WO2014087545A1 - Virtual machine deployment method and computer system - Google Patents

Virtual machine deployment method and computer system Download PDF

Info

Publication number
WO2014087545A1
WO2014087545A1 PCT/JP2012/081846 JP2012081846W WO2014087545A1 WO 2014087545 A1 WO2014087545 A1 WO 2014087545A1 JP 2012081846 W JP2012081846 W JP 2012081846W WO 2014087545 A1 WO2014087545 A1 WO 2014087545A1
Authority
WO
WIPO (PCT)
Prior art keywords
bandwidth
virtual machine
value
deployment
virtual
Prior art date
Application number
PCT/JP2012/081846
Other languages
French (fr)
Japanese (ja)
Inventor
成昊 金
崇博 大平
Original Assignee
株式会社日立製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社日立製作所 filed Critical 株式会社日立製作所
Priority to PCT/JP2012/081846 priority Critical patent/WO2014087545A1/en
Publication of WO2014087545A1 publication Critical patent/WO2014087545A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061Partitioning or combining of resources
    • G06F9/5077Logical partitioning of resources; Management or configuration of virtualized resources
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45562Creating, deleting, cloning virtual machine instances

Definitions

  • the present invention relates to a computer system and an information processing method, and more particularly to deployment of a virtual machine.
  • Patent Document 1 discloses an example of a method of selecting and deploying from a plurality of physical machines a physical machine that can secure the resource amount of the virtual machine 1 when newly deploying the virtual machine to the physical machine.
  • Patent Document 2 discloses an example of a method for predicting a load change of a virtual machine and allocating appropriate computer resources.
  • a virtualized environment will be constructed and operated in stages.
  • a virtualized system when a virtualized system is operating, it is necessary to newly virtualize another system and deploy it to the same physical machine.
  • the network bandwidth of the physical machine and virtual machine is temporarily used in large quantities, and the network bandwidth of the physical machine and the virtual machine already running Since a part is used for transferring the image file, there is a problem that the performance is deteriorated.
  • An object of the present invention is to guarantee the performance of an existing virtual machine when a new virtual machine is deployed.
  • the computer system according to the present invention is a computer system that includes a physical machine having a hypervisor that operates a virtual machine and a deployment management server that is responsible for deploying the virtual machine that manages the disk image file of the virtual machine.
  • the bandwidth management mechanism that controls the network bandwidth and sends the bandwidth usage status to the deployment management server, and receives virtual machine image files and writes them to disk when deploying virtual machines
  • a deployment daemon deployment daemon
  • a daemon is a program that operates as a background process.
  • the deployment management server has a deployment bandwidth control mechanism that controls the network transmission bandwidth of the image file.
  • the bandwidth management mechanism of the physical machine controls the network bandwidth of the virtual machine of the physical machine in consideration of the bandwidth usage status of the virtual machine, and the deployment bandwidth control mechanism of the deployment management server controls the network bandwidth of transferring the virtual machine image file. It is characterized by that.
  • the present invention is also configured as a deployment method executed by the computer system.
  • the deployment process when deploying a new virtual machine or updating a virtual machine, uses a part of the bandwidth of the existing virtual machine as in the past, thereby improving the performance of the existing virtual machine.
  • the network bandwidth used for deployment and the network bandwidth of the existing virtual machine it is possible to secure the necessary network bandwidth and guarantee the real-time property of the virtual machine.
  • FIG. 1 It is a figure which shows the structural example of a deployment application process. It is a figure which shows the structural example of a deployment daemon process. It is a figure which shows a bandwidth management table. It is a figure which shows a band history table. It is a figure which shows a band accommodation destination table. It is a figure which shows a virtual machine image transfer flag table. It is a figure which shows a deployment bandwidth table. It is a figure which shows a process outline. It is a figure which shows the relationship between the magnitude relationship of a parameter, and each process.
  • This example describes the initialization process and the virtual machine deployment process.
  • FIG. 1 shows a configuration of a computer system 10 according to an embodiment.
  • the computer system 10 includes a physical machine 100 that operates the virtual machine 120 and a deployment management server 200 that performs control for realizing the virtual machine on the physical machine 100.
  • Both the physical machine 100 and the deployment management server 200 are realized by a processing device having a CPU, a memory, an input / output terminal, and a storage device.
  • the physical machine 100 includes a hypervisor 110, a disk 170 for recording information, and a communication IF 160, and is connected to the deployment management server 200 via a network 260.
  • the hypervisor 110 includes a deployment daemon 130 that receives virtual machine image files from the virtual machines 120 and 121 and the deployment management server 200 and writes them to the disk 170.
  • the hypervisor 110 also schedules the virtual machines 120 and 121, and each virtual machine.
  • a bandwidth management mechanism 150 that manages bandwidth used by the machine and the deployment daemon 130 is provided.
  • the bandwidth management mechanism 150 performs bandwidth status monitoring and exchange of bandwidth (values) between virtual machines, records bandwidth management information used by each virtual machine and the deployment daemon 130 in the bandwidth management table 151, and The bandwidth used by the virtual machine and the deployment daemon 130 is recorded in the bandwidth history table 152.
  • the bandwidth management mechanism 150 controls the bandwidth used by each virtual machine based on the bandwidth management table 151.
  • the deployment management server 200 stores the virtual machine image file 210 for realizing the virtual machine on the physical machine, the communication IF 240, the deployment bandwidth control mechanism 230 for controlling the network bandwidth, and the disk image file 210 of the virtual machine 120 in the physical machine 100.
  • a deployment application 250 to be transferred and a display 220 for displaying the transfer status of the image file are provided.
  • the deployment bandwidth control mechanism 230 controls the transfer bandwidth of the virtual machine image file 210.
  • Each of the hypervisor 110 of the physical machine 100 and the deployment bandwidth control mechanism 230 and the deployment application 250 of the deployment management server 200 is realized by software (program), stored in each memory, and executed by the CPU. It is also possible to store these software in a computer-readable storage medium, read the software stored in the storage medium into a memory, and execute it on the CPU.
  • the bandwidth management table 151 shown in FIG. 13 is a table for managing information related to the bandwidth of each virtual machine and deployment daemon.
  • the management target 151a, the bandwidth threshold 151b, the margin value 151c, the accommodation value 151d, and these values Is composed of a band value 151e.
  • the definitions of the threshold value 151b, the margin value 151c, and the accommodation value 151d will be described later.
  • the bandwidth history table 152 shown in FIG. 14 is a table for managing how much bandwidth each virtual machine or deployment daemon is currently using, and includes a management target 152a and a used bandwidth 152b.
  • the bandwidth accommodation destination table 153 illustrated in FIG. 15 is a table for managing the correspondence between the virtual machine and the deployment daemon and other virtual machines and deployment daemons that allow bandwidth, the bandwidth accommodation source 153a and the bandwidth accommodation destination. 153b.
  • the virtual machine image transfer table 154 shown in FIG. 16 is a table for managing the presence / absence of transfer of virtual machine images for each virtual machine.
  • the deployment bandwidth table 231 shown in FIG. 17 is a table for managing a bandwidth value to be secured at the time of deployment for each virtual machine, and includes a virtual machine 231a and a deployment bandwidth value 231b.
  • FIG. 2 Initialization processing (FIG. 2): Virtual machine bandwidth allocation processing (FIG. 3) is performed.
  • the bandwidth is a movable bandwidth
  • the following mutual bandwidth interchange processing (FIG. 4) is repeatedly executed at regular time intervals during operation, and the virtual machine and the deployment are performed based on the magnitude relationship between the bandwidth to be used and the threshold. Flex bandwidth between daemons.
  • FIG. 6 When a virtual machine image file is being transferred (FIG. 6): The following processing is performed depending on whether the use bandwidth of the virtual machine is larger than the threshold.
  • FIG. 19 is a diagram showing the relationship between the magnitude relationship of the parameters related to the band and each of the above processes.
  • the processing executed during non-transfer or transfer of a virtual machine image file differs depending on the magnitude relationship between the use bandwidth of the virtual machine and the threshold and the interchange value of the virtual machine is 0 / less.
  • the numbers of the process flow diagrams described later are shown.
  • the initialization processing sequence is configured by each virtual machine bandwidth allocation processing (S301), deployment daemon bandwidth allocation processing (S302 to S305), priority setting processing (S306), and bandwidth value initial setting processing (S307, S308). Is done.
  • the bandwidth management mechanism 150 allocates a network bandwidth to be used for each virtual machine.
  • the bandwidth management mechanism 150 initializes the network bandwidth used by the deployment daemon 130 on the physical machine 100.
  • a predetermined value is set in the threshold 151b of the deployment daemon 130, and “0” is set in the margin value 151c.
  • the accommodation value 151d of the deployment daemon 130 is set to a value obtained by adding the margin value 151c of each virtual machine in the bandwidth management table 151 to the accommodation value 151d (S302).
  • the margin value 151c of each virtual machine in the bandwidth management table 151 is subtracted from the accommodation value 151d of each virtual machine in the bandwidth management table 151, and the subtraction result is set to the accommodation value 151d of each virtual machine (S303).
  • “deploy daemon” is set in the bandwidth accommodation source of the bandwidth accommodation destination table 153 (S304).
  • the bandwidth management mechanism 150 recalculates the bandwidth value 151e (threshold 151b + margin value 151c + accommodation value 151d) of each virtual machine and deployment daemon, and sets the bandwidth value in the bandwidth management table 151 (S305).
  • the threshold 151b and the margin value 151c are values of the bandwidth possessed by the virtual machine itself, and the accommodation value 151d is a bandwidth value accommodated from another virtual machine.
  • the priority setting process (S306) for the virtual machine and the deployment daemon will be described.
  • the schedule mechanism 140 that executes the priority setting process controls the execution priority of each virtual machine. This corresponds to a scheduler of a conventional OS, and is a part of the function executed by the hypervisor 110 in this embodiment.
  • the scheduling mechanism 140 on the physical machine 100 sets the scheduling priority of each virtual machine.
  • the schedule mechanism 140 sets the priority of the deployment daemon 130 to the lowest level. This is to minimize the influence of the operation of the deployment daemon 130 on each virtual machine and to guarantee the real-time property of the virtual machine 120.
  • the priority of the packet addressed to the deployment daemon 130 is set to the lowest level, and the setting is made so that the packet addressed to each virtual machine is processed with priority.
  • the bandwidth management mechanism 150 controls the bandwidth of each virtual machine and the deployment daemon according to the bandwidth value 151e of the bandwidth management table 151 (S307).
  • the bandwidth management mechanism 150 transmits the bandwidth value of the deployment daemon to the bandwidth control mechanism of the deployment management server (S308).
  • Steps S307 and S308 are processes for applying each value set in steps S301 to S306 first (before actual operation).
  • “controlling the bandwidth” means setting a bandwidth value in the hypervisor 110 that monitors bandwidth usage or restricts usage. Steps S307 and S308 correspond to the processing of FIG. 9 included in FIG. 4 to be described later.
  • step S301 of the initialization processing sequence in the physical machine 100 of the computer system 10 will be described.
  • the bandwidth management mechanism 150 on the physical machine 100 allocates a network bandwidth to be used for each virtual machine according to the requirements of the virtual machine. ,initialize.
  • the bandwidth management mechanism 150 records the threshold value and the margin value of the network bandwidth allocated to each virtual machine in the bandwidth management table 151 (S301-1).
  • the threshold 151b is a basic value of a virtual machine that is an index for performing processing for obtaining bandwidth from other virtual machines when there is a virtual machine that uses the bandwidth with a value larger than that. This is the value of the bandwidth required for proper operation.
  • the margin value 151c is a margin with respect to a threshold when designing a bandwidth used by the virtual machine, and is also a value at which the virtual machines can accommodate the bandwidth.
  • the virtual machine accommodation value 151d in the bandwidth management table 151 is initialized with "0" (S301-2).
  • the accommodation value 151 d is a numerical value indicating “a value of a bandwidth accommodated from another virtual machine”. It is set to “0” at the initialization stage. If the value of the accommodation value 151d is negative, it indicates that the bandwidth has been accommodated by another virtual machine.
