the computer control unit sets a stop timing at which the internal time of the virtual machine management device matches the stop time. Stop the virtual machine,
The communication device control unit stores the data received in the reception waiting time by the virtual communication device in the storage area after the reception waiting time has elapsed from the stop timing. The virtual machine control device according to claim 5. A virtual machine control device according to any one of claims 1 to 7,
A virtual machine control system comprising: a virtual machine management device that transmits a stop command for each of a plurality of virtual machines. In a virtual computer control method of a virtual computer control device in which a virtual communication device that performs communication processing and a virtual computer that communicates data via the virtual communication device are incorporated,
The instruction receiving unit receives a stop instruction for the virtual machine from a virtual machine management apparatus that manages a plurality of virtual machines including the virtual machine,
When the computer control unit receives the command to stop the virtual computer, the command receiving unit stops the virtual computer,
When the command receiving unit receives a command to stop the virtual machine, the communication device control unit receives the data transmitted by the other virtual machine before the other virtual machine is stopped by the command to stop the other virtual machine. A virtual computer control method for a virtual computer control device, wherein data received during the reception waiting time by the virtual communication device is stored in a predetermined storage area after a predetermined reception waiting time for reception has elapsed. . A virtual computer control program for causing a virtual computer control device to execute the virtual computer control method according to claim 9.

Images