  • the bandwidth value 151e of each virtual machine in the bandwidth management table 151 is calculated and recorded as the sum of the threshold value 151b, the margin value 151c, and the accommodation value 151d (S301-3).
  • the bandwidth management mechanism 150 controls the network bandwidth of each virtual machine with the bandwidth value 151e calculated here.
  • the threshold value which is the bandwidth value required for the basic operation of the virtual machine
  • the margin value and the accommodation value that can cope with the fluctuation of the bandwidth during operation are assigned to each virtual machine.
  • the bandwidth to be managed by each virtual machine is determined.
  • bandwidth is not interchanged between a plurality of virtual machines and between a virtual machine and a deployment daemon.
  • the bandwidth of each virtual machine 120 is controlled by the threshold value 151 b recorded in the bandwidth management table 151.
  • the deployment daemon 130 controls the bandwidth by the sum of the margin values 151c of the virtual machines recorded in the bandwidth management table 151.
  • the bandwidth history table 152 described in FIG. 1 is not necessary.
  • the inter-band interchange processing described below is not performed. Therefore, the bandwidth once allocated by other virtual machines during operation does not fluctuate.
  • ⁇ Operation example in movable band As an example of the embodiment, an operation example in the movable band is shown.
  • the bandwidth is mutually interchanged between virtual machines or between a virtual machine and a deployment daemon.
  • the virtual machine that does not use the bandwidth very much has the bandwidth interchanged. Thereby, the virtual machine and the deployment daemon can effectively use the bandwidth of the entire system.
  • Band mutual interchange processing is performed periodically with a period of, for example, about 1 second while the physical machine 100 is in operation. By repeatedly executing this bandwidth interchange processing during operation, the bandwidth demand of each virtual machine that changes during operation and the appropriate bandwidth allocation according to the temporary transfer of virtual machine image files (between virtual machines) (Band interchange) at any time.
  • the bandwidth interchange processing is performed by other virtual machines. Perform processing to accommodate the bandwidth.
  • a bandwidth whose total amount is limited can be effectively utilized among a plurality of virtual machines, and a necessary bandwidth can be temporarily allocated from the virtual machine to the deployment daemon 130 when transferring the virtual machine image file.
  • the bandwidth management mechanism 150 on the physical machine 100 refers to the virtual machine image transfer flag table 154 in FIG. 16 and determines whether the virtual machine image file is being transferred in the virtual machine (S401).
  • the mutual bandwidth interchange process (S402) during the virtual machine image transfer shown in FIG. 6 is performed for each virtual machine. Thereafter, a bandwidth history recording process (S404) is performed.
  • the bandwidth management mechanism 150 of the physical machine 100 records the bandwidth used by each virtual machine and the deployment daemon in the bandwidth history table 152 of FIG.
  • the bandwidth mutual interchange processing (S403) during non-transfer of the virtual machine image of FIG. 5 is performed for each virtual machine. Thereafter, a bandwidth history recording process (S404) is performed.
  • the mutual bandwidth interchange processing during non-transfer of the virtual machine image is processing that is performed when the virtual machine image is not transferred, and the bandwidth is interchanged between the virtual machines.
  • the bandwidth used by the deployment daemon that transfers virtual machine images can also be used effectively between virtual machines.
  • the bandwidth management mechanism 150 of the physical machine 100 determines whether or not the bandwidth used by the virtual machine is less than the threshold 151b with reference to the bandwidth history table 152 (S403-1).
  • the bandwidth interchange processing during non-transfer of the virtual machine image is terminated.
  • the bandwidth management mechanism 150 determines whether or not the accommodation value 151d of the virtual machine is 0 or more with reference to the bandwidth management table 151 (S403). -2).
  • the virtual machine bandwidth securing process (S403-4) in FIG. 8 is performed, and the process is terminated.
  • the virtual machine bandwidth return processing (S403-3) in FIG. 7 is performed, and the processing is terminated.
  • the virtual machine bandwidth return process is a process in which the virtual machine returns itself the bandwidth that has been accommodated by other virtual machines or deployment daemons. As a result, it is possible to cope with a sudden increase in the bandwidth required for processing by returning the bandwidth that has been accommodated elsewhere.
  • the bandwidth return destination virtual machine is referred to as an interchange source virtual machine.
  • the bandwidth management mechanism 150 adds and sets its own margin value 151c to the virtual machine accommodation value 151d (S501).
  • the bandwidth management mechanism 150 refers to the bandwidth accommodation destination table 153 of FIG. 15 to select a virtual machine of the accommodation source 153a, and from the accommodation value 151d of the virtual machine of the accommodation source 153b, the margin value 151c of the accommodation source virtual machine Is subtracted and set again (S502).
  • the bandwidth management mechanism 150 recalculates and sets the bandwidth value 151e of each virtual machine (S503).
  • the bandwidth value 151e is calculated in the bandwidth management table 151 as the threshold value 151b + the margin value 151c + the accommodation value 151d.
  • the calculated bandwidth value is set in the bandwidth value 151e of the bandwidth management table 151.
  • the bandwidth management mechanism 150 resets the accommodation destination 153b of the virtual machine of the accommodation source 153a of the bandwidth accommodation destination table 153 (S504).
  • the bandwidth management mechanism 150 controls the bandwidth of each virtual machine according to the bandwidth value 151e of the bandwidth management table 151 (S505).
  • the bandwidth management mechanism 150 transmits the bandwidth value 151e of the deployment daemon to the deployment bandwidth control mechanism 230 of the deployment management server 200 (S506).
  • the virtual machine bandwidth securing process is a process for securing the bandwidth by having the bandwidth required by the virtual machine interchanged from other virtual machines whose used bandwidth is less than the threshold 151b. As a result, it is guaranteed to cope with a sudden increase in bandwidth required for processing.
  • a virtual machine that performs the virtual machine bandwidth securing process is referred to as a bandwidth accommodation destination virtual machine.
  • the bandwidth management mechanism 150 refers to the bandwidth history table 151 and the bandwidth accommodation destination table 153.
  • the bandwidth management mechanism 150 uses less bandwidth than the threshold 151b and does not allow bandwidth to other virtual machines 120 and 121 or the deployment daemon 130. It is determined whether there is a virtual machine (S601).
  • the virtual machine bandwidth securing process is terminated.
  • the bandwidth management mechanism 150 uses a virtual machine that uses less bandwidth than the threshold 151b and does not accommodate bandwidth.
  • the bandwidth history table 152 in FIG. 14 is referred to, and the one having the smallest bandwidth is selected (S602).
  • the virtual machine selected here is referred to as an accommodation source virtual machine.
  • the bandwidth management mechanism 150 subtracts the accommodation value 151d of the selected accommodation source virtual machine from the accommodation value 151d of the accommodation source virtual machine selected in step S602, and sets it in the bandwidth management table 151 (S603).
  • the bandwidth management mechanism 150 adds and sets the margin value 151c of the accommodation source virtual machine selected in step S602 to the accommodation value 151d of the bandwidth accommodation destination virtual machine (S604).
  • the bandwidth management mechanism 150 sets the accommodation destination virtual machine in the item 153a of the accommodation source virtual machine selected in the process of step S602 in the bandwidth accommodation destination table 153 (S605).
  • the bandwidth management mechanism 150 of the physical machine 100 determines whether or not the bandwidth used by the virtual machine is less than the threshold 151b with reference to the bandwidth history table 152 (S402-1).
  • the bandwidth interchange processing (S402-4) to the deployment daemon 130 of FIG. 9 is performed, and the bandwidth mutual interchange processing during the virtual machine image transfer is terminated.
  • the bandwidth management mechanism 150 determines whether or not the accommodation value 151d of the virtual machine is 0 or more with reference to the bandwidth management table 151 (S402-). 2).
  • the bandwidth mutual accommodation processing during the virtual machine image transfer ends.
  • the virtual machine bandwidth return processing (S403-3) in FIG. 7 is performed, and the processing is terminated.
  • the bandwidth accommodation processing to the deployment daemon 130 is processing for accommodating bandwidth to the deployment daemon 130 when the bandwidth usage of the virtual machine is small and there is room. As a result, since the bandwidth is accommodated from the spare virtual machine to the deployment daemon 130, it is not necessary to increase the bandwidth of the entire system for transferring the virtual machine image file.
  • the bandwidth management mechanism 150 refers to the bandwidth management table 151, and subtracts its own margin value 151c from the virtual machine accommodation value 151d (S701).
  • the bandwidth management mechanism 150 adds the margin value 151c of the virtual machine to the accommodation value 151d of the deployment daemon 130 and resets the setting (S702).
  • the bandwidth management mechanism 150 sets a deployment daemon in the virtual machine accommodation destination 153b of the bandwidth accommodation destination table 153 (S703).
  • the bandwidth management mechanism 150 controls the bandwidth of each of the virtual machines 120 and 121 and the deployment daemon 130 according to the bandwidth value 151e of the bandwidth management table (S704).
  • the bandwidth management mechanism 150 transmits the bandwidth value 151e of the deployment daemon to the deployment bandwidth control mechanism 230 of the deployment management server 200 (S705).
  • the deployment bandwidth control mechanism 230 of the deployment management server 200 controls the transfer bandwidth of the virtual machine image file with the bandwidth value 151e of the deployment daemon received from the bandwidth management mechanism 150 (S706).
  • the deployment bandwidth control mechanism process is a process performed by the deployment bandwidth control mechanism 230 of the deployment management server 200 to control the bandwidth of the deployment daemon 130. Since the bandwidth of the deployment daemon 130 is controlled by the bandwidth obtained as a result of the bandwidth interchange between the virtual machines in the physical machine 100, the operation of the virtual machine is not affected.
  • the deployment bandwidth control mechanism 230 records the bandwidth value 151e of the deployment daemon 130 received from the bandwidth management mechanism 150 of the physical machine 100 in the deployment bandwidth table 231 of FIG. 17 (S802).
  • the deployment bandwidth control mechanism 230 controls the bandwidth of the deployment application 250 with the deployment bandwidth value 231b of the deployment bandwidth table 231 (S803).
  • the deploy application process is a process for the deploy application 250 of the deployment management server 200 to transfer the virtual machine file image 210 to the physical server 100.
  • the deployment management server 200 collectively manages various virtual machine file images 210 and provides them to each of the virtual machines, thereby eliminating the need for the physical machine 100 to prepare various virtual machine file images 210.
  • the deployment application 250 starts transferring the virtual machine image file 210 to the deployment daemon 130 of the physical machine 100 (S902).
  • the deployment application 250 calculates the transfer completion time by dividing the remaining transfer amount of the virtual machine image file 210 by the current use bandwidth, and displays it on the display 220 (S904). Return to the virtual machine image file transfer determination process.
  • the deployment daemon process is a process in which the physical machine deployment daemon 130 receives a virtual machine image file from the deployment application 250 of the deployment management server 200 and writes it to the disk 170.
  • the virtual machine image file can be shared between the virtual machines, and the virtual machine image used by one virtual machine can be re-used by another virtual machine. It can be used, and the transfer of the virtual machine image file between the deployment management server 200 and the physical machine 100 and the bandwidth interchange processing associated with the transfer can be omitted.
  • the deployment daemon 130 writes the received virtual machine image file 210 to the disk 170 (S1003).
  • the deploy daemon 130 records “1” in the flag 154b of the transfer flag table 154 (S1004).
  • the virtual machine image file is transferred from the deployment management server 200, but the virtual machine image file can also be transferred from an external medium such as a DVD to the deployment daemon 130.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Stored Programmes (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A physical machine that has a hypervisor for running a virtual machine, the physical machine further having: a bandwidth management mechanism for controlling a network bandwidth and transmitting bandwidth usage to a deployment management server; deployment daemon software for receiving an image file for a virtual machine when the virtual machine is deployed, and writing the image file to a disk; and a scheduling mechanism for performing scheduling for the virtual machine and the deployment daemon. A deployment management server for managing the virtual machine image file has a deployment bandwidth control mechanism for controlling the network transmission bandwidth of the image file. The deployment bandwidth control mechanism in the deployment management server controls the network bandwidth.

Description

仮想マシンデプロイ方法および計算機システムVirtual machine deployment method and computer system
 本発明は、計算機システム及び情報処理方法に係り、特に仮想マシンのデプロイに関する。 The present invention relates to a computer system and an information processing method, and more particularly to deployment of a virtual machine.
 高信頼な処理と高度なリアルタイム性(応答性能)が求められるミッションクリティカルな計算機システムを仮想化する際には、仮想マシンの性能を保証する必要がある。通常、仮想マシンに十分な計算機資源を割り当てることで性能を担保している。そのため、仮想マシンの性能を担保するためのデプロイが行われる。デプロイ(deploy)とは、ネットワーク経由で利用されるソフトウエアなどのコンポーネントを利用可能な状態にすることである。例えば、予め仮想マシンに必要なメモリ量、ディスク量、CPU使用時間を算出し、割り当てることにより性能を保証する。特許文献1には、仮想マシンを物理マシンに新規デプロイする際に、仮想マシン1の資源量が確保可能な物理マシンを複数の物理マシンから選択し、デプロイする方法の例が開示されている。また、特許文献2には、仮想マシンの負荷変化を予測し適切な計算機資源を割り当てる方式の例が開示されている。 When virtualizing a mission-critical computer system that requires highly reliable processing and advanced real-time performance (response performance), it is necessary to guarantee the performance of the virtual machine. Usually, performance is secured by allocating sufficient computer resources to a virtual machine. Therefore, deployment for ensuring the performance of the virtual machine is performed. “Deploy” refers to making a component such as software used via a network available. For example, the memory capacity, the disk capacity, and the CPU usage time necessary for the virtual machine are calculated and allocated in advance to guarantee the performance. Patent Document 1 discloses an example of a method of selecting and deploying from a plurality of physical machines a physical machine that can secure the resource amount of the virtual machine 1 when newly deploying the virtual machine to the physical machine. Patent Document 2 discloses an example of a method for predicting a load change of a virtual machine and allocating appropriate computer resources.
特開2011-198332号公報JP 2011-198332 A 特開2011-170787号公報JP 2011-170787 A
 長期にわたるシステムの運用・保守にあたっては、仮想化環境を段階的に構築して運用ことになる。つまり、仮想化されたシステムが動作している際に、新たに別のシステムを仮想化し、同一の物理マシンにデプロイする必要が出てくる。そのような場合、同一の物理マシンに別の仮想マシンを実現するために、大容量の仮想マシンイメージファイルを物理マシンに配置する必要がでてくる。しかし、ネットワークを用いて、この計算機イメージファイルを転送する場合、一時的に物理マシンや仮想マシンのネットワーク帯域を大量に使用してしまい、物理マシン、および、すでに動作中の仮想マシンのネットワーク帯域の一部がイメージファイルの転送に利用されるので、その性能が劣化してしまうという課題がある。 For long-term system operation and maintenance, a virtualized environment will be constructed and operated in stages. In other words, when a virtualized system is operating, it is necessary to newly virtualize another system and deploy it to the same physical machine. In such a case, in order to realize another virtual machine on the same physical machine, it is necessary to arrange a large-capacity virtual machine image file on the physical machine. However, when transferring this computer image file using a network, the network bandwidth of the physical machine and virtual machine is temporarily used in large quantities, and the network bandwidth of the physical machine and the virtual machine already running Since a part is used for transferring the image file, there is a problem that the performance is deteriorated.
 本発明の目的は、新規に仮想マシンをデプロイする際、既存の仮想マシンの性能を保証することにある。 An object of the present invention is to guarantee the performance of an existing virtual machine when a new virtual machine is deployed.
 本発明に係る計算機システムは、仮想マシンを動作させるハイパーバイザを有する物理マシンと、仮想マシンのディスクイメージファイルを管理する仮想マシンのデプロイを担当するデプロイ管理サーバによって構成された計算機システムである。 The computer system according to the present invention is a computer system that includes a physical machine having a hypervisor that operates a virtual machine and a deployment management server that is responsible for deploying the virtual machine that manages the disk image file of the virtual machine.
 物理マシンには、ネットワーク帯域を制御し、帯域使用状況をデプロイ管理サーバに送信する帯域管理機構と、仮想マシンのデプロイの際に仮想マシンのイメージファイルの受信、および、ディスクへの書きだしをおこなうデプロイデーモン(deploy daemon)と、仮想マシン、および、デプロイデーモンのスケジュールを行うスケジュール機構を有する。デーモンは、バックグラウンドプロセスとして動作するプログラムである。デプロイ管理サーバには、イメージファイルのネットワーク送信帯域を制御するデプロイ帯域制御機構を有する。 For physical machines, the bandwidth management mechanism that controls the network bandwidth and sends the bandwidth usage status to the deployment management server, and receives virtual machine image files and writes them to disk when deploying virtual machines A deployment daemon (deploy daemon), a virtual machine, and a schedule mechanism for scheduling the deployment daemon are included. A daemon is a program that operates as a background process. The deployment management server has a deployment bandwidth control mechanism that controls the network transmission bandwidth of the image file.
 仮想マシンの帯域使用状況を考慮して物理マシンの帯域管理機構が物理マシンの仮想マシンのネットワーク帯域を制御し、デプロイ管理サーバのデプロイ帯域制御機構が仮想マシンイメージファイルを転送するネットワーク帯域を制御することを特徴とする。 The bandwidth management mechanism of the physical machine controls the network bandwidth of the virtual machine of the physical machine in consideration of the bandwidth usage status of the virtual machine, and the deployment bandwidth control mechanism of the deployment management server controls the network bandwidth of transferring the virtual machine image file. It is characterized by that.
 本発明はまた、上記計算機システムで実行されるデプロイ方法としても構成される。 The present invention is also configured as a deployment method executed by the computer system.
 本発明によれば、新規仮想マシンのデプロイ、あるいは仮想マシンの更新の際に、従来のようにデプロイ処理が既存の仮想マシンの帯域の一部を利用することにより、既存の仮想マシンの性能を悪化させることを防ぎ、デプロイに使用するネットワーク帯域、および、既存仮想マシンのネットワーク帯域を制御することにより、必要なネットワーク帯域を確保して仮想マシンのリアルタイム性を保証することができる。 According to the present invention, when deploying a new virtual machine or updating a virtual machine, the deployment process uses a part of the bandwidth of the existing virtual machine as in the past, thereby improving the performance of the existing virtual machine. By preventing the deterioration and controlling the network bandwidth used for deployment and the network bandwidth of the existing virtual machine, it is possible to secure the necessary network bandwidth and guarantee the real-time property of the virtual machine.
計算機システムの構成例を示す図である。It is a figure which shows the structural example of a computer system. 初期化処理を説明するフローチャートである。It is a flowchart explaining an initialization process. 仮想マシン用帯域割り当て処理を説明するフローチャートである。It is a flowchart explaining the bandwidth allocation process for virtual machines. 帯域相互融通処理を説明するフローチャートである。It is a flowchart explaining a band mutual accommodation process. 仮想マシンイメージ非転送中の帯域相互融通処理を説明するフローチャートである。10 is a flowchart for explaining mutual bandwidth interchange processing during virtual machine image non-transfer. 仮想マシンイメージ転送中の帯域相互融通処理を説明するフローチャートである。It is a flowchart explaining the band mutual accommodation process during virtual machine image transfer. 仮想マシン帯域返還処理の構成例を示す図である。It is a figure which shows the structural example of a virtual machine zone | band return process. 仮想マシン帯域確保処理の構成例を示す図である。It is a figure which shows the structural example of a virtual machine bandwidth ensuring process. デプロイデーモンへの帯域融通処理の構成例を示す図である。It is a figure which shows the structural example of the band accommodation process to a deployment daemon. デプロイ帯域制御機構処理の構成例を示す図である。It is a figure which shows the structural example of a deployment bandwidth control mechanism process. デプロイアプリ処理の構成例を示す図である。It is a figure which shows the structural example of a deployment application process. デプロイデーモン処理の構成例を示す図である。It is a figure which shows the structural example of a deployment daemon process. 帯域管理テーブルを示す図である。It is a figure which shows a bandwidth management table. 帯域履歴テーブルを示す図である。It is a figure which shows a band history table. 帯域融通先テーブルを示す図である。It is a figure which shows a band accommodation destination table. 仮想マシンイメージ転送フラグテーブルを示す図である。It is a figure which shows a virtual machine image transfer flag table. デプロイ帯域テーブルを示す図である。It is a figure which shows a deployment bandwidth table. 処理概要を示す図である。It is a figure which shows a process outline. パラメータの大小関係と各処理との関係を示す図である。It is a figure which shows the relationship between the magnitude relationship of a parameter, and each process.
 以下、図面を参照して、本発明の実施形態について説明する。 Hereinafter, embodiments of the present invention will be described with reference to the drawings.
 本実施例は、初期化処理および仮想マシンのデプロイ処理について説明するものである。 This example describes the initialization process and the virtual machine deployment process.
 <システム構成>
 図1は、一実施形態による計算機システム10の構成を示す。
<System configuration>
FIG. 1 shows a configuration of a computer system 10 according to an embodiment.
 計算機システム10は、仮想マシン120を動作させる物理マシン100、および、物理マシン100で仮想マシンを実現するための制御を行うデプロイ管理サーバ200から構成される。物理マシン100及びデプロイ管理サーバ200のいずれも、CPU、メモリ、入出力端末及び記憶装置を有する処理装置によって実現される。 The computer system 10 includes a physical machine 100 that operates the virtual machine 120 and a deployment management server 200 that performs control for realizing the virtual machine on the physical machine 100. Both the physical machine 100 and the deployment management server 200 are realized by a processing device having a CPU, a memory, an input / output terminal, and a storage device.
 物理マシン100は、ハイパーバイザ110、情報を記録するディスク170、および、通信IF160を備え、デプロイ管理サーバ200とネットワーク260を介して接続されている。ハイパーバイザ110は、仮想マシン120、121、および、デプロイ管理サーバ200から仮想マシンイメージファイルを受け取ってディスク170に書き出すデプロイデーモン130を備え、また、それらのスケジューリングを行うスケジュール機構140、および、各仮想マシンとデプロイデーモン130が使用する帯域を管理する帯域管理機構150を備える。帯域管理機構150は、帯域状況の監視、及び仮想マシン間での帯域(の値)のやり取りを行い、各仮想マシンとデプロイデーモン130が使用する帯域管理情報を帯域管理テーブル151に記録し、各仮想マシンとデプロイデーモン130が使用中の帯域を帯域履歴テーブル152に記録する。また、帯域管理機構150は、帯域管理テーブル151に基づき、各仮想マシンが使用する帯域を制御する。 The physical machine 100 includes a hypervisor 110, a disk 170 for recording information, and a communication IF 160, and is connected to the deployment management server 200 via a network 260. The hypervisor 110 includes a deployment daemon 130 that receives virtual machine image files from the virtual machines 120 and 121 and the deployment management server 200 and writes them to the disk 170. The hypervisor 110 also schedules the virtual machines 120 and 121, and each virtual machine. A bandwidth management mechanism 150 that manages bandwidth used by the machine and the deployment daemon 130 is provided. The bandwidth management mechanism 150 performs bandwidth status monitoring and exchange of bandwidth (values) between virtual machines, records bandwidth management information used by each virtual machine and the deployment daemon 130 in the bandwidth management table 151, and The bandwidth used by the virtual machine and the deployment daemon 130 is recorded in the bandwidth history table 152. The bandwidth management mechanism 150 controls the bandwidth used by each virtual machine based on the bandwidth management table 151.
 デプロイ管理サーバ200は、物理マシンで仮想マシンを実現するための仮想マシンのイメージファイル210、通信IF240、ネットワーク帯域を制御するデプロイ帯域制御機構230、仮想マシン120のディスクイメージファイル210を物理マシン100に転送するデプロイアプリ250、及びそのイメージファイルの転送状況を表示するためのディスプレイ220を備える。デプロイ帯域制御機構230は、仮想マシンのイメージファイル210の転送帯域を制御する。 The deployment management server 200 stores the virtual machine image file 210 for realizing the virtual machine on the physical machine, the communication IF 240, the deployment bandwidth control mechanism 230 for controlling the network bandwidth, and the disk image file 210 of the virtual machine 120 in the physical machine 100. A deployment application 250 to be transferred and a display 220 for displaying the transfer status of the image file are provided. The deployment bandwidth control mechanism 230 controls the transfer bandwidth of the virtual machine image file 210.
 物理マシン100のハイパーバイザ110、及びデプロイ管理サーバ200のデプロイ帯域制御機構230とデプロイアプリ250のそれぞれはソフトウエア(プログラム)で実現され、それぞれのメモリに格納され、CPUによって実行される。また、これらのソフトウエアを計算機で読み取り可能な記憶媒体に格納し、記憶媒体に格納されたソフトウエアをメモリに読み込んでCPUで実行することもできる。 Each of the hypervisor 110 of the physical machine 100 and the deployment bandwidth control mechanism 230 and the deployment application 250 of the deployment management server 200 is realized by software (program), stored in each memory, and executed by the CPU. It is also possible to store these software in a computer-readable storage medium, read the software stored in the storage medium into a memory, and execute it on the CPU.
 <各テーブルの説明>     
 図13に示す帯域管理テーブル151は、各仮想マシンやデプロイデーモンの帯域に関する情報を管理するテーブルであり、管理の対象151a、帯域の閾値151b、余裕値151c及び融通値151d、更に、これらの値を合計した帯域値151eで構成される。閾値151b、余裕値151c、及び融通値151dの定義は後述する。
<Description of each table>
The bandwidth management table 151 shown in FIG. 13 is a table for managing information related to the bandwidth of each virtual machine and deployment daemon. The management target 151a, the bandwidth threshold 151b, the margin value 151c, the accommodation value 151d, and these values Is composed of a band value 151e. The definitions of the threshold value 151b, the margin value 151c, and the accommodation value 151d will be described later.
 図14に示す帯域履歴テーブル152は、各仮想マシンやデプロイデーモンが現在どの位の帯域を使用しているかを管理するテーブルであり、管理の対象152a、及び使用帯域152bで構成される。 The bandwidth history table 152 shown in FIG. 14 is a table for managing how much bandwidth each virtual machine or deployment daemon is currently using, and includes a management target 152a and a used bandwidth 152b.
 図15に示す帯域融通先テーブル153は、当該仮想マシンやデプロイデーモンと、帯域を融通している他の仮想マシンやデプロイデーモンとの対応を管理するテーブルであり、帯域融通元153a及び帯域融通先153bで構成される。 The bandwidth accommodation destination table 153 illustrated in FIG. 15 is a table for managing the correspondence between the virtual machine and the deployment daemon and other virtual machines and deployment daemons that allow bandwidth, the bandwidth accommodation source 153a and the bandwidth accommodation destination. 153b.
 図16に示す仮想マシンイメージ転送テーブル154は、仮想マシンイメージの転送の有無を仮想マシン毎に管理するテーブルであり、転送先の仮想マシン154a及び転送の有無を示すフラグ154b(転送ありの場合は「1」)で構成される。 The virtual machine image transfer table 154 shown in FIG. 16 is a table for managing the presence / absence of transfer of virtual machine images for each virtual machine. The transfer destination virtual machine 154a and the flag 154b indicating the presence / absence of transfer (if there is transfer) “1”).
 図17に示すデプロイ帯域テーブル231は、仮想マシン毎にデプロイの際に確保すべき帯域値を管理するテーブルであり、仮想マシン231a及びデプロイ帯域値231bで構成される。 The deployment bandwidth table 231 shown in FIG. 17 is a table for managing a bandwidth value to be secured at the time of deployment for each virtual machine, and includes a virtual machine 231a and a deployment bandwidth value 231b.
 <処理概要>
 図18に基づいて、物理マシン100及びデプロイ管理サーバ200における処理概要を説明する。( )内は関連する図である。
<Process overview>
Based on FIG. 18, an outline of processing in the physical machine 100 and the deployment management server 200 will be described. Figures in parentheses are related figures.
 物理マシン:
 (1)初期化処理(図2):仮想マシン用帯域割り当て処理(図3)を行う。
Physical machine:
(1) Initialization processing (FIG. 2): Virtual machine bandwidth allocation processing (FIG. 3) is performed.
 (2)帯域が固定帯域の場合は以下の処理を行わず、固定帯域での運用を行う。 (2) When the bandwidth is a fixed bandwidth, the following processing is not performed and the operation is performed in the fixed bandwidth.
 (3)帯域が可動帯域の場合は、運用中に、一定時間間隔で以下の帯域相互融通処理(図4)を繰り返し実行し、使用する帯域と閾値との大小関係に基づいて仮想マシン及びデプロイデーモンの間で帯域を融通する。 (3) If the bandwidth is a movable bandwidth, the following mutual bandwidth interchange processing (FIG. 4) is repeatedly executed at regular time intervals during operation, and the virtual machine and the deployment are performed based on the magnitude relationship between the bandwidth to be used and the threshold. Flex bandwidth between daemons.
  (3-1)仮想マシンイメージファイルが非転送中の場合(図5):
   仮想マシンの帯域の融通値が0以上かどうかに応じて以下の処理を行う。
(3-1) When the virtual machine image file is not being transferred (FIG. 5):
The following processing is performed according to whether the bandwidth interchange value of the virtual machine is 0 or more.
    (0以下)他に融通していた帯域を自身に返還させる仮想マシン帯域返還処理(図7)
    (0以上)余裕のある帯域を他から融通してもらう仮想マシン帯域確保処理(図8)
  (3-2)仮想マシンイメージファイルが転送中の場合(図6):
   仮想マシンの使用帯域が閾値よりも大きいかどうかに応じて以下の処理を行う。
(0 or less) Virtual machine bandwidth return processing to return the bandwidth that was otherwise accommodated to itself (FIG. 7)
(0 or more) Virtual machine bandwidth reservation processing to allow extra bandwidth (Figure 8)
(3-2) When a virtual machine image file is being transferred (FIG. 6):
The following processing is performed depending on whether the use bandwidth of the virtual machine is larger than the threshold.
    (>閾値)他に融通していた帯域を自身に返還させる仮想マシン帯域返還処理(図7)
    (<閾値)イメージファイルの転送に必要な帯域をデプロイデーモンに融通する帯域融通処理(図9)
 デプロイ管理サーバ:
 (1)デプロイデーモンの帯域を制御(図10)
 (2)デプロイ開始命令に基づいて仮想マシンイメージファイルを物理マシンに転送(図11)→物理マシンのデプロイデーモンがファイルをディスクに格納(図12)
 図19は、帯域に関するパラメータの大小関係と上記の各処理との関係を示す図である。
(> Threshold) Virtual machine bandwidth return processing to return the bandwidth that was otherwise accommodated to itself (FIG. 7)
(<Threshold) Bandwidth interchange processing for accommodating the bandwidth required for image file transfer to the deployment daemon (FIG. 9)
Deployment management server:
(1) Control the bandwidth of the deployment daemon (Figure 10)
(2) Transfer virtual machine image file to physical machine based on deployment start command (FIG. 11) → physical machine deployment daemon stores file on disk (FIG. 12)
FIG. 19 is a diagram showing the relationship between the magnitude relationship of the parameters related to the band and each of the above processes.
 後述するように、仮想マシンの使用帯域と閾値との大小関係、及び仮想マシンの融通値が0以上/以下に応じて仮想マシンイメージファイルの非転送中あるいは転送中に実行される処理が異なる。図19の( )内には後述する処理フローの図の番号を示す。 As will be described later, the processing executed during non-transfer or transfer of a virtual machine image file differs depending on the magnitude relationship between the use bandwidth of the virtual machine and the threshold and the interchange value of the virtual machine is 0 / less. In () of FIG. 19, the numbers of the process flow diagrams described later are shown.
 <初期化処理シーケンス>
 図2を参照して、計算機システム10の物理マシン100における初期化処理シーケンスについて説明する。
<Initialization processing sequence>
An initialization process sequence in the physical machine 100 of the computer system 10 will be described with reference to FIG.
 初期化処理シーケンスは、各仮想マシン用帯域割り当て処理(S301)、デプロイデーモン用帯域割り当て処理(S302~S305)、優先度設定処理(S306)、及び帯域値初期設定処理(S307、S308)により構成される。 The initialization processing sequence is configured by each virtual machine bandwidth allocation processing (S301), deployment daemon bandwidth allocation processing (S302 to S305), priority setting processing (S306), and bandwidth value initial setting processing (S307, S308). Is done.
 各仮想マシン用帯域割り当て処理(S301、図3にて詳細に説明する)では、各仮想マシンの要件(要求される処理性能、処理の重要度や緊急度など)に合わせ、物理マシン100上の帯域管理機構150が、各仮想マシン用に使用するネットワーク帯域を割り当てる。 In each virtual machine bandwidth allocation process (S301, which will be described in detail in FIG. 3), on the physical machine 100 in accordance with the requirements of each virtual machine (required processing performance, importance of processing, urgency, etc.) The bandwidth management mechanism 150 allocates a network bandwidth to be used for each virtual machine.
 デプロイデーモン用帯域割り当て処理(S302)では、物理マシン100上で、帯域管理機構150がデプロイデーモン130の使用するネットワーク帯域を初期化する。まず、デプロイデーモン130の閾値151bに所定の値を設定し、余裕値151cに「0」を設定する。前記デプロイデーモン130の融通値151dには、帯域管理テーブル151の各仮想マシンの余裕値151cを融通値151dに加算した値とする(S302)。前記帯域管理テーブル151の各仮想マシンの余裕値151cを、帯域管理テーブル151の各仮想マシンの融通値151dから減算し、減算結果を各仮想マシンの融通値151dに設定する(S303)。また、帯域融通先テーブル153の帯域融通元には「デプロイデーモン」を設定する(S304)。帯域管理機構150は、各仮想マシンおよびデプロイデーモンの帯域値151e(閾値151b+余裕値151c+融通値151d)を再計算し、帯域管理テーブル151に設定する(S305)。閾値151b及び余裕値151cは、仮想マシン自身が持っている帯域の値であり、融通値151dは他の仮想マシンから融通された帯域の値である。 In the deployment daemon bandwidth allocation process (S302), the bandwidth management mechanism 150 initializes the network bandwidth used by the deployment daemon 130 on the physical machine 100. First, a predetermined value is set in the threshold 151b of the deployment daemon 130, and “0” is set in the margin value 151c. The accommodation value 151d of the deployment daemon 130 is set to a value obtained by adding the margin value 151c of each virtual machine in the bandwidth management table 151 to the accommodation value 151d (S302). The margin value 151c of each virtual machine in the bandwidth management table 151 is subtracted from the accommodation value 151d of each virtual machine in the bandwidth management table 151, and the subtraction result is set to the accommodation value 151d of each virtual machine (S303). Further, “deploy daemon” is set in the bandwidth accommodation source of the bandwidth accommodation destination table 153 (S304). The bandwidth management mechanism 150 recalculates the bandwidth value 151e (threshold 151b + margin value 151c + accommodation value 151d) of each virtual machine and deployment daemon, and sets the bandwidth value in the bandwidth management table 151 (S305). The threshold 151b and the margin value 151c are values of the bandwidth possessed by the virtual machine itself, and the accommodation value 151d is a bandwidth value accommodated from another virtual machine.
 仮想マシンおよびデプロイデーモンへの優先度設定処理(S306)について説明する。優先度設定処理を実行するスケジュール機構140は、各仮想マシンの実行の優先度を制御している。これは、従来のOSのスケジューラに対応しており、本実施例では、ハイパーバイザ110が実行する機能の一部である。各仮想マシンの要件に合わせ、物理マシン100上のスケジュール機構140は各仮想マシンのスケジューリング優先度を設定する。また、スケジュール機構140はデプロイデーモン130の優先度を最低レベルに設定する。これは、デプロイデーモン130の動作が各仮想マシンに与える影響を最小限にし、仮想マシン120のリアルタイム性を保証するためである。また、ネットワークパケットの優先度についても、デプロイデーモン130あてのパケットの優先度は最低レベルとし、各仮想マシンあてのパケットが優先して処理されるよう設定を行うものとする。 The priority setting process (S306) for the virtual machine and the deployment daemon will be described. The schedule mechanism 140 that executes the priority setting process controls the execution priority of each virtual machine. This corresponds to a scheduler of a conventional OS, and is a part of the function executed by the hypervisor 110 in this embodiment. In accordance with the requirements of each virtual machine, the scheduling mechanism 140 on the physical machine 100 sets the scheduling priority of each virtual machine. The schedule mechanism 140 sets the priority of the deployment daemon 130 to the lowest level. This is to minimize the influence of the operation of the deployment daemon 130 on each virtual machine and to guarantee the real-time property of the virtual machine 120. Also, regarding the priority of the network packet, the priority of the packet addressed to the deployment daemon 130 is set to the lowest level, and the setting is made so that the packet addressed to each virtual machine is processed with priority.
 帯域管理機構150が、各仮想マシン及びデプロイデーモンの帯域を帯域管理テーブル151の帯域値151eに合わせ、制御する (S307)。帯域管理機構150が、デプロイデーモンの帯域値を、デプロイ管理サーバの帯域制御機構に送信する (S308)。ステップS307及びS308は、ステップS301~S306で設定した各値を最初に(実際の運用前に)適用する処理である。ここで、「帯域を制御する」とは、帯域の利用状況の監視、あるいは利用の制限を行うハイパーバイザ110に帯域の値を設定することである。また、ステップS307、S308は、後述する図4に含まれる図9の処理に対応する。 The bandwidth management mechanism 150 controls the bandwidth of each virtual machine and the deployment daemon according to the bandwidth value 151e of the bandwidth management table 151 (S307). The bandwidth management mechanism 150 transmits the bandwidth value of the deployment daemon to the bandwidth control mechanism of the deployment management server (S308). Steps S307 and S308 are processes for applying each value set in steps S301 to S306 first (before actual operation). Here, “controlling the bandwidth” means setting a bandwidth value in the hypervisor 110 that monitors bandwidth usage or restricts usage. Steps S307 and S308 correspond to the processing of FIG. 9 included in FIG. 4 to be described later.
 <仮想マシン用帯域初期化処理>
 図3を参照して、計算機システム10の物理マシン100における初期化処理シーケンスのステップS301の詳細について説明する。
<Virtual machine bandwidth initialization processing>
With reference to FIG. 3, the details of step S301 of the initialization processing sequence in the physical machine 100 of the computer system 10 will be described.
 仮想マシン用帯域割り当て処理(S301)では、最初に(実際の運用前に)、仮想マシンの要件に合わせ、物理マシン100上の帯域管理機構150が、各仮想マシン用に使用するネットワーク帯域を割り当て、初期化する。 In the virtual machine bandwidth allocation process (S301), first, before the actual operation, the bandwidth management mechanism 150 on the physical machine 100 allocates a network bandwidth to be used for each virtual machine according to the requirements of the virtual machine. ,initialize.
 帯域管理機構150は各仮想マシンに割り当てたネットワーク帯域の、閾値および余裕値を帯域管理テーブル151に記録する(S301-1)。ネットワーク帯域の設定には、各仮想マシンへの割り込みによる負荷も考慮し、リアルタイム性が確保可能なように設計する。ここでいう閾値151bとは、それ以上の値で帯域を使用している仮想マシンがあった場合に、他の仮想マシンから帯域を融通してもらう処理を行う指標となる、仮想マシンの基本的な動作に必要な帯域の値である。また、余裕値151cとは、仮想マシンが使用する帯域を設計する際の閾値に対するマージンであり、かつ、仮想マシンが互いに帯域を融通する値でもある。 The bandwidth management mechanism 150 records the threshold value and the margin value of the network bandwidth allocated to each virtual machine in the bandwidth management table 151 (S301-1). When setting the network bandwidth, consider the load caused by interrupts to each virtual machine, and design it to ensure real-time performance. Here, the threshold 151b is a basic value of a virtual machine that is an index for performing processing for obtaining bandwidth from other virtual machines when there is a virtual machine that uses the bandwidth with a value larger than that. This is the value of the bandwidth required for proper operation. Further, the margin value 151c is a margin with respect to a threshold when designing a bandwidth used by the virtual machine, and is also a value at which the virtual machines can accommodate the bandwidth.
 帯域管理テーブル151の仮想マシンの融通値151dを「0」で初期化する(S301-2)。ここで、融通値151dとは、「他の仮想マシンから融通された帯域の値」を表す数値である。初期化の段階では「0」とする。融通値151dの値が負の場合は、他の仮想マシンに帯域を融通したことを表す。 The virtual machine accommodation value 151d in the bandwidth management table 151 is initialized with "0" (S301-2). Here, the accommodation value 151 d is a numerical value indicating “a value of a bandwidth accommodated from another virtual machine”. It is set to “0” at the initialization stage. If the value of the accommodation value 151d is negative, it indicates that the bandwidth has been accommodated by another virtual machine.
 帯域管理テーブル151の各仮想マシンの帯域値151eについては、閾値151b、余裕値151c、および、融通値151dの和として算出し、記録する(S301-3)。ここで算出する帯域値151eにて帯域管理機構150は各仮想マシンのネットワーク帯域を制御する。 The bandwidth value 151e of each virtual machine in the bandwidth management table 151 is calculated and recorded as the sum of the threshold value 151b, the margin value 151c, and the accommodation value 151d (S301-3). The bandwidth management mechanism 150 controls the network bandwidth of each virtual machine with the bandwidth value 151e calculated here.
 上記の初期化処理により、各仮想マシンに対して、仮想マシンの基本的な動作に必要な帯域の値である閾値だけでなく、動作時の帯域の変動に対応できる余裕値や融通値が割り当てられ、各仮想マシンで管理すべき帯域が定められる。 By the above initialization process, not only the threshold value, which is the bandwidth value required for the basic operation of the virtual machine, but also the margin value and the accommodation value that can cope with the fluctuation of the bandwidth during operation are assigned to each virtual machine. The bandwidth to be managed by each virtual machine is determined.
 <固定帯域での運用例>
 実施例の一例として、固定帯域での運用例を示す。固定帯域での運用においては、複数の仮想マシン間、および、仮想マシンとデプロイデーモン間で帯域を相互に融通し合うことはしない。各仮想マシン120は、帯域管理テーブル151に記録した閾値151bで帯域が制御される。また、デプロイデーモン130は、帯域管理テーブル151に記録した各仮想マシンの余裕値151cの和で帯域が制御される。この際、図1に記載した帯域履歴テーブル152は不要になる。また、以下に述べる帯域相互融通処理等も行わない。従って、運用時に他の仮想マシンによって、一度割り当てられた帯域が変動することがない。
<Operation example with fixed bandwidth>
As an example of the embodiment, an operation example in a fixed band is shown. In operation with a fixed bandwidth, bandwidth is not interchanged between a plurality of virtual machines and between a virtual machine and a deployment daemon. The bandwidth of each virtual machine 120 is controlled by the threshold value 151 b recorded in the bandwidth management table 151. Further, the deployment daemon 130 controls the bandwidth by the sum of the margin values 151c of the virtual machines recorded in the bandwidth management table 151. At this time, the bandwidth history table 152 described in FIG. 1 is not necessary. In addition, the inter-band interchange processing described below is not performed. Therefore, the bandwidth once allocated by other virtual machines during operation does not fluctuate.
 <可動帯域での運用例>
 実施例の一例として、可動帯域での運用例を示す。可動帯域での運用においては、仮想マシン間、あるいは、仮想マシンとデプロイデーモンとの間で帯域を相互に融通し合う。各仮想マシン、および、デプロイデーモンは帯域を多く使用している場合には、帯域をあまり使用していない仮想マシンから帯域を融通してもらう。これにより、仮想マシン、および、デプロイデーモンはシステム全体の帯域を有効に活用することができる。
<Operation example in movable band>
As an example of the embodiment, an operation example in the movable band is shown. In operation in the movable band, the bandwidth is mutually interchanged between virtual machines or between a virtual machine and a deployment daemon. When each virtual machine and the deployment daemon use a large amount of bandwidth, the virtual machine that does not use the bandwidth very much has the bandwidth interchanged. Thereby, the virtual machine and the deployment daemon can effectively use the bandwidth of the entire system.
 可動帯域での運用には、以下に述べる帯域相互融通処理等を行う。 For the operation in the movable band, the following mutual bandwidth processing is performed.
 <帯域相互融通処理>
 図4を参照して、計算機システム10の物理マシン100における、実際の運用中に実行される帯域相互融通処理について説明する。
<Bandwidth interchange processing>
With reference to FIG. 4, a mutual bandwidth interchange process executed during actual operation in the physical machine 100 of the computer system 10 will be described.
 帯域相互融通処理は物理マシン100が稼働中は例えば1秒程度の周期で定期的に行われる。この帯域相互融通処理を運用中に繰り返し実行することによって、運用中に変化する各仮想マシンの帯域の需要、及び一時的な仮想マシンイメージファイルの転送に応じた適切な帯域の配分(仮想マシン間での帯域の融通)が随時行われる。 Band mutual interchange processing is performed periodically with a period of, for example, about 1 second while the physical machine 100 is in operation. By repeatedly executing this bandwidth interchange processing during operation, the bandwidth demand of each virtual machine that changes during operation and the appropriate bandwidth allocation according to the temporary transfer of virtual machine image files (between virtual machines) (Band interchange) at any time.
 帯域相互融通処理は、各仮想マシン120、121間、または、デプロイデーモン130と仮想マシン120、121間において、仮想マシンの帯域の使用量が少なく余裕がある場合は、他の仮想マシンに自身の帯域を融通する処理を行う。その結果、総量が限られている帯域を複数の仮想マシン間で有効活用でき、仮想マシンイメージファイルの転送の際は仮想マシンからデプロイデーモン130に、必要な帯域を一時的に割り当てることができる。 When the bandwidth usage of the virtual machine is small between the virtual machines 120 and 121, or between the deployment daemon 130 and the virtual machines 120 and 121 and there is a margin, the bandwidth interchange processing is performed by other virtual machines. Perform processing to accommodate the bandwidth. As a result, a bandwidth whose total amount is limited can be effectively utilized among a plurality of virtual machines, and a necessary bandwidth can be temporarily allocated from the virtual machine to the deployment daemon 130 when transferring the virtual machine image file.
 物理マシン100上の帯域管理機構150は図16の仮想マシンイメージ転送フラグテーブル154を参照し、当該仮想マシンにおいて仮想マシンイメージファイルが転送中かどうかを判断する(S401)。 The bandwidth management mechanism 150 on the physical machine 100 refers to the virtual machine image transfer flag table 154 in FIG. 16 and determines whether the virtual machine image file is being transferred in the virtual machine (S401).
 仮想マシンイメージを転送中の場合は、各仮想マシンについて、図6の仮想マシンイメージ転送中の帯域相互融通処理(S402)を行う。その後、帯域履歴記録処理(S404)を行う。 When the virtual machine image is being transferred, the mutual bandwidth interchange process (S402) during the virtual machine image transfer shown in FIG. 6 is performed for each virtual machine. Thereafter, a bandwidth history recording process (S404) is performed.
 帯域履歴記録処理では、物理マシン100の帯域管理機構150が、各仮想マシンおよびデプロイデーモンが使用している帯域を、図14の帯域履歴テーブル152に記録する。 In the bandwidth history recording process, the bandwidth management mechanism 150 of the physical machine 100 records the bandwidth used by each virtual machine and the deployment daemon in the bandwidth history table 152 of FIG.
 仮想マシンイメージを転送中でない場合は、各仮想マシンについて図5の仮想マシンイメージ非転送中の帯域相互融通処理(S403)を行う。その後、帯域履歴記録処理(S404)を行う。 If the virtual machine image is not being transferred, the bandwidth mutual interchange processing (S403) during non-transfer of the virtual machine image of FIG. 5 is performed for each virtual machine. Thereafter, a bandwidth history recording process (S404) is performed.
 <仮想マシンイメージ非転送中の帯域相互融通処理>
 図5を参照して、計算機システム10の物理マシン100における仮想マシンイメージ非転送中の帯域相互融通処理(S403)について説明する。
<Band interchange processing during non-transfer of virtual machine images>
With reference to FIG. 5, the mutual bandwidth interchange processing (S403) during the virtual machine image non-transfer in the physical machine 100 of the computer system 10 will be described.
 仮想マシンイメージ非転送中の帯域相互融通処理は、仮想マシンイメージを転送していない場合に行われる、仮想マシン間で相互に帯域を融通しあう処理である。その結果、仮想マシンイメージの転送を行うデプロイデーモンが利用する帯域も仮想マシン間で有効に利用できる。 The mutual bandwidth interchange processing during non-transfer of the virtual machine image is processing that is performed when the virtual machine image is not transferred, and the bandwidth is interchanged between the virtual machines. As a result, the bandwidth used by the deployment daemon that transfers virtual machine images can also be used effectively between virtual machines.
 仮想マシンイメージ非転送中の帯域相互融通処理は、仮想マシン毎に行われる。 帯 域 Mutual bandwidth interchange processing during non-transfer of virtual machine images is performed for each virtual machine.
 物理マシン100の帯域管理機構150は、仮想マシンが使用している帯域が閾値151bより少ないかどうかを、帯域履歴テーブル152を参照し判断する(S403-1)。 The bandwidth management mechanism 150 of the physical machine 100 determines whether or not the bandwidth used by the virtual machine is less than the threshold 151b with reference to the bandwidth history table 152 (S403-1).
 仮想マシンが使用している帯域が閾値151bより少ない場合は、仮想マシンイメージ非転送中の帯域相互融通処理を終了する。 If the bandwidth used by the virtual machine is less than the threshold value 151b, the bandwidth interchange processing during non-transfer of the virtual machine image is terminated.
 仮想マシンが使用している帯域が閾値151b以上の場合は、その仮想マシンの融通値151dが0以上であるかどうかを、帯域管理機構150が、帯域管理テーブル151を参照し、判断する(S403-2)。 When the bandwidth used by the virtual machine is equal to or greater than the threshold 151b, the bandwidth management mechanism 150 determines whether or not the accommodation value 151d of the virtual machine is 0 or more with reference to the bandwidth management table 151 (S403). -2).
 仮想マシンの融通値151dが0以上である場合は、図8の仮想マシン帯域確保処理(S403-4)を行い、処理を終了する。 When the virtual machine accommodation value 151d is 0 or more, the virtual machine bandwidth securing process (S403-4) in FIG. 8 is performed, and the process is terminated.
 仮想マシンの融通値151dが0以上でない場合は、図7の仮想マシン帯域返還処理(S403-3)を行い、処理を終了する。 If the virtual machine accommodation value 151d is not 0 or more, the virtual machine bandwidth return processing (S403-3) in FIG. 7 is performed, and the processing is terminated.
 <仮想マシン帯域返還処理>
 図7を参照して、計算機システム10の物理マシン100における仮想マシン帯域返還処理(S403-3)について説明する。
<Virtual machine bandwidth return processing>
With reference to FIG. 7, the virtual machine bandwidth return process (S403-3) in the physical machine 100 of the computer system 10 will be described.
 仮想マシン帯域返還処理は、仮想マシンが、他の仮想マシンやデプロイデーモンに融通していた帯域を自身に返還させる処理である。その結果、他に融通していた帯域を返還することで、処理に必要な帯域の急激な増加に対処できる。 The virtual machine bandwidth return process is a process in which the virtual machine returns itself the bandwidth that has been accommodated by other virtual machines or deployment daemons. As a result, it is possible to cope with a sudden increase in the bandwidth required for processing by returning the bandwidth that has been accommodated elsewhere.
 仮想マシン帯域返還処理において、帯域の返還先仮想マシンを融通元仮想マシンと呼称する。
帯域管理機構150は仮想マシンの融通値151dに自身の余裕値151cを加算して設定しなおす(S501)。
In the virtual machine bandwidth return process, the bandwidth return destination virtual machine is referred to as an interchange source virtual machine.
The bandwidth management mechanism 150 adds and sets its own margin value 151c to the virtual machine accommodation value 151d (S501).
 帯域管理機構150は、図15の帯域融通先テーブル153を参照し、融通元153aの仮想マシンを選択し、その融通元153bの仮想マシンの融通値151dから、融通元の仮想マシンの余裕値151cを減算して設定しなおす(S502)。 The bandwidth management mechanism 150 refers to the bandwidth accommodation destination table 153 of FIG. 15 to select a virtual machine of the accommodation source 153a, and from the accommodation value 151d of the virtual machine of the accommodation source 153b, the margin value 151c of the accommodation source virtual machine Is subtracted and set again (S502).
 帯域管理機構150は、各仮想マシンの帯域値151eを再計算し設定する(S503)。ここで、帯域値151eは、帯域管理テーブル151において、閾値151b+余裕値151c+融通値151dとして計算する。計算した帯域値は帯域管理テーブル151の帯域値151eに設定する。 The bandwidth management mechanism 150 recalculates and sets the bandwidth value 151e of each virtual machine (S503). Here, the bandwidth value 151e is calculated in the bandwidth management table 151 as the threshold value 151b + the margin value 151c + the accommodation value 151d. The calculated bandwidth value is set in the bandwidth value 151e of the bandwidth management table 151.
 帯域管理機構150が、帯域融通先テーブル153の融通元153aの仮想マシンの融通先153bをリセットする (S504)。 The bandwidth management mechanism 150 resets the accommodation destination 153b of the virtual machine of the accommodation source 153a of the bandwidth accommodation destination table 153 (S504).
 帯域管理機構150は、各仮想マシンの帯域を帯域管理テーブル151の帯域値151eに合わせて制御する(S505)。 The bandwidth management mechanism 150 controls the bandwidth of each virtual machine according to the bandwidth value 151e of the bandwidth management table 151 (S505).
 帯域管理機構150が、デプロイデーモンの帯域値151eを、デプロイ管理サーバ200のデプロイ帯域制御機構230に送信する (S506)。 The bandwidth management mechanism 150 transmits the bandwidth value 151e of the deployment daemon to the deployment bandwidth control mechanism 230 of the deployment management server 200 (S506).
 <仮想マシン帯域確保処理>
 図8を参照して、計算機システム10の物理マシン100における仮想マシン帯域確保処理(S403-4)について説明する。
<Virtual machine bandwidth reservation processing>
With reference to FIG. 8, the virtual machine bandwidth securing process (S403-4) in the physical machine 100 of the computer system 10 will be described.
 仮想マシン帯域確保処理は仮想マシンが必要とする帯域を、使用中帯域が閾値151bより少ない他の仮想マシンから融通してもらうことで、帯域を確保する処理である。その結果、処理に必要な帯域の急激な増加への対処が保証される。 The virtual machine bandwidth securing process is a process for securing the bandwidth by having the bandwidth required by the virtual machine interchanged from other virtual machines whose used bandwidth is less than the threshold 151b. As a result, it is guaranteed to cope with a sudden increase in bandwidth required for processing.
 仮想マシン帯域確保処理において、仮想マシン帯域確保処理を行う仮想マシンを帯域融通先の仮想マシンと呼称する。 In the virtual machine bandwidth securing process, a virtual machine that performs the virtual machine bandwidth securing process is referred to as a bandwidth accommodation destination virtual machine.
 帯域管理機構150は帯域履歴テーブル151および帯域融通先テーブル153を参照し、使用している帯域が閾値151bより少なく、かつ、他の仮想マシン120、121またはデプロイデーモン130に帯域を融通していない仮想マシンがあるかどうかを判断する(S601)。 The bandwidth management mechanism 150 refers to the bandwidth history table 151 and the bandwidth accommodation destination table 153. The bandwidth management mechanism 150 uses less bandwidth than the threshold 151b and does not allow bandwidth to other virtual machines 120 and 121 or the deployment daemon 130. It is determined whether there is a virtual machine (S601).
 使用している帯域が閾値151bより少なく、かつ、帯域を融通していない仮想マシンがない場合は、仮想マシン帯域確保処理を終了する。 If the used bandwidth is less than the threshold 151b and there is no virtual machine that does not accommodate the bandwidth, the virtual machine bandwidth securing process is terminated.
 使用している帯域が閾値151bより少なく、かつ、帯域を融通していない仮想マシンがある場合は、帯域管理機構150は使用している帯域が閾値151bより少なく、帯域を融通していない仮想マシンのうち、図14の帯域履歴テーブル152を参照して、使用している帯域が一番小さいものを選択する(S602)。ここで選択した仮想マシンを融通元仮想マシンと呼称する。 When there is a virtual machine that uses less bandwidth than the threshold 151b and does not accommodate the bandwidth, the bandwidth management mechanism 150 uses a virtual machine that uses less bandwidth than the threshold 151b and does not accommodate bandwidth. Among them, the bandwidth history table 152 in FIG. 14 is referred to, and the one having the smallest bandwidth is selected (S602). The virtual machine selected here is referred to as an accommodation source virtual machine.
 帯域管理機構150はステップS602で選択した融通元仮想マシンの融通値151dから、選択した融通元仮想マシンの融通値151dを減算して、帯域管理テーブル151に設定する(S603)。 The bandwidth management mechanism 150 subtracts the accommodation value 151d of the selected accommodation source virtual machine from the accommodation value 151d of the accommodation source virtual machine selected in step S602, and sets it in the bandwidth management table 151 (S603).
 帯域管理機構150が、帯域融通先の仮想マシンの融通値151dに、ステップS602で選択した融通元仮想マシンの余裕値151cを加算して設定する(S604)。 The bandwidth management mechanism 150 adds and sets the margin value 151c of the accommodation source virtual machine selected in step S602 to the accommodation value 151d of the bandwidth accommodation destination virtual machine (S604).
 帯域管理機構150が、帯域融通先テーブル153の、ステップS602の処理で選択した融通元仮想マシンの項目153aに、融通先仮想マシンを設定する(S605)。 The bandwidth management mechanism 150 sets the accommodation destination virtual machine in the item 153a of the accommodation source virtual machine selected in the process of step S602 in the bandwidth accommodation destination table 153 (S605).
 <仮想マシンイメージ転送中の帯域相互融通処理>
 図6を参照して、計算機システム10の物理マシン100における仮想マシンイメージ転送中の帯域相互融通処理(S402)について説明する。本処理は、仮想マシンの帯域に余裕があればデプロイデーモン130に帯域を融通する処理である。その結果、帯域に余裕がある仮想マシンの動作に影響を与えることなく、デプロイデーモン130は仮想マシンイメージファイルを転送することができる。
<Band interchange processing during virtual machine image transfer>
With reference to FIG. 6, the mutual bandwidth interchange process (S402) during virtual machine image transfer in the physical machine 100 of the computer system 10 will be described. This process is a process for accommodating the bandwidth to the deployment daemon 130 if the virtual machine has a sufficient bandwidth. As a result, the deployment daemon 130 can transfer the virtual machine image file without affecting the operation of the virtual machine having a sufficient bandwidth.
 物理マシン100の帯域管理機構150は、仮想マシンが使用している帯域が閾値151bより少ないかどうかを、帯域履歴テーブル152を参照し判断する(S402-1)。 The bandwidth management mechanism 150 of the physical machine 100 determines whether or not the bandwidth used by the virtual machine is less than the threshold 151b with reference to the bandwidth history table 152 (S402-1).
 仮想マシンが使用している帯域が閾値151bより少ない場合は、図9のデプロイデーモン130への帯域融通処理(S402-4)を行い、仮想マシンイメージ転送中の帯域相互融通処理を終了する。 If the bandwidth used by the virtual machine is less than the threshold 151b, the bandwidth interchange processing (S402-4) to the deployment daemon 130 of FIG. 9 is performed, and the bandwidth mutual interchange processing during the virtual machine image transfer is terminated.
 仮想マシンが使用している帯域が閾値151b以上の場合は、その仮想マシンの融通値151dが0以上であるかどうかを、帯域管理機構150が帯域管理テーブル151を参照し、判断する(S402-2)。 If the bandwidth used by the virtual machine is equal to or greater than the threshold 151b, the bandwidth management mechanism 150 determines whether or not the accommodation value 151d of the virtual machine is 0 or more with reference to the bandwidth management table 151 (S402-). 2).
 仮想マシンの融通値151dが0以上である場合は、仮想マシンイメージ転送中の帯域相互融通処理を終了する。 When the virtual machine accommodation value 151d is 0 or more, the bandwidth mutual accommodation processing during the virtual machine image transfer ends.
 仮想マシンの融通値151dが0以上でない場合は、図7の仮想マシン帯域返還処理(S403-3)を行い、処理を終了する。 If the virtual machine accommodation value 151d is not 0 or more, the virtual machine bandwidth return processing (S403-3) in FIG. 7 is performed, and the processing is terminated.
 <デプロイデーモンへの帯域融通処理>
 図9を参照して、計算機システム10の物理マシン100における仮想マシンイメージ転送中のデプロイデーモンへの帯域融通処理(S402-4)について説明する。
<Band interchange processing to deployment daemon>
With reference to FIG. 9, the bandwidth accommodation processing (S402-4) to the deployment daemon during the virtual machine image transfer in the physical machine 100 of the computer system 10 will be described.
 デプロイデーモン130への帯域融通処理は、仮想マシンの帯域使用量が少なく、余裕がある場合に、デプロイデーモン130に対して帯域を融通する処理である。その結果、余裕のある仮想マシンからデプロイデーモン130に帯域を融通するので、仮想マシンイメージファイルの転送のためにシステム全体の帯域を増やす必要がない。 The bandwidth accommodation processing to the deployment daemon 130 is processing for accommodating bandwidth to the deployment daemon 130 when the bandwidth usage of the virtual machine is small and there is room. As a result, since the bandwidth is accommodated from the spare virtual machine to the deployment daemon 130, it is not necessary to increase the bandwidth of the entire system for transferring the virtual machine image file.
 帯域管理機構150は、帯域管理テーブル151を参照し、仮想マシンの融通値151dに自身の余裕値151cを減算して設定しなおす(S701)。 The bandwidth management mechanism 150 refers to the bandwidth management table 151, and subtracts its own margin value 151c from the virtual machine accommodation value 151d (S701).
 帯域管理機構150は、デプロイデーモン130の融通値151dに仮想マシンの余裕値151cを加算して設定しなおす(S702)。 The bandwidth management mechanism 150 adds the margin value 151c of the virtual machine to the accommodation value 151d of the deployment daemon 130 and resets the setting (S702).
 帯域管理機構150は、帯域融通先テーブル153の仮想マシンの融通先153bにデプロイデーモンを設定する(S703)。 The bandwidth management mechanism 150 sets a deployment daemon in the virtual machine accommodation destination 153b of the bandwidth accommodation destination table 153 (S703).
 帯域管理機構150が、各仮想マシン120、121および、デプロイデーモン130の帯域を帯域管理テーブルの帯域値151eに合わせ、制御する(S704)。 The bandwidth management mechanism 150 controls the bandwidth of each of the virtual machines 120 and 121 and the deployment daemon 130 according to the bandwidth value 151e of the bandwidth management table (S704).
 帯域管理機構150が、デプロイデーモンの帯域値151eを、デプロイ管理サーバ200のデプロイ帯域制御機構230に送信する(S705)。 
 デプロイ管理サーバ200のデプロイ帯域制御機構230が帯域管理機構150より受信したデプロイデーモンの帯域値151eで、仮想マシンイメージファイルの転送帯域を制御する(S706)。
The bandwidth management mechanism 150 transmits the bandwidth value 151e of the deployment daemon to the deployment bandwidth control mechanism 230 of the deployment management server 200 (S705).
The deployment bandwidth control mechanism 230 of the deployment management server 200 controls the transfer bandwidth of the virtual machine image file with the bandwidth value 151e of the deployment daemon received from the bandwidth management mechanism 150 (S706).
 <デプロイ帯域制御機構処理>
 図10を参照して、計算機システム10のデプロイ管理サーバ200におけるデプロイ帯域制御機構処理230について説明する。
<Deployment bandwidth control mechanism processing>
The deployment bandwidth control mechanism process 230 in the deployment management server 200 of the computer system 10 will be described with reference to FIG.
 デプロイ帯域制御機構処理は、デプロイ管理サーバ200のデプロイ帯域制御機構230がデプロイデーモン130の帯域を制御するために行う処理である。物理マシン100内の仮想マシン間で帯域を融通し合った結果得られた帯域によってデプロイデーモン130の帯域を制御するので、仮想マシンの動作には影響を与えない。 The deployment bandwidth control mechanism process is a process performed by the deployment bandwidth control mechanism 230 of the deployment management server 200 to control the bandwidth of the deployment daemon 130. Since the bandwidth of the deployment daemon 130 is controlled by the bandwidth obtained as a result of the bandwidth interchange between the virtual machines in the physical machine 100, the operation of the virtual machine is not affected.
 物理マシンの帯域管理機構150から帯域値設定のためのパケットが届いているかを判断する(S801)。 It is determined whether a packet for setting a bandwidth value has arrived from the bandwidth management mechanism 150 of the physical machine (S801).
 パケットが届いていない場合、処理の開始時点に戻る。 If the packet has not arrived, return to the start of processing.
 パケットが届いている場合、デプロイ帯域制御機構230は図17のデプロイ帯域テーブル231に、物理マシン100の帯域管理機構150から受信したデプロイデーモン130の帯域値151eを記録する(S802)。 When the packet has arrived, the deployment bandwidth control mechanism 230 records the bandwidth value 151e of the deployment daemon 130 received from the bandwidth management mechanism 150 of the physical machine 100 in the deployment bandwidth table 231 of FIG. 17 (S802).
 デプロイ帯域制御機構230は、デプロイ帯域テーブル231のデプロイ帯域値231bで、デプロイアプリ250の帯域を制御する(S803)。 The deployment bandwidth control mechanism 230 controls the bandwidth of the deployment application 250 with the deployment bandwidth value 231b of the deployment bandwidth table 231 (S803).
 <デプロイアプリ処理>
 図11を参照して、計算機システム10のデプロイ管理サーバ200におけるデプロイアプリ処理250について説明する。
<Deploy application processing>
With reference to FIG. 11, the deployment application processing 250 in the deployment management server 200 of the computer system 10 will be described.
 デプロイアプリ処理は、デプロイ管理サーバ200のデプロイアプリ250が、仮想マシンファイルイメージ210を物理サーバ100に転送するための処理である。デプロイ管理サーバ200が種々の仮想マシンファイルイメージ210を一括管理して仮想マシンのそれぞれに提供することで、物理マシン100が種々の仮想マシンファイルイメージ210を用意する必要がなくなる。 The deploy application process is a process for the deploy application 250 of the deployment management server 200 to transfer the virtual machine file image 210 to the physical server 100. The deployment management server 200 collectively manages various virtual machine file images 210 and provides them to each of the virtual machines, thereby eliminating the need for the physical machine 100 to prepare various virtual machine file images 210.
 ユーザからデプロイ開始命令が発行されたかを判断する。(S901)。 Judge whether a deployment start command has been issued by the user. (S901).
 デプロイ開始命令が発行されていない場合は、処理の開始時点に戻る。 If the deployment start command has not been issued, return to the start of processing.
 デプロイ開始命令が発行された場合、デプロイアプリ250は、物理マシン100のデプロイデーモン130に対して、仮想マシンイメージファイル210の転送を開始する(S902)。 When the deployment start command is issued, the deployment application 250 starts transferring the virtual machine image file 210 to the deployment daemon 130 of the physical machine 100 (S902).
 仮想マシンイメージファイル210の転送が終了したかを判断する(S903)。 It is determined whether the transfer of the virtual machine image file 210 is completed (S903).
 転送が終了した場合は処理を終了する。 If the transfer ends, the process ends.
 転送が終了していない場合は、デプロイアプリ250は、仮想マシンイメージファイル210の残り転送量と、現在の使用帯域で除算して転送完了時間を計算し、ディスプレイ220に表示する(S904)。仮想マシンイメージファイルの転送判断処理に戻る。 If the transfer has not been completed, the deployment application 250 calculates the transfer completion time by dividing the remaining transfer amount of the virtual machine image file 210 by the current use bandwidth, and displays it on the display 220 (S904). Return to the virtual machine image file transfer determination process.
 <デプロイデーモン処理>
 図12を参照して、計算機システム10の物理マシン100におけるデプロイデーモン処理について説明する。
<Deploy daemon processing>
The deployment daemon process in the physical machine 100 of the computer system 10 will be described with reference to FIG.
 デプロイデーモン処理は物理マシンのデプロイデーモン130が、デプロイ管理サーバ200のデプロイアプリ250から仮想マシンイメージファイルを受信して、ディスク170に書きだす処理である。デプロイ管理サーバ200から送信された仮想マシンイメージファイルをディスク170に書きだすことにより、仮想マシン間で仮想マシンイメージファイルを共有でき、一度ある仮想マシンが利用した仮想マシンイメージを他の仮想マシンが再利用でき、デプロイ管理サーバ200と物理マシン100との間の仮想マシンイメージファイルの転送、及び転送に伴う帯域の融通処理を割愛できる。 The deployment daemon process is a process in which the physical machine deployment daemon 130 receives a virtual machine image file from the deployment application 250 of the deployment management server 200 and writes it to the disk 170. By writing the virtual machine image file transmitted from the deployment management server 200 to the disk 170, the virtual machine image file can be shared between the virtual machines, and the virtual machine image used by one virtual machine can be re-used by another virtual machine. It can be used, and the transfer of the virtual machine image file between the deployment management server 200 and the physical machine 100 and the bandwidth interchange processing associated with the transfer can be omitted.
 デプロイ管理サーバ200のデプロイアプリ250から仮想マシンイメージファイル210が送られてきているかを判断する(S1001)。 It is determined whether the virtual machine image file 210 is sent from the deployment application 250 of the deployment management server 200 (S1001).
 ファイルが送られていない場合、図16の転送フラグテーブル154のフラグ154bに「0」を記録する(S1002)。 If no file has been sent, “0” is recorded in the flag 154b of the transfer flag table 154 in FIG. 16 (S1002).
 ファイルが送られている場合、デプロイデーモン130は、受信した仮想マシンイメージファイル210をディスク170に書きだす(S1003)。 If the file has been sent, the deployment daemon 130 writes the received virtual machine image file 210 to the disk 170 (S1003).
 デプロイデーモン130は転送フラグテーブル154のフラグ154bに「1」を記録する(S1004)。 The deploy daemon 130 records “1” in the flag 154b of the transfer flag table 154 (S1004).
 <計算機システムの拡張性>
 以上、計算機システム10内の物理マシン100が1台の場合の例について説明したが、物理マシンは1台に限定されない。また、物理マシン100上で動作する仮想マシン120、121が2台の場合の例について説明したが、仮想マシンは2台に限定されない。物理マシン100とデプロイ管理サーバ200を結ぶネットワーク260を1台の場合と同様に適切に構成することにより、2台以上の物理マシン100から構成される計算機システムに拡張可能である。
<Extensibility of computer system>
The example in the case where there is one physical machine 100 in the computer system 10 has been described above, but the number of physical machines is not limited to one. In addition, although an example in which there are two virtual machines 120 and 121 operating on the physical machine 100 has been described, the number of virtual machines is not limited to two. By appropriately configuring the network 260 that connects the physical machine 100 and the deployment management server 200 as in the case of a single machine, the network system can be expanded to a computer system including two or more physical machines 100.
 本実施例では、デプロイ管理サーバ200から仮想マシンイメージファイルを転送していたが、DVDなどの外部メディアからデプロイデーモン130に仮想マシンイメージファイルを転送することもできる。 In this embodiment, the virtual machine image file is transferred from the deployment management server 200, but the virtual machine image file can also be transferred from an external medium such as a DVD to the deployment daemon 130.
 また、本実施例においては、仮想マシンのデプロイに特記して、物理マシンや既存仮想マシンへの干渉を防ぐ方法について説明したが、デプロイに限らず、仮想マシンのマイグレーションやバックアップについても、本実施例の形態を適用することができる。 In this example, the method for preventing interference with physical machines and existing virtual machines has been described with special reference to virtual machine deployment. However, not only for deployment, but also for migration and backup of virtual machines. Example forms can be applied.
 以上、一実施形態について説明したが、本発明は上記実施形態に限定されるものではなく、その要旨を逸脱しない範囲で種々変更可能である。 As mentioned above, although one embodiment has been described, the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the scope of the invention.
10:計算機システム、100:物理マシン、110:ハイパーバイザ、120:仮想マシン1、121:仮想マシン2、130:デプロイデーモン、140:スケジュール機構、150:帯域管理機構、151:帯域管理テーブル、152:帯域履歴テーブル、153:帯域融通テーブル、160、240:通信IF、170:ディスク、200:デプロイ管理サーバ、210:仮想マシンイメージファイル、220:ディスプレイ、230:デプロイ帯域制御機構、260:ネットワーク 10: computer system, 100: physical machine, 110: hypervisor, 120: virtual machine 1, 121: virtual machine 2, 130: deployment daemon, 140: schedule mechanism, 150: bandwidth management mechanism, 151: bandwidth management table, 152 : Bandwidth history table, 153: bandwidth accommodation table, 160, 240: communication IF, 170: disk, 200: deployment management server, 210: virtual machine image file, 220: display, 230: deployment bandwidth control mechanism, 260: network

Claims (15)

  1.  複数の仮想マシンを動作させる物理マシンと、前記物理マシンで前記仮想マシンを実現するための制御を行うデプロイ管理サーバとをネットワークを介して接続した計算機システムにおける仮想マシンデプロイ方法であって、前記物理マシンは、
     前記複数の仮想マシンのそれぞれに対して、前記仮想マシンの基本的な動作に必要な帯域の値である閾値を含む、帯域に関する値の初期値を設定し、
     一定時間間隔で、前記仮想マシンが使用する帯域と前記閾値との大小関係に基づいて、前記仮想マシンと、前記仮想マシンを前記物理マシンで実現するための仮想マシンイメージファイルを前記デプロイ管理サーバから受け取るデプロイデーモンとの間で帯域を融通する帯域相互融通処理を行う、
     ことを特徴とする仮想マシンデプロイ方法。
    A virtual machine deployment method in a computer system in which a physical machine that operates a plurality of virtual machines and a deployment management server that controls the physical machine to realize the virtual machine are connected via a network. Machine
    For each of the plurality of virtual machines, an initial value of a band-related value including a threshold value that is a band value required for basic operation of the virtual machine is set.
    Based on the magnitude relationship between the bandwidth used by the virtual machine and the threshold at regular time intervals, the virtual machine and a virtual machine image file for realizing the virtual machine on the physical machine are obtained from the deployment management server. Perform mutual bandwidth interchange processing to allow bandwidth interchange with the receiving deployment daemon.
    A virtual machine deployment method characterized by that.
  2.  前記帯域相互融通処理は、
     前記仮想マシンイメージファイルが非転送中の場合、前記仮想マシン間で相互に帯域を融通し合う第一の帯域相互融通処理を行い、
     前記仮想マシンイメージファイルが転送中の場合、前記仮想マシンの帯域に余裕があれば前記デプロイデーモンに帯域を融通する第二の帯域相互融通処理を行い、
     前記第一あるいは第二の帯域相互融通処理の後で、前記複数の仮想マシンのそれぞれ、及び前記デプロイデーモンが使用している帯域を記録する、
     ことを特徴とする請求項1記載の仮想マシンデプロイ方法。
    The band mutual accommodation processing is:
    When the virtual machine image file is not being transferred, a first bandwidth mutual interchange process is performed in which bandwidths are mutually interchanged between the virtual machines,
    If the virtual machine image file is being transferred, if there is a margin in the virtual machine bandwidth, perform a second bandwidth mutual accommodation process to accommodate the bandwidth to the deployment daemon,
    Record the bandwidth used by each of the plurality of virtual machines and the deployment daemon after the first or second bandwidth interchange processing.
    The virtual machine deployment method according to claim 1.
  3.  前記第一の帯域相互融通処理は、
     前記仮想マシンイメージファイルが非転送中の場合、前記帯域に関する値に含まれる、他の仮想マシンから融通された帯域の値である融通値が0以上でなければ、
     他の仮想マシンに融通していた帯域を自身の仮想マシンに返還させる仮想マシン帯域返還処理を行い、
     前記融通値が0以上であれば、
     前記閾値に対するマージンである余裕値に基づいて、余裕のある他の仮想マシンから帯域を融通してもらう仮想マシン帯域確保処理を行う、
     ことを特徴とする請求項2記載の仮想マシンデプロイ方法。
    The first inter-band interchange process is:
    When the virtual machine image file is not being transferred, an accommodation value included in the value related to the bandwidth, which is a bandwidth value accommodated from another virtual machine, is not 0 or more,
    Perform the virtual machine bandwidth return processing to return the bandwidth that has been accommodated to other virtual machines to your virtual machine,
    If the accommodation value is 0 or more,
    Based on a margin value that is a margin with respect to the threshold value, a virtual machine bandwidth securing process is performed in which the bandwidth is accommodated from another virtual machine having a margin,
    The virtual machine deployment method according to claim 2.
  4.  前記帯域確保処理は、使用している帯域が前記閾値より少なく、帯域を他に融通していない仮想マシンから受け取った前記余裕値を当該仮想マシンの前記融通値に加算することを特徴とする請求項3記載の仮想マシンデプロイ方法。 The bandwidth securing process is characterized in that the margin value received from a virtual machine that uses less bandwidth than the threshold and does not accommodate any other bandwidth is added to the accommodation value of the virtual machine. Item 4. The virtual machine deployment method according to Item 3.
  5.  前記第二の帯域相互融通処理は、
     前記仮想マシンの使用帯域が前記閾値よりも大きければ、
     他の仮想マシンに融通していた帯域を自身の仮想マシンに返還させる仮想マシン帯域返還処理を行い、
     前記仮想マシンの使用帯域が前記閾値よりも大きくなければ、
     前記仮想イメージファイルの転送に必要な帯域を前記デプロイデーモンに融通する帯域融通処理を行う、
     ことを特徴とする請求項2記載の仮想マシンデプロイ方法。
    The second interband interchange process is:
    If the bandwidth used by the virtual machine is greater than the threshold,
    Perform the virtual machine bandwidth return processing to return the bandwidth that has been accommodated to other virtual machines to your virtual machine,
    If the bandwidth used by the virtual machine is not greater than the threshold,
    Performing bandwidth accommodation processing for accommodating the bandwidth necessary for transferring the virtual image file to the deployment daemon;
    The virtual machine deployment method according to claim 2.
  6.  前記帯域融通処理は、仮想マシンの前記融通値から減算した前記余裕値を前記デプロイデーモンの前記融通値に加算して再設定し、 
     前記閾値、前記余裕値、及び前記融通値の合計である帯域値を前記デプロイ管理サーバに送信し、
     前記帯域値を、前記仮想マシン及び前記デプロイデーモンの帯域管理テーブルに設定する、
    ことを特徴とする請求項5記載の仮想マシンデプロイ方法。
    In the bandwidth accommodation process, the margin value subtracted from the accommodation value of the virtual machine is added to the accommodation value of the deployment daemon and reset,
    Sending the bandwidth value that is the sum of the threshold value, the margin value, and the accommodation value to the deployment management server,
    Setting the bandwidth value in a bandwidth management table of the virtual machine and the deployment daemon;
    The virtual machine deployment method according to claim 5.
  7.  前記仮想マシン帯域返還処理は、当該仮想マシンの前記融通値に前記余裕値を加算し、
     選択された前記仮想マシンまたは前記デプロイデーモンの前記融通値から前記当該仮想マシンの前記余裕値を減算することを特徴とする請求項3及び請求項5に記載の仮想マシンデプロイ方法。
    The virtual machine bandwidth return process adds the margin value to the accommodation value of the virtual machine,
    6. The virtual machine deployment method according to claim 3, wherein the margin value of the virtual machine is subtracted from the flexibility value of the selected virtual machine or the deployment daemon.
  8.  複数の仮想マシンを動作させる物理マシンと、
     前記物理マシンで前記仮想マシンを実現するための制御を行うデプロイ管理サーバと、
     前記物理マシンと前記デプロイ管理サーバと接続するネットワークを有し、
     前記物理マシンは、
     前記複数の仮想マシンのそれぞれに対して、前記仮想マシンの基本的な動作に必要な帯域の値である閾値を含む、帯域に関する値の初期値を設定する初期化処理手段、及び
     一定時間間隔で、前記仮想マシンが使用する帯域と前記閾値との大小関係に基づいて、前記仮想マシンと、前記仮想マシンを前記物理マシンで実現するための仮想マシンイメージファイルを前記デプロイ管理サーバから受け取るデプロイデーモンとの間で帯域を融通する帯域相互融通手段を有する、
     ことを特徴とする計算機システム。
    A physical machine that runs multiple virtual machines;
    A deployment management server that performs control for realizing the virtual machine on the physical machine;
    A network connecting the physical machine and the deployment management server;
    The physical machine is
    An initialization processing means for setting an initial value of a bandwidth-related value, including a threshold value that is a bandwidth value necessary for basic operation of the virtual machine, for each of the plurality of virtual machines; A deployment daemon that receives, from the deployment management server, a virtual machine image file for realizing the virtual machine on the physical machine based on a magnitude relationship between a bandwidth used by the virtual machine and the threshold value. Having bandwidth mutual accommodation means for accommodating bandwidth between
    A computer system characterized by that.
  9.  前記帯域相互融通手段は、
     前記仮想マシンイメージファイルが非転送中の場合、前記仮想マシン間で相互に帯域を融通し合う第一の帯域相互融通処理を行い、
     前記仮想マシンイメージファイルが転送中の場合、前記仮想マシンの帯域に余裕があれば前記デプロイデーモンに帯域を融通する第二の帯域相互融通処理を行い、
     前記第一あるいは第二の帯域相互融通処理の後で、前記複数の仮想マシンのそれぞれ、及び前記デプロイデーモンが使用している帯域を記録する、
     ことを特徴とする請求項8記載の計算機システム。
    The band mutual interchange means is:
    When the virtual machine image file is not being transferred, a first bandwidth mutual interchange process is performed in which bandwidths are mutually interchanged between the virtual machines,
    If the virtual machine image file is being transferred, if there is a margin in the virtual machine bandwidth, perform a second bandwidth mutual accommodation process to accommodate the bandwidth to the deployment daemon,
    Record the bandwidth used by each of the plurality of virtual machines and the deployment daemon after the first or second bandwidth interchange processing.
    The computer system according to claim 8.
  10.  前記第一の帯域相互融通処理は、
     前記仮想マシンイメージファイルが非転送中の場合、前記帯域に関する値に含まれる、他の仮想マシンから融通された帯域の値である融通値が0以上でなければ、
     他の仮想マシンに融通していた帯域を自身の仮想マシンに返還させる仮想マシン帯域返還処理を行い、
     前記融通値が0以上であれば、
     前記閾値に対するマージンである余裕値に基づいて、余裕のある他の仮想マシンから帯域を融通してもらう仮想マシン帯域確保処理を行う、
     ことを特徴とする請求項9記載の計算機システム。
    The first inter-band interchange process is:
    When the virtual machine image file is not being transferred, an accommodation value included in the value related to the bandwidth, which is a bandwidth value accommodated from another virtual machine, is not 0 or more,
    Perform the virtual machine bandwidth return processing to return the bandwidth that has been accommodated to other virtual machines to your virtual machine,
    If the accommodation value is 0 or more,
    Based on a margin value that is a margin with respect to the threshold value, a virtual machine bandwidth securing process is performed in which the bandwidth is accommodated from another virtual machine having a margin,
    The computer system according to claim 9.
  11.  前記第二の帯域相互融通処理は、
     前記仮想マシンの使用帯域が前記閾値よりも大きければ、
     他の仮想マシンに融通していた帯域を自身の仮想マシンに返還させる仮想マシン帯域返還処理を行い、
     前記仮想マシンの使用帯域が前記閾値よりも大きくなければ、
     前記仮想イメージファイルの転送に必要な帯域を前記デプロイデーモンに融通する帯域融通処理を行う、
     ことを特徴とする請求項9記載の計算機システム。
    The second interband interchange process is:
    If the bandwidth used by the virtual machine is greater than the threshold,
    Perform the virtual machine bandwidth return processing to return the bandwidth that has been accommodated to other virtual machines to your virtual machine,
    If the bandwidth used by the virtual machine is not greater than the threshold,
    Performing bandwidth accommodation processing for accommodating the bandwidth necessary for transferring the virtual image file to the deployment daemon;
    The computer system according to claim 9.
  12.  前記帯域融通処理は、仮想マシンの前記融通値から減算した前記余裕値を前記デプロイデーモンの前記融通値に加算して再設定し、 
     前記閾値、前記余裕値、及び前記融通値の合計である帯域値を前記デプロイ管理サーバに送信し、
     前記帯域値を、前記仮想マシン及び前記デプロイデーモンの帯域管理テーブルに設定する、
    ことを特徴とする請求項11記載の計算機システム。
    In the bandwidth accommodation process, the margin value subtracted from the accommodation value of the virtual machine is added to the accommodation value of the deployment daemon and reset,
    Sending the bandwidth value that is the sum of the threshold value, the margin value, and the accommodation value to the deployment management server,
    Setting the bandwidth value in a bandwidth management table of the virtual machine and the deployment daemon;
    The computer system according to claim 11, wherein:
  13.  複数の仮想マシンを動作させる物理マシンと、前記物理マシンで前記仮想マシンを実現するための制御を行うデプロイ管理サーバとを、ネットワークを介して接続した計算機システムにおける仮想マシンデプロイ方法であって、前記物理マシンは、
     前記複数の仮想マシンのそれぞれに対して、前記仮想マシンの基本的な動作に必要な帯域の値である閾値を含む、帯域に関する値の初期値を設定し、
     前記設定した帯域に関する値の初期値に基づいて、前記複数の仮想マシンのそれぞれの帯域、及び前記デプロイ管理サーバから仮想マシンイメージファイルを受け取るデプロイデーモンの帯域を制御することを特徴とする仮想マシンデプロイ方法。
    A virtual machine deployment method in a computer system in which a physical machine that operates a plurality of virtual machines and a deployment management server that performs control for realizing the virtual machine by the physical machine are connected via a network, The physical machine
    For each of the plurality of virtual machines, an initial value of a band-related value including a threshold value that is a band value required for basic operation of the virtual machine is set.
    A virtual machine deployment characterized by controlling a bandwidth of each of the plurality of virtual machines and a bandwidth of a deployment daemon that receives a virtual machine image file from the deployment management server based on an initial value of the set bandwidth value. Method.
  14.  前記初期値の設定の際に、
     前記複数の仮想マシンのそれぞれに帯域を割り当て、
     前記デプロイデーモンに帯域を割り当て、
     前記複数の仮想マシンにそれぞれの要件に応じた優先度を設定し、前記デプロイデーモンに最低の優先度を設定し、
     前記割り当てた帯域を前記複数の仮想マシンにそれぞれに設定し、
     前記デプロイデーモンに割り当てた帯域値を前記デプロイ管理サーバに送信する、
    ことを特徴とする請求項13記載の仮想マシンデプロイ方法。
    When setting the initial value,
    A bandwidth is allocated to each of the plurality of virtual machines;
    Allocate bandwidth to the deployment daemon,
    Set priorities according to the requirements of the plurality of virtual machines, set the lowest priority to the deployment daemon,
    Set the allocated bandwidth for each of the plurality of virtual machines,
    Sending the bandwidth value assigned to the deployment daemon to the deployment management server;
    The virtual machine deployment method according to claim 13.
  15.  前記複数の物理マシンのそれぞれに帯域を割り当てる際に、
     前記仮想マシンの基本的な動作に必要な帯域の値である閾値、前記閾値に対するマージンである余裕値、及び他の仮想マシンから融通された帯域の値である融通値の合計である帯域値を、前記複数の仮想マシンのそれぞれに設定することを特徴とする請求項14記載の仮想マシンデプロイ方法。
    When allocating bandwidth to each of the plurality of physical machines,
    A bandwidth value that is a sum of a threshold value that is a bandwidth value necessary for basic operation of the virtual machine, a margin value that is a margin for the threshold value, and an accommodation value that is a bandwidth value accommodated from other virtual machines. The virtual machine deployment method according to claim 14, wherein the virtual machine deployment method is set for each of the plurality of virtual machines.
PCT/JP2012/081846 2012-12-07 2012-12-07 Virtual machine deployment method and computer system WO2014087545A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2012/081846 WO2014087545A1 (en) 2012-12-07 2012-12-07 Virtual machine deployment method and computer system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2012/081846 WO2014087545A1 (en) 2012-12-07 2012-12-07 Virtual machine deployment method and computer system

Publications (1)

Publication Number Publication Date
WO2014087545A1 true WO2014087545A1 (en) 2014-06-12

Family

ID=50882989

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/081846 WO2014087545A1 (en) 2012-12-07 2012-12-07 Virtual machine deployment method and computer system

Country Status (1)

Country Link
WO (1) WO2014087545A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003157177A (en) * 2001-11-22 2003-05-30 Hitachi Ltd Virtual computer system and program
JP2005115653A (en) * 2003-10-08 2005-04-28 Ntt Data Corp Virtual machine management device and program
JP2005191688A (en) * 2003-12-24 2005-07-14 Hitachi Software Eng Co Ltd Communication band lending and borrowing system
JP2006166297A (en) * 2004-12-10 2006-06-22 Fumio Kobayashi Communication band control apparatus, communication band control method and program
JP2007316724A (en) * 2006-05-23 2007-12-06 Hitachi Ltd Computer system, management computer, and program distribution method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003157177A (en) * 2001-11-22 2003-05-30 Hitachi Ltd Virtual computer system and program
JP2005115653A (en) * 2003-10-08 2005-04-28 Ntt Data Corp Virtual machine management device and program
JP2005191688A (en) * 2003-12-24 2005-07-14 Hitachi Software Eng Co Ltd Communication band lending and borrowing system
JP2006166297A (en) * 2004-12-10 2006-06-22 Fumio Kobayashi Communication band control apparatus, communication band control method and program
JP2007316724A (en) * 2006-05-23 2007-12-06 Hitachi Ltd Computer system, management computer, and program distribution method

Similar Documents

Publication Publication Date Title
US10572290B2 (en) Method and apparatus for allocating a physical resource to a virtual machine
EP3270289B1 (en) Container-based multi-tenant computing infrastructure
KR102281739B1 (en) Resource scheduling method, scheduling server, cloud computing system, and storage medium
JP6190969B2 (en) Multi-tenant resource arbitration method
EP3553657A1 (en) Method and device for allocating distributed system task
US9069465B2 (en) Computer system, management method of computer resource and program
US20130339956A1 (en) Computer system and optimal arrangement method of virtual machine in computer system
US11755369B2 (en) Techniques for container scheduling in a virtual environment
WO2017179537A1 (en) Software update control device, software update control system, software update control method, and recording medium having software update control program stored thereon
WO2019056771A1 (en) Distributed storage system upgrade management method and device, and distributed storage system
JPWO2015087442A1 (en) Storage system migration method and migration method
CN113382077B (en) Micro-service scheduling method, micro-service scheduling device, computer equipment and storage medium
CN109992373B (en) Resource scheduling method, information management method and device and task deployment system
JP6010975B2 (en) Job management apparatus, job management method, and program
JP2012181578A (en) Update control device and program
CN114489517B (en) Offline acceleration migration method, system, equipment and storage medium
CN116881012A (en) Container application vertical capacity expansion method, device, equipment and readable storage medium
WO2014087545A1 (en) Virtual machine deployment method and computer system
JP6191361B2 (en) Information processing system, information processing system control method, and control program
WO2020022018A1 (en) Resource allocation device, resource management system, and resource allocation program
JP6836536B2 (en) Storage system and IO processing control method
WO2018173300A1 (en) I/o control method and i/o control system
JP2016062535A (en) Information processing system, control method thereof, and control program
US20240061698A1 (en) Managing the assignment of virtual machines to non-uniform memory access nodes
JP6235973B2 (en) server

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12889684

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12889684

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP