WO2016110950A1 - 計算機システム、管理システム、及び、リソース管理方法 - Google Patents
計算機システム、管理システム、及び、リソース管理方法 Download PDFInfo
- Publication number
- WO2016110950A1 WO2016110950A1 PCT/JP2015/050236 JP2015050236W WO2016110950A1 WO 2016110950 A1 WO2016110950 A1 WO 2016110950A1 JP 2015050236 W JP2015050236 W JP 2015050236W WO 2016110950 A1 WO2016110950 A1 WO 2016110950A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- instance
- physical
- resource
- computer
- information
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
Definitions
- the present invention generally relates to a technology of a computer system, a management system, and a resource management method.
- IaaS Intelligent as a Service
- a predetermined system environment for example, a WEB server system, a file server system, and a DB system.
- a user can construct a system environment without constructing a physical computer or the like by using IaaS through a terminal connected to the Internet.
- An IaaS provider operates a computer system including a plurality of physical computers, network devices, and storage devices, and provides IaaS.
- the system environment is constructed on this computer system.
- a method for constructing a system environment on a computer system a bare metal method, a virtualization method, and an LPAR (Logical Partition) method are known.
- the bare metal method is a method in which all physical resources of a physical device (CPU (Central Processing Unit) and memory) of a physical computer are allocated to one system environment.
- CPU Central Processing Unit
- memory memory
- the virtualization method is a method of virtualizing physical resources of physical devices of a physical computer (constructing virtual devices) and allocating them to a plurality of system environments.
- the LPAR method is a method of dividing physical resources of a physical device included in a physical computer and allocating a part of the divided physical resources only to one system environment among a plurality of system environments.
- Patent Document 1 describes this LPAR technology.
- the utilization rate of physical resources in the physical computer is improved.
- the physical resources of one physical computer are shared by a plurality of system environments, if a certain system environment in a plurality of system environments places a high load on the physical resources, the performance of the other system environment is reduced. May be reduced.
- the system environment is constructed by the bare metal method, such a situation does not occur, but the utilization rate of physical resources in the physical computer decreases.
- the problem is to provide a computer system, a management system, and a resource management method for constructing a system environment so that the utilization rate of physical resources in the physical computer is increased.
- the computer system includes a plurality of computers and a management system that manages the plurality of computers.
- a management system manages an instance realized by allocating a physical resource, which is a physical resource, to multiple computers, and includes information for configuring the instance.
- an instance implementation method corresponding to the instance configuration information is selected from a plurality of different instance implementation methods.
- the plurality of computers provide physical resources to the instance based on the selected instance realization method.
- a system environment can be constructed so that the utilization rate of physical resources in a physical computer is high.
- the structural example of the computer system which concerns on a present Example is shown.
- the example of the physical device with which a physical computer is provided is shown.
- a configuration example of a bare metal host, a virtual machine host, and an LPAR host is shown.
- a configuration example in which a plurality of LPAR hosts are connected and used is shown.
- 2 shows a configuration example of a storage apparatus.
- the structural example of the function of a management computer is shown.
- the structural example of an instance information table is shown.
- the structural example of a resource information table is shown.
- the structural example of a template information table is shown.
- the structural example of a topology information table is shown.
- the structural example of a candidate table is shown.
- the modification of the structure of a candidate table is shown. It is a flowchart which shows the example of the creation process of an instance.
- the structural example of a template utilization frequency table is shown. An example of processing for creating derived template information is shown.
- the processing unit (function) may be described using the expression “kkk unit”.
- the processing unit is executed by a computer program being executed by a processor (for example, a CPU (Central Processing Unit). It may be realized by a hardware circuit (for example, an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit)), which is determined when a processing unit is realized by a processor. Since the processing is appropriately performed using storage resources (for example, memory) and / or communication interface devices (for example, communication ports), the processing unit has few processors.
- the processing described with the processing unit as the subject may be processing performed by a processor or an apparatus having the processor, and the processor is a hardware circuit that performs part or all of the processing.
- the program may be installed in the processor from a program source, which may be, for example, a program distribution computer or a storage medium that can be read by the computer.
- a program source which may be, for example, a program distribution computer or a storage medium that can be read by the computer.
- a plurality of processing units may be combined into one processing unit, or one processing unit may be divided into a plurality of processing units.
- the management system may be composed of one or more computers.
- the management computer displays information (specifically, the management computer displays information on its own display device, or the management computer transmits display information to a remote display computer)
- the management computer is the management system.
- the plurality of computers may include a display computer when the display computer performs display
- the management system is the management system.
- the management computer also includes an interface device connected to one or more I / O devices including a display device, a storage resource (for example, a memory) that stores information, and a processor connected to the interface device and the storage resource.
- the display device may be a display device included in the management computer or a display computer connected to the management computer.
- the I / O device may be an I / O device (for example, a keyboard and a pointing device or a touch panel) included in the management computer, a display computer connected to the management computer, or another computer. That the management computer “displays the display information” is to display the display information on the display device, and this may be to display the display information on the display device of the management computer.
- the management computer may transmit display information to the display computer (in the latter case, the display information is displayed by the display computer).
- the management computer inputting / outputting information may be inputting / outputting information to / from an I / O device of the management computer, or a remote computer connected to the management computer (for example, a display)
- Information may be input / output to / from the computer.
- the information output may be a display of information.
- FIG. 1 shows a configuration example of a computer system 1 according to the present embodiment.
- the computer system 1 includes a plurality of physical computers 10, a storage device 100, and a management computer 200 that manages them.
- Each device 10, 100, 200 is connected by a predetermined network 55, 65, 252, and can transmit and receive data to and from each other.
- Each physical computer 10, management computer 200, and storage device 100 may be connected to a SAN (Storage Area Network) 55 through an HBA (Host Bus Adapter).
- the SAN 55 may be an FC SAN including an FC (FibreChannel) switch 50, an HBA 51, and the like.
- the HBA may be connected to another HBA or the FC switch 50 through a port.
- the SAN 55 may be configured other than the fiber channel.
- the SAN 55 may be configured by iSCSI, FCoE, Infini-band, or the like.
- Each physical computer 10 and management computer 200 may be connected to a predetermined network through NIC (Network Interface Card).
- the network may be configured by an Ethernet (registered trademark) switch 60 and a cable.
- Each physical computer 10 may transmit and receive application data, control information, and the like through a network.
- Protocols such as TCP / IP (Transmission Control Protocol / Internet Protocol) may be used for communication on the network.
- the network is divided into a LAN 65, a service network 250, and management networks 251 and 252.
- the service client 71 of the client terminal 70 and the application 12 implemented on the physical computer 10 may transmit and receive data necessary for the information processing service through the service network 250.
- the control components of each device may transmit and receive control data for managing the configuration of each device through the management networks 251 and 252.
- the plurality of physical computers 10 and management computers 200 configured in the data center may transmit and receive various data via the LAN 65.
- These network configurations may be physically separated, or may be logically separated by layer 3 switch settings or layer 2 switch settings (VLAN (Virtual LAN), virtual local area network).
- VLAN Virtual LAN
- the client terminal 70 which is a kind of computer can send and receive data to and from the computer system 1 through the service network 250 and / or the management network 251.
- the user can request the computer system 1 to configure the system environment through the client terminal 70.
- a system environment configured on the computer system 1 is referred to as an instance 11.
- the instance 11 may be configured by the guest OS 13 and the application 12 or may be configured only by the guest OS 13.
- Examples of the application 12 include a WEB server, a file server and a database server, or a combination thereof.
- the instance 11 is realized on the physical computer 10 constituting the computer system 1.
- a method for realizing the instance 11 there is a first method for allocating a physical resource provided by one physical computer 10 to a plurality of instances in a sharable manner.
- the instance 11 implementation method includes a second implementation method in which all or a part of physical resources provided by one computer is allocated to one instance in an occupying manner.
- An example of the first implementation method is a virtualization method.
- Examples of the second implementation method include a bare metal method and an LPAR method.
- the physical computer 10 configured with the bare metal method is referred to as a bare metal host 10a
- the physical computer 10 configured with the virtualization method is referred to as a virtual machine host 10b
- the physical computer 10 configured with the LPAR method is referred to as an LPAR host 10c. Details of these will be described later (see FIG. 3).
- the client terminal 70 may have a service client 71 and a management client 72 as functions.
- the service client 71 receives a service related to information processing from the application 12 on the instance 11.
- the management client 72 is connected to the management computer 200 and manages the configuration of the computer system 1.
- These service client 71 and management client 72 may be dedicated programs or general-purpose programs executed by a WEB browser or the like.
- the service client 71 and the management client 72 may be arranged together in one client terminal 70, or may be arranged separately in a plurality of client terminals 70.
- the storage apparatus 100 provides each instance 11 with a data storage area.
- the storage apparatus 100 includes a CPU, a memory, a storage device, and the like, and a program for controlling the storage device may be executed. Examples of storage devices include HDD (Hard Disk Drive) and SSD (Solid State Drive).
- the storage apparatus 100 may configure a volume 101 that is a logical storage area from a physical storage area of the storage device, and provide the volume 101 to each instance 11.
- the guest OS 13 related to the bare metal host 10 a may execute data read and write on the volume 101.
- the virtualization mechanism 24 may divide the volume 101 to configure a plurality of virtual volumes. Then, the guest OS 13b related to the virtual machine host 10b may read and write data to the virtual volume assigned to itself.
- the management computer 200 manages instances realized by allocating physical resources, which are physical resources, from a plurality of physical computers. Further, the management computer 200 is information including settings for configuring an instance, and is based on the instance configuration information including setting of necessity of occupying the physical resource of the instance. Then, an instance realization method corresponding to the instance configuration information is selected.
- the instance configuration information may be any of instance configuration requirements described later and instance information stored in the instance information table.
- the management computer 200 manages the physical computer 10, the storage device 100, the network devices 50, 60, etc. that make up the computer system 1.
- the management computer 200 has functions related to the application management unit 201, the integrated resource management unit 202, the computer management unit 203, the network management unit 204, and the storage management unit 205. These functions 201 to 205 may be collectively referred to as a management function.
- the application management unit 201 provides an I / F for managing the application 12 on the instance 11 from the client terminal 70.
- the computer management unit 203 manages the configuration of the physical computer 10.
- the network management unit 204 manages the configuration of the network device (for example, the FC switch 50 and the network adapter).
- the storage management unit 205 manages the configuration of the storage apparatus 100.
- the integrated resource management unit 202 has a function of simplifying complex resource management across these elements 201 to 205. The user can easily perform complex resource management through the integrated resource management unit 202.
- FIG. 2 shows an example of a physical device included in the physical computer 10.
- the physical computer 10 includes a CPU 15, a memory 18, an I / O (Input / Output) controller 31, a graphic controller 32, a disk controller 33, a USB (Universal Serial Bus) I / F 19, a VGA (Video Graphics Array) I as physical devices.
- / F20, SAS (Serial Attached SCSI) I / F 154, storage disk 105, NIC 34, HBA 35, and PCIe (Peripheral Component Interconnect Express) switch 17 may be provided.
- the CPU 15 and the memory 18 may be connected to the I / O controller 31, the graphic controller 32, and the disk controller 33 through the data bus 16. Further, the CPU 15 and the memory 18 may be connected to the NIC 34 and the HBA 35 through the data bus 16 and the PCIe switch 17.
- Each controller 31, 32, 33 controls the physical device of the connection destination.
- a USB I / F 19 may be connected to the I / O controller 31.
- a VGA I / F 20 may be connected to the graphic controller 32.
- a SAS I / F 154 may be connected to the disk controller 33.
- An HDD may be connected to the SAS I / F 154 as a kind of the storage disk 105.
- Each physical device may transmit / receive data to / from each other by PCIe through the data bus 16 and the PCIe switch 17.
- FIG. 3 shows a configuration example of the bare metal host 10a, the virtual machine host 10b, and the LPAR host 10c.
- a physical resource 14a provided by a physical device (CPU 15a, memory 18a, NIC 34a, HBA 35a, etc.) is directly assigned to the instance 11a realized in the bare metal host 10a. Therefore, the guest OS 13a related to the instance 11a can occupy all the physical resources 14a provided by the physical device of the bare metal host 10a.
- the utilization rate of the physical resource 14a of the bare metal host 10a is the use of physical resources 14b and 14c of the virtual machine host 10b and the LPAR host 10c described later. It tends to be lower than the rate. However, since there is no instance sharing the physical resource 14a, the instance 11a is not affected by other instances.
- the utilization rate is a value representing how many physical resources are used in terms of time or space among the available physical resources. Details of the utilization rate will be described later (see FIG. 8).
- the virtual machine host 10b includes a virtualization mechanism 24.
- the virtualization mechanism 24 may be configured by either hardware or software.
- the virtualization mechanism 24 divides the physical resource 14b of the physical device to configure a plurality of virtual devices.
- a resource provided by this virtual device is referred to as a virtual resource 41.
- the virtualization mechanism 24 allocates a virtual resource 41b obtained by dividing the physical resource 14b to each of the plurality of instances 11b realized in the virtual machine host 10b. Therefore, the guest OS 13b of each instance 11b can use only the virtual resources 41b allocated by the virtualization mechanism 24 among the physical resources 14b.
- the virtual machine host 10b is often configured. The reason is as follows. In the virtual machine host 10b, since the physical resource 14b is shared and used from the plurality of instances 11b, the utilization rate of the physical resource 14b of the virtual machine host 10b is compared with the utilization rate of the physical resource 14a of the bare metal host 10a. It tends to be expensive. When increasing the physical resource 14a allocated to the instance 11a, it is necessary to add a physical device in the bare metal host 10a. However, in the virtual machine host 10b, the setting of the virtualization mechanism 24 may be changed so as to increase the allocation amount of the physical resource 14b to the virtual resource 41b.
- the instance 11b may be affected by other instances.
- the processing loads of a plurality of instances 11b on the virtual machine host 10b increase simultaneously, the physical resource 14b shared by the plurality of instances 11b may become a performance bottleneck. Accordingly, when it is desired to construct an instance (for example, a DB server) having stable I / O performance, it is not appropriate to construct the instance 11b in the virtual machine host 10b.
- an instance for example, a DB server
- the LPAR host 10 c has an LPAR mechanism 37.
- the LPAR mechanism 37 may be configured by either hardware or software.
- the physical device provides the physical resource 14c, but the predetermined physical device can further provide a plurality of independent physical resources (referred to as “independent physical resources”). Independent physical resources are not affected by the status of other independent physical resources.
- each core of the CPU 15 can be an independent physical resource.
- each partition obtained by dividing the address space of the memory 18 can be an independent physical resource.
- each spindle (for example, a plurality of HDDs) of the storage apparatus 100 can be an independent physical resource.
- the physical device is the network adapter 34, 35, etc., the ports 61, 51 that each has are independent physical resources.
- the I / Fs 19, 20, 154 connected to the controllers 31, 32, 33 can be independent physical resources. Further, the physical address of the data cache, the RAID group, the HDD, the physical port on the network switch, and the like can be independent physical resources.
- a physical device having an SR-IOV (Single Root Input Output Virtualization) function can multiplex logical addresses on the physical device side and provide a logically independent physical area to the instance 11. Therefore, for a physical device having an SR-IOV function, each of its logically independent physical areas may be an independent physical resource.
- SR-IOV Single Root Input Output Virtualization
- each of its logically independent physical areas may be an independent physical resource.
- the NIC 34 has an SR-IOV function, it is physically one NIC 34, but each NIC that is logically handled as a plurality can be an independent physical resource.
- the LPAR mechanism 37 can control whether an independent physical resource is allocated exclusively to one instance 11 or shared to a plurality of instances 11.
- the LAPR mechanism 37 may configure a plurality of virtual resources from the independent physical resources and assign the virtual resources to the instances 11.
- An independent physical resource allocated exclusively to one instance 11 is referred to as an occupied resource 42.
- An independent physical resource that is shared to a plurality of instances 11 is referred to as a shared resource.
- the structural difference between the bare metal host 10a and the virtual machine host 10b is the presence or absence of the virtualization mechanism 24. Accordingly, the virtualization mechanism 24 may be introduced into the bare metal host 10a to become the virtual machine host 10b, and conversely, the virtualization mechanism 24 may be deleted from the virtual machine host 10b to become the bare metal host 10a.
- the difference in configuration between the bare metal host 10 a and the LPAR host 10 c is the presence or absence of the LPAR mechanism 37. Therefore, the LPAR mechanism 37 may be introduced into the bare metal host 10a to make the LPAR host 10c, and conversely, the LPAR mechanism 37 may be deleted from the LPAR host 10c to become the bare metal host 10a.
- the difference in configuration between the virtual machine host 10b and the LPAR host 10c is a difference between the virtualization mechanism 24 and the LPAR mechanism 37. Accordingly, the virtualization mechanism 24 of the virtual machine host 10b is changed to the LPAR mechanism 37 to be the LPAR host 10c, and conversely, the LPAR mechanism 37 of the LPAR host 10c is changed to the virtualization mechanism 24 to be the virtual machine host 10b. You can do it.
- FIG. 4 shows a configuration example in which a plurality of LPAR hosts 10c are connected and used.
- the LPAR mechanisms 37 of the respective LPAR hosts 10c may cooperate with each other so that the physical resources 14 of the respective LPAR hosts 10c can be mutually used.
- the data bus 16-1 of the LPAR host 10c-1 and the data bus 16-2 of the LPAR host 10c-2 are connected by an interconnect 16-3.
- the LPAR mechanism 37 can allocate an independent physical resource of the LPAR host 10c-2 different from the LPAR host 10c-1 in which the instance 11c is configured to the instance 11c. For example, in FIG. 4, the LPAR mechanism 37 assigns the core of the CPU 15-2 of the LPAR host 10c-2 to the instance 11c configured in the LPAR host 10c-1.
- one of the plurality of LPAR hosts 10c is selected as a representative node, and the identifier of the representative node may be used as an identifier of the plurality of LPAR hosts 10c. .
- the interconnect 16-3 has a larger communication delay than the data bus 16. Therefore, the LPAR mechanism 37 preferentially allocates the physical resource 14-1 connected by the data bus 16-1 in the LPAR host 10c-1 in which the instance 11c is configured to the instance 11c, and the allocation is performed. If this is not possible, the physical resource 14-2 ahead of the interconnect 16-3 in another LPAR host 10c-2 may be allocated to the instance 11.
- the guest OS 13c on the instance 11c has a set of physical resources 14-1 (for example, a set of the CPU 15-1 and the memory 18-1) connected to the data bus 16-1 in order to cope with such a communication delay or the like. ) May be detected.
- the guest OS 13c corresponding to the NUMA (Non-Uniform Memory Access) architecture has such a function.
- FIG. 5 shows a configuration example of the storage apparatus 100.
- the storage apparatus 100 stores data requested to be written from the physical computer 10 (or an instance 11 of the physical computer 10; the same applies to FIG. 5 below), and passes the data requested to be read to the physical computer 10.
- the storage apparatus 100 includes, for example, a processor 151, a memory 152, a cache memory 153, a SATA (Serial ATA) I / F 154, an HBA 159, a NIC 155, a storage medium 105, and a data bus to which these elements are connected. 156.
- the processor 151, the memory 152, the cache memory 153, the SATA I / F 154, the HBA 159, and the NIC 155 may be collectively referred to as the storage controller 150.
- the storage apparatus 100 may form a RAID group 104 from a plurality of storage media 105.
- the HBA 159 is connected to the SAN 55 and transmits / receives a write request and a read request to / from the physical computer 10.
- the NIC 155 is connected to the management network 252 and transmits and receives control requests and the like to and from the management computer 200.
- the SATA I / F 154 is connected to the RAID group 104 and transmits / receives read data and write data to / from the RAID group 154.
- the cache memory 153 has a higher I / O speed than the storage medium 105.
- the cache memory 153 temporarily stores read data and write data transmitted / received via the SATA I / F 154. This is for speeding up the response of the storage controller 150 to the physical computer 10.
- the processor 151 executes the response program 153a, the control program 153b, and the like stored in the memory 152, and realizes various functions of the storage apparatus 100.
- the response program 153a is a program that returns a response to the read request and the write request transmitted from the physical computer 10.
- the control program 153 b is a program that configures the logical volume 101 from the storage area of the RAID group 104 and provides the volume 101 to the physical computer 10.
- the control program 153 b may configure the storage pool 106 from the storage area of the RAID group 104 and configure the virtual volume 107 from the storage pool 106. Then, when storing the write data in the virtual volume 107, the control program 153b may reserve an area for storing the write data in the storage pool 106. This is because a plurality of physical computers 10 can use the storage area of the RAID group 104 without waste.
- the volume 101 and the virtual volume 107 may be called storage resources.
- a storage resource can be an independent physical resource.
- the control program 153b may manage to which physical computer 10 (instance 11) the storage resource is allocated and / or as an occupied resource or a shared resource.
- FIG. 6 shows a functional configuration example of the management computer 200.
- the management computer 200 includes, for example, an integrated resource management unit 202, a computer management unit 203, a network management unit 204, and a storage management unit 205 as one type of function.
- the integrated resource management unit 202 includes an instance configuration unit 211, a requirement reception unit 210, an instance information table 212, a resource information table 213, a template information table 214, a template usage frequency table 218, a topology information table 215, A candidate table 217 and a resource operation information database 216 may be included.
- the resource operation information database 216 may store history information including information on past movable performance of each device.
- the instance configuration unit 211 Based on the information stored in each of the tables 212 to 217, the instance configuration unit 211 satisfies the instance configuration requirements and sets each physical computer 10 to have the highest utilization rate of the physical resources 14 of the physical computer 10 as much as possible. Control.
- the instance configuration requirements may include information on requirements that the user requests from the instance 11.
- the instance configuration requirement may include information on a resource that is a constituent element of the instance 11 and information on a function, performance, and quality required for the resource.
- the instance configuration unit 211 may use a resource for which high quality is specified in the instance configuration requirement as an occupied resource, and other resources as shared resources.
- the requirement receiving unit 210 receives the instance configuration requirement from the management client 72 of the client terminal 70. Then, the requirement reception unit 210 passes the received instance configuration requirement to the instance configuration unit 211 as appropriate. For example, when the same resource is requested in a plurality of instance configuration requirements, the requirement receiving unit 210 passes the previously received instance configuration requirement to the instance configuration unit 211 first, and temporarily holds the received instance configuration requirement. May be.
- instance information table 212 for each instance 11 configured in the computer system 1, instance information that is information related to the configuration of the instance 11 is stored. Details of the instance information table 212 will be described later (see FIG. 7).
- the resource information table 213 stores, for each physical device, independent physical resource information that is information related to the independent physical resource that the physical device has. Details of the resource information table 213 will be described later (see FIG. 8).
- the template information table 214 stores template information. The user can create an instance configuration requirement more easily by using the template information. Details of the template information table 214 will be described later (see FIG. 9).
- the topology information table 215 stores, for each instance 11 configured in the computer system 1, information (referred to as “topology information”) indicating the relationship between the resource configuring the instance 11 and the physical device related to the resource. Is done. Details of the topology information table 215 will be described later (see FIG. 10).
- candidates for the physical computer 10 to which the instance 11 is deployed are stored during the process of creating the instance 11. Details of the candidate table 217 will be described later (see FIG. 11).
- the template usage frequency table 218 stores information on the usage frequency of template information. Details of the template usage frequency table 218 will be described later (see FIG. 13).
- FIG. 7 shows a configuration example of the instance information table 212.
- the instance information table 212 stores instance information of each instance 11 configured in the computer system 1.
- the instance information may include an instance ID, common information 212-1, calculation resource information 212-2, network resource information 212-3, storage resource information 212-4, and function information 212-5.
- the instance ID is information for identifying the instance 11.
- the common information 212-1 includes information common to the instance 11 of the instance ID.
- the common information 212-1 includes, for example, a template name, an instance name, an instance realization method, and a user name.
- the template name indicates the name of the template information based on the instance 11 of the instance ID.
- the instance name indicates the name of the instance 11 of the instance ID.
- the instance implementation method indicates whether the instance 11 of the instance ID is realized by, for example, the “bare metal method”, the “virtualization method”, or the “LPAR method”.
- the user name indicates the name of the user who owns the instance 11 of the instance ID.
- the computation resource information 212-2 includes physical resource information related to the computation assigned to the instance 11 of the instance ID.
- the computing resource information 212-2 includes an allocation mode, an operating frequency, and the number of cores for each CPU resource allocated to the instance 11.
- the allocation mode related to the CPU resource indicates whether the CPU resource allocated to the instance 11 of the instance ID is an occupied resource or a shared resource.
- a CPU resource whose allocation mode is “occupied” is not used by instances other than the instance 11 of the instance ID.
- a CPU resource whose allocation mode is “shared” may be used by instances other than the instance 11 of the instance ID.
- the operating frequency indicates the operating frequency of this CPU resource.
- the number of cores indicates the number of cores that the CPU resource has.
- the network resource information 212-3 includes information on physical resources related to the network assigned to the instance 11 of the instance ID.
- the network resource information includes, for each NIC resource allocated to the instance 11, an allocation mode, a band, the number of ports, and a VLAN ID.
- the allocation mode related to the NIC resource indicates whether the NIC resource allocated to the instance 11 of the instance ID is an occupied resource or a shared resource.
- the NIC resource whose allocation mode is “occupied” is not used from instances other than the instance 11 of the instance ID. There is a possibility that the NIC resource whose allocation mode is “shared” is also used by instances other than the instance 11 of the instance ID.
- Band indicates the maximum band of this NIC resource.
- the number of ports indicates the number of ports that the NIC resource has.
- the VLAN ID indicates the ID of the VLAN set in this NIC resource.
- the storage resource information 212-4 includes information on physical resources related to storage allocated to the instance 11 of the instance ID.
- the storage resource information 212-4 includes an allocation mode, a capacity, a hierarchy, and a redundant path for each virtual storage allocated to the instance 11.
- the allocation mode related to storage indicates whether the storage resource allocated to the instance 11 of the instance ID is an occupied resource or a shared resource. Storage resources whose allocation mode is “occupied” are not used by instances other than the instance 11 of the instance ID. A storage resource whose allocation mode is “shared” may be used by instances other than the instance 11 of the instance ID.
- Capacity indicates the maximum storage capacity of this storage resource.
- the hierarchy indicates the hierarchy (tier) of this storage resource.
- the redundant path indicates the number of redundant paths for this storage resource.
- the function information 212-5 includes information related to the function set in the instance 11 of the instance ID.
- the function information 212-5 includes a migration function and a high availability function.
- the migration function is a function related to the migration of the instance 11 of the instance ID.
- the instance 11 whose migration function is “non-stop” has a function necessary for non-stop migration.
- the high availability function is a function related to the availability of the instance 11 of the instance ID. For example, the instance 11 in which the high availability function is “valid” has a function necessary for the high availability operation.
- the user can specify all or part of the above-mentioned instance information items using the instance configuration requirements. That is, the instance configuration requirement may include the same items as the above-described instance information.
- the management computer 200 may automatically set items not specified by the user in the instance configuration requirements. Items for which an invalid value is designated by the user in the instance configuration requirement may be automatically corrected by the management computer 200. When the value specified by the user in the instance configuration requirement is inconsistent between items, the management computer 200 may automatically correct the value. “N / A” may be set in the resource item that is not allocated to the instance 11.
- the instance 11 having the instance ID “IST-A0-0002” is realized by the “LPAR method”.
- CPU resource “CPU # 0” assigned to this instance 11 is “shared”. Therefore, the arithmetic processing by CPU # 0 in this instance 11 may be affected by other instances.
- the NIC resource “NIC # 0” assigned to this instance 11 is “occupied”. Therefore, the communication processing by NIC # 0 in this instance 11 is not affected by other instances.
- the storage resource “storage # 0” allocated to this instance 11 is “shared”. Therefore, the I / O for storage # 0 in this instance 11 may be affected by other instances.
- FIG. 8 shows a configuration example of the resource information table 213.
- the resource information table 213 stores, for each physical device that each physical computer 10 has, information related to the independent physical resource that the physical device has (referred to as “independent physical resource information”).
- the physical device possessed by the physical computer 10 may have the following independent physical resources.
- the independent physical resource related to the CPU may be each core of the CPU.
- each core group of cores # 0 to # 7, cores # 8 to # 11, cores # 12 to # 17, and cores # 28 to # 31 in the CPU is one independent physical resource. .
- the independent physical resource related to the memory may be one partition among the storage areas of the memory divided into a plurality of partitions. For example, in the case of FIG. 8, each partition of memory addresses # 00 to # 9F, addresses # A0 to #CF, and addresses # D0 to #FF is one independent physical resource.
- the independent physical resource related to the NIC may be each slot of the NIC.
- each of the NIC slots # 0, # 1, and # 2 is one independent physical resource.
- the physical device included in the storage apparatus 100 may have the following independent physical resources.
- the independent physical resource related to the processor may be each processor.
- each of MP # 1 and # 2 is one independent physical resource.
- the independent physical resource related to the cache memory may be each slot.
- each of the slots A0 and A1 is one independent physical resource.
- the independent physical resource related to the port may be each port.
- each of the ports P00-01, P02-10, and P11-15 is one independent physical resource.
- the independent physical resource related to the RAID group may be each RAID group.
- the RAID group “RAID # 00” is one independent physical resource.
- the independent physical resource information includes, as field values, for example, device ID 213-1, physical device name 213-2, independent physical resource name 213-3, allocation mode 213-4, and utilization rate 213-5. Have.
- the device ID 213-1 is information for identifying devices such as the physical computer 10 and the storage device 100.
- the physical device name 213-2 indicates the name of the physical device that the device having the device ID 213-1 has.
- the independent physical resource name 213-3 indicates the name of the independent physical resource that the physical device having the physical device name 213-2 has.
- Allocation mode 213-4 indicates whether the independent physical resource with independent physical resource name 213-3 is in the “occupied”, “shared”, or “unallocated” state.
- An independent physical resource whose allocation mode is “occupied” is allocated to one instance 11.
- An independent physical resource whose allocation mode is “shared” may be allocated to a plurality of instances 11.
- An independent physical resource whose allocation mode is “unallocated” is not allocated to any instance 11.
- the utilization rate 213-5 is information indicating the utilization rate of the independent physical resource of the independent physical resource name 213-3.
- the utilization rate related to the CPU may be a utilization rate per unit time in a core group that is an independent physical resource.
- the utilization rate related to the memory may be a utilization rate per partition which is an independent physical resource.
- the utilization rate related to the NIC may be a utilization rate per unit time in one slot which is an independent physical resource.
- the utilization rate related to the processor may be a utilization rate per unit time in one processor which is an independent physical resource.
- the utilization rate related to the cache memory may be a utilization rate per unit time in one slot which is an independent physical resource.
- the utilization rate related to the port may be a utilization rate per unit time in one port which is an independent physical resource.
- the utilization rate related to the RAID group may be a utilization rate per storage capacity of one RAID group which is an independent physical resource.
- FIG. 9 shows a configuration example of the template information table 214.
- the template information table 214 stores template information.
- the template information may be set with a default value that can be used to create an instance configuration requirement.
- the template information may be created in advance by an operator of the computer system 1 or the like.
- the template information table 214 may store template information of various patterns according to user needs.
- the template information table 214 may store various patterns of template information, such as by performance of the instance 11, by use, or by cost. Different default values may be set for template information having different patterns.
- selectable default values In the template information, selectable default values, upper and lower limit values, default values that cannot be changed, and the like may be set.
- a default value appropriately sized for an application running on the instance 11 may be set. This default value may be calculated on the basis of a configuration in which performance requirements that can be exhibited by an application resource requirement and a prior operation test are estimated.
- default values suitable for the configuration of the physical computer 10 and the storage apparatus 100 of the computer system 1 may be set.
- a default value that increases the utilization rate of the entire physical device of the computer system 1 may be set.
- a default value obtained by equally dividing the physical resource 14 of each physical device may be set in the template information.
- “occupied” and “shared” options may be set as default values of the allocation mode.
- templates of “occupied” and “shared” may be set as the default value of the allocation mode.
- options of “occupied” and “shared” may be set as the default value of the allocation mode.
- “occupy” may be set as the default value of the allocation mode.
- “shared” may be set as the default value of the allocation mode.
- Template information is not limited to the configuration shown in FIG.
- the template information may have a configuration in which occupation / sharing of a specific independent physical resource such as “CPU resource occupation type” can be set in the common information 214-1.
- the user can easily create an instance configuration requirement by using template information having a setting close to a desired instance configuration. Further, the business operator of the computer system 1 can easily make a plan for expansion and change of the resource configuration of the computer system 1 by having the user use the template information.
- FIG. 10 shows a configuration example of the topology information table 215.
- the instance 11 arranged in the physical computer 10 and the information related to the physical resource used from the instance 11 are stored in association with each other.
- the topology information table 215 includes, as field values, for example, an instance ID 215-1, a resource pool ID 215-2, information about a physical computer 215-3, an Ethernet switch ID 215-4, an FC switch ID 215-5, and a storage device Information 215-6.
- the physical computer information 215-3 includes, for example, a physical computer ID and information on the CPU core and NIC related to the independent physical resource.
- the information 215-6 related to the storage device includes, for example, information related to the storage device ID, the port ID related to the independent physical source, the cache ID, the volume ID, and the RAID group ID.
- the instance ID 215-1 is information for identifying the instance 11 already arranged in the physical computer 10.
- the resource pool ID 215-2 indicates to which resource pool the instance 11 of the instance ID 215a, the physical computer 10, and the like belong.
- the physical computer ID indicates the physical computer 10 having a physical resource assigned to the instance 11 of the instance ID 215-1. That is, the instance 11 with the instance ID 215-1 is arranged in the physical computer 10 with this physical computer ID.
- the CPU core indicates a core allocated to the instance 11 of the instance ID 215-1 among the cores related to the independent physical resources of the CPU included in the physical computer 10 having the physical computer ID.
- NIC indicates a slot allocated to the instance 11 of the instance ID 215-1 among the slots related to the independent physical resources of the NIC included in the physical computer 10 having the physical computer ID.
- the Ethernet switch ID 215-4 indicates the Ethernet switch used by the instance 11 of the instance ID 215-1. That is, a physical computer 10 having a physical computer ID is connected to the Ethernet switch.
- FC switch ID 215-5 indicates an FC switch used by instance 11 of instance ID 215-1. That is, the physical computer 10 with the physical computer ID is connected to this FC switch.
- Storage device ID indicates a storage device 100 having a storage resource assigned to the instance 11 of the instance ID 215-1.
- the port ID, cache ID, volume ID, and RAID group ID indicate the port, cache, volume, and RAID group related to the storage device used by the instance 11 of the instance ID 215-1.
- the performance of the instance 11 decreases as the network topology in which the physical computer 10 in which the instance 11 is arranged and the independent physical resource allocated to the instance 11 is farther from each other. Therefore, it is desirable that an independent physical resource close to the network topology is assigned to the instance 11.
- the topology information table 215 is referred to when it is desired to know the network topology distance between physical devices. For example, it can be said that two physical computers 10 having the same FC switch are close in network topology.
- the topology information table 215 is referred to when it is desired to know the physical devices and physical resources allocated to the allocated instance 11.
- CPU cores having the same physical computer ID and not associated with any instance ID can be said to be “unassigned”.
- FIGS. 11A and 11B show a configuration example of the candidate table 217.
- FIG. The management computer 200 may employ any candidate table 217 shown in FIGS. 11A and 11B.
- the candidate table 217 stores a plurality of instance configuration candidates.
- the candidate ID 217-1 is information for identifying the instance configuration candidate.
- the candidate table 217 may store information for calculating an evaluation index of the instance configuration of the candidate ID 217-1.
- the evaluation index is a value serving as an index for determining one instance configuration from among a plurality of instance configuration candidates.
- the evaluation index in the physical computer 10 may be the surplus resource 217-2 in each of the CPU, memory, and NIC.
- the evaluation index in the storage apparatus 100 may be the surplus resource 217-3 in each of the spindle, port, and cache.
- the surplus resources may be indicated by the ratio of the surplus resources in the entire resource as indicated by 217a-2 and 217a-3 in FIG. 11A, or independently as indicated by 217b-2 and 217b-3 in FIG. 11B. It may be the number of unassigned independent physical resources in the total number of physical resources.
- FIG. 12 is a flowchart illustrating an example of the creation process of the instance 11.
- the user accesses the requirement receiving unit 210 of the management computer 200 through the management client 72 of the client terminal 70. Then, the user selects desired template information from a plurality of template information stored in the template information table 214 (S330).
- the user creates an instance configuration requirement using the selected template information.
- a default value, an upper limit value and a lower limit value, and / or options may be set in the template information item.
- the user may modify each item based on the setting and create an instance configuration requirement.
- the requirement receiving unit 210 may notify the user that it is illegal.
- the instance configuration unit 211 determines whether or not “occupied” is designated in any allocation mode of the instance configuration requirements (S331). It can be said that the resource for which “occupied” is designated requires the performance to be guaranteed for the resource. It can be said that the resource for which “occupancy” is not designated does not require the performance guarantee for the resource.
- the process when “Occupancy” is not specified in any allocation mode of the instance configuration requirement (S331: NO) is as follows.
- the instance configuration unit 211 creates the instance 11 based on the virtualization method (S332). For example, the instance configuration unit 211 creates and arranges the instance 11 in the virtual machine host 10b. This is because this user requires performance guarantee for any resource, so even if the instance 11 is created and arranged in the virtual machine host 10b, this user's requirement is satisfied. Further, the business operator of the computer system 1 can improve the utilization rate of physical resources in the computer system 1 by creating and arranging the instance 11 in the virtual machine host 10b.
- the process when “Occupied” is designated in any allocation mode of the instance configuration requirement is as follows.
- a resource for which “occupied” is designated as the instance configuration requirement allocation mode is referred to as an “occupied designated resource”.
- the instance configuration unit 211 searches the resource information table 213 for the device ID 213-1 in which the allocation mode 213-4 of the independent physical resource name 213-3 corresponding to the exclusive designated resource is “unallocated” (S333). .
- the instance configuration unit 211 searches the resource information table 213 for a physical computer having a CPU whose allocation mode includes “unallocated”. Good.
- the instance configuration unit 211 searches the resource information table 213 for a physical computer having a NIC that includes “unallocated” in the allocation mode. Good.
- the instance configuration unit 211 searches the resource information table 213 for a storage apparatus 100 having a RAID group that includes “unallocated” in the allocation mode. May be.
- the instance configuration unit 211 when “occupied” is specified for both the computation resource and the network resource, the instance configuration unit 211 “unassigned” in the allocation mode of both the CPU and the NIC from the resource information table 213. You may search the physical computer which contains.
- the instance configuration unit 211 may reflect the history information in the resource operation information database 216 in the utilization rate 213-5 of the resource information table 213 before executing the process of S333.
- the instance configuration unit 211 selects one device (referred to as “target device”) from the devices searched in the process of S333. Then, the instance configuration unit 211 searches the topology information table 215 for a device connected to the target device (referred to as “connected device”).
- the instance configuration unit 211 determines whether any of the searched connection devices can provide a physical resource to a resource for which “shared” is specified in the instance configuration requirement (referred to as “shared specified resource”). Is determined (S334).
- the instance configuration unit 211 may make the determination by referring to information related to the connection device in the resource information table 213. If the instance configuration unit 211 cannot search for a connection device from the topology information table 215, the instance configuration unit 211 may also search for a connection device that is not stored in the topology information table 215.
- the reason why the connection device capable of providing the physical resource to the instance 11 is searched from the topology information table 215 is as follows.
- the topology information table 215 stores devices that provide physical resources to the arranged instances 11. Therefore, by preferentially allocating the physical resource of the device stored in the topology information table 215 to the newly created instance 11, the utilization rate of the physical resource per device can be increased.
- the instance configuration unit 211 may select a connection device that is closest to the target device in terms of network topology. As a result, the instance configuration unit 211 can reduce overhead due to a communication delay between the target device and the connection device as much as possible.
- the process when any of the connection devices can provide the physical resource to the shared designated resource is as follows.
- the instance configuration unit 211 assigns a candidate ID 217-1 to the set of the target device and the connection device, and registers the set related to the candidate ID 217-1 in the candidate table 217 (S335). Then, the instance configuration unit 211 proceeds to the process of S336.
- connection devices can provide the physical resource to the shared designated resource (S334: NO)
- the instance configuration unit 211 proceeds directly to the process of S336.
- the instance configuration unit 211 determines whether or not an unselected device remains among the devices searched in the process of S333 (S336). If an unselected device remains (S336: YES), the instance configuration unit 211 returns to the process of S333, selects one of the unselected devices, and repeats the process.
- the instance configuration unit 211 selects one device ID 217-1 from the device IDs 217-1 stored in the candidate table 217 (S337). For example, the instance configuration unit 211 selects a set of device IDs 217-1 that maximizes the utilization rate of physical resources.
- the instance configuration unit 221 may evaluate the known candidate device higher than the unknown candidate device with respect to the utilization rate of the physical resource.
- An unknown candidate device is a candidate device that does not provide physical resources to any instance 11.
- a known candidate device is a candidate device that provides physical resources to any instance 11.
- the reason for this evaluation is that the unknown candidate device does not operate the LPAR mechanism 37, and there is a possibility that a switching cost for newly introducing the LPAR mechanism 37 may occur. On the other hand, in the known candidate apparatus, since the LPAR mechanism 37 is operating, there is no possibility that such a cost will occur.
- the instance configuration unit 211 may select a set of the bare metal host 10a instead of the LPAR host 10c in the process of S337. This is because the LPAR mechanism 37 can be introduced into the bare metal host 10a so as to be the LPAR host 10c. Further, the instance configuration unit 211 may move another instance 11 operating in the set selected from the candidate table in the process of S337 to another LPAR host as necessary.
- the surplus resource 217b-2 of this physical computer is extremely small (for example, the surplus resource 217b-2
- the instance configuration The unit 211 may operate this physical computer not using the LPAR method but using the bare metal method.
- the instance 11 to which a physical resource is allocated from a physical computer operated in a bare metal system can obtain relatively high performance because there is no LAPR mechanism 37 overhead.
- the predetermined license cost for the instance 11 may be lowered.
- this instance 11 is cost-effective and meets instance configuration requirements.
- the instance configuration unit 211 creates and arranges the instance 11 in the physical computer selected in the process of S337. As described above, the instance configuration unit 211 may introduce the LPAR mechanism 37 into the selected physical computer 10 as necessary.
- the above-described change in the physical computer implementation method and / or movement of other instances may be performed each time a new instance 11 is created and arranged.
- the utilization rate of the physical resource of the computer system 1 is optimized according to the situation at that time. That is, in the computer system 1, the physical resources in the entire computer system 1 even when the addition, change and deletion of the physical device related to the physical computer 10 and the addition, configuration change and deletion of the instance 11 occur.
- the utilization rate of can be kept as high as possible.
- IaaS creation, configuration change, and deletion of an instance are requested asynchronously by many users.
- the utilization rate of physical resources in the entire computer system 1 is improved.
- different instances of the owning user share the physical resources of one physical computer. Therefore, when the processing load of an instance owned by a certain user increases, the performance of the instance owned by another user may deteriorate.
- the LPAR host 10c does not “occupy” all physical resources of the physical computer, but can set “occupied” and “shared” in units of independent physical resources that are part of the physical resource. Easy to increase rate. However, it is difficult to effectively increase the utilization rate unless the number of physical devices included in the LPAR host 10c and the utilization rate of each physical device are known. In addition, the physical device of the LPAR host 10c may be changed. A user who uses IaaS does not know the physical device that the LPAR host 10c has. Therefore, with the instance configuration requirements arbitrarily created by the user, it is difficult to effectively increase the utilization rate of physical resources in the LPAR host 10c.
- template information suitable for the configuration of the computer system 1 is provided to the user, and the instance configuration requirement is created by using the template information, so that the utilization rate of physical resources in the computer system 1 can be increased. Can be effectively increased.
- the user can create an instance configuration requirement more easily by using the template information.
- FIG. 13 shows a configuration example of the template usage frequency table 218.
- the information stored in the template usage frequency table 218 is used when creating more appropriate template information.
- the template information needs to be created so as to conform to both the physical device configuration of the physical computer 10 and the instance configuration requirements desired by the user. If the template information does not conform to the configuration of the physical device of the physical computer 10, when the template information is used, the utilization rate of physical resources may be reduced.
- the template information has only an option for specifying “occupied” or “shared” in two NICs.
- the user can only specify “occupied” in two NICs.
- the instance 11 configured in this manner “occupies” the physical resources of two NICs, there is a high possibility that only one NIC physical resource is used. In this case, the utilization rate of NIC physical resources in the physical computer 10 decreases.
- the configuration of the physical device that the physical computer 10 has may be changed.
- a physical device whose support period has ended may be replaced with another physical device.
- the contract contents relating to the physical device may be updated due to the convenience of the supplier of the physical device or a change in demand.
- the template usage frequency table 218 may store template information and information related to the usage frequency of template information derived from the template information (referred to as “derived template information”).
- Derived template information may be created based on template configuration requirements created by changing the content of the original template information.
- the default value set in the derived template information may be calculated based on how the user has changed the default value set in the original template information. For this calculation, a statistical method such as cluster analysis may be applied.
- the management computer 200 can create derivative template information including default values and the like reflecting many users' requests.
- the template usage frequency table 218 has, for example, a template ID 218-1, a derived template ID 218-1, a usage frequency 218-3, a difference 218-4, and an approximate model 218-5 as field values.
- the usage frequency 218-3 is a value indicating how often the template information or the derived template information (referred to as “template information etc.”) is used by the user.
- the usage frequency 218-3 may be the total number of times that the template information or the like is used, or the number of users that use the template information or the like.
- the usage frequency 218-3 may be a usage frequency of template information or the like in a predetermined period.
- the usage frequency 218-3 may be added when the template information or the like is used to create the instance 11.
- the usage frequency 218-3 may be subtracted when the instance 11 created based on the template information or the like is deleted or reconfigured.
- the difference 218-4 is information indicating the difference of the physical resource 14 between the original template information and the derived template information.
- the difference may be calculated based on the change made by the user in the process of S330 in FIG.
- the approximate model 218-5 is information indicating which of the resource configurations (referred to as “resource configuration models”) of the physical computers that are set in the template information etc. is approximate to the resource configuration of a plurality of different models. It is.
- the record of the template ID 218-1 “TMP3023-9002” shown in FIG. 13 indicates that the approximate model 218-5 is “1 / 2-A model”. This is because the resource configuration set in the template information of this template ID 218-1 “TMP3023-9002” approximates the configuration of one-half of the physical resources of the resource configuration model “A model”. Show. For example, when the resource configuration model “A model” has four NICs, two NICs are set in the template information of the approximate model “1 / 2-A model”.
- the resource configuration model may indicate the configuration of any physical resource of the physical computer 10, the network device, and the storage device 100 that make up the computer system 1.
- the resource configuration model may be expressed by a model number defined by the manufacturer of the device or a product name.
- FIG. 14 shows an example of processing for creating derived template information.
- the instance configuration unit 211 creates derived template information based on the template information (S340).
- the instance configuration unit 211 may classify the configuration of the instance 11 based on items changed by the user from the template information, and create derived template information. Then, the instance configuration unit 211 stores a record related to the derived template information in the template usage frequency table 218.
- the instance configuration unit 211 calculates a use frequency and an approximate model for the created derived template information. Then, the instance configuration unit 211 stores the calculation result in the use frequency 218-3 and approximate model 218-5 of the record relating to the generated derived template information in the template use frequency table 218 (S341).
- the instance configuration unit 211 searches the template usage frequency table 218 for derived template information having a higher usage frequency 218-3 than the original template information (S342).
- the derived template information searched here is more suitable for the user's needs than the original template information created by the operator of the computer system 1. In other words, it can be said that there is room for improvement in the original template information.
- the instance constructing unit 211 rejects the derived template information (S344), and ends this processing. It's okay.
- the instance configuration unit 211 executes the following processing for the searched derived template information.
- the instance configuration unit 211 refers to the template usage frequency table 218 and determines whether the granularity of division in the approximate model 218-5 is smaller in the derived template information than in the original template information (S343). That is, it is determined whether the allocation of the physical resource 14 can be set in a smaller unit in the derived template information than in the original template information.
- the instance configuration unit 211 adopts the derived template information as template information (S345). ), This process may be terminated.
- the instance configuration unit 211 may store this derived template information in the template information table 214 as new template information.
- the instance configuration unit 211 may replace the original template information with the derived template information in the template information table 214.
- the instance configuration unit 211 discards the derived template information (S344). ), This process may be terminated.
- the usage frequency 218-3 “190” of the derived template information “derivation 2” from the original template information “TMP3023-9002” is the usage frequency 218 of the original template information. -3 It is higher than “182”.
- the approximate model 218-5 “1 / 3-B model” of the derived template information “derivative 2” is divided more than the approximate model 218-5 “1 / 2-A model” of the original template information.
- the particle size is small.
- the derivation template information “derivation 2” of the original template information “TMP3023-9002” may be adopted as new instance information.
- the instance configuration unit 211 may execute the following process. The instance configuration unit 211 selects an instance to be examined from the instance information table 212. Then, the instance configuration unit 211 determines whether or not this instance is configured based on the original template information of the derived template information. If this determination is affirmative, the instance configuration unit 211 may change the configuration of this instance to match the setting of the derived template information based on the user's approval.
- the template information adapted to the user's needs is automatically obtained based on the usage frequency 218-3 of the original template information and the usage frequency 218-3 of the derived template information changed from the original template information. Can be created.
- the utilization rate of physical resources can be improved.
- the operator of the computer system 1 can improve the cost effectiveness in the operation of the computer system 1.
- the utilization rate of physical resources can be improved.
- the derived template information that matches the change is automatically Can be created.
- Computer system 10 Physical computer 70: Client terminal 100: Storage device 200: Management computer 24: Virtualization mechanism 37: LPAR mechanism
Abstract
Description
Claims (13)
- 複数の計算機と、前記複数の計算機を管理する管理システムと、を含む計算機システムであって、
前記管理システムは、
前記複数の計算機から物理的なリソースである物理リソースが割り当てられることによって実現されるインスタンスを管理し、
前記インスタンスを構成するための設定を含む情報であって、前記インスタンスの前記物理リソースの占有の要否の設定を含むインスタンス構成情報に基づいて、複数の異なるインスタンスの実現方式の中から、前記インスタンス構成情報に対応するインスタンスの実現方式を選択し、
前記複数の計算機は、
前記選択されたインスタンスの実現方式に基づいて、前記インスタンスに前記物理リソースを提供する
計算機システム。 - 前記複数の異なるインスタンスの実現方式には、一つの計算機が提供する物理リソースを複数のインスタンスに共有可能に割り当てる第1の実現方式と、一つの計算機が提供する物理リソースの全部又は一部を一つのインスタンスに占有可能に割り当てる第2の実現方式と、が含まれている
請求項1に記載の計算機システム。 - 前記管理システムは、前記インスタンス構成情報に前記物理リソースの占有を要する旨の設定が含まれていない場合、前記インスタンス構成情報に対応するインスタンスの実現方式として、前記第1の実現方式を選択する
請求項2に記載の計算機システム。 - 前記管理システムは、前記インスタンス構成情報に前記物理リソースの占有を要する旨の設定が含まれている場合、前記インスタンス構成情報に対応するインスタンスの実現方式として、前記第2の実現方式を選択する
請求項2に記載の計算機システム。 - 前記インスタンス構成情報における占有の要否の設定は、種類の異なる物理リソース毎に設定される
請求項4に記載の計算機システム。 - 前記管理システムは、前記インスタンス構成情報において、前記種類の異なる物理リソースの内の第1の種類の物理リソースに占有を要する旨が設定されている場合、何れのインスタンスにも割り当てられていない前記第1の種類の物理リソースを、前記インスタンスに割り当てる
請求項5に記載の計算機システム。 - 前記管理システムは、前記インスタンス構成情報において、前記種類の異なる物理リソースの内の第2の種類の物理リソースに占有を要する旨が設定されていない場合、何れかのインスタンスに割り当てられている前記第2の種類の物理リソースを、優先的に前記インスタンスに割り当てる
請求項5に記載の計算機システム。 - 前記管理システムは、前記第2の実現方式が選択された計算機であって、この計算機の何れのインスタンスにも割り当てられていない物理リソースが所定の閾値未満であり、且つ、この計算機が二つ以上のインスタンスに物理リソースを提供していない場合、この計算機が提供する物理リソースの全部を一つのインスタンスに占有可能に割り当てる
請求項4に記載の計算機システム。 - 前記管理システムは、前記インスタンスに、前記インスタンスを実現する第1の計算機とは異なる第2の計算機の物理リソースを提供する場合、前記第1の計算機からのネットワークトポロジ的な距離の近さに基づいて、前記第2の計算機を決定する
請求項1に記載の計算機システム。 - 前記管理システムは、前記インスタンス構成情報に含まれる設定に関して、デフォルト設定を含むテンプレート情報を有し、
前記デフォルト設定は、前記複数の計算機が提供可能な物理リソースに基づいて決定されている
請求項1に記載の計算機システム。 - 前記管理システムは、前記テンプレート情報の利用頻度、及び、前記テンプレート情報のデフォルト設定から変更された設定に基づいて、新たなデフォルト設定を含むテンプレート情報を作成する
請求項10に記載の計算機システム。 - 複数の計算機を管理する管理システムであって、
前記複数の計算機が物理的なリソースである物理リソースが割り当てられることによって実現されるインスタンスを管理し、
前記インスタンスを構成するための設定を含む情報であって、前記インスタンスの前記物理リソースの占有の要否の設定を含むインスタンス構成情報に基づいて、複数の異なるインスタンスの実現方式の中から、前記インスタンス構成情報に対応するインスタンスの実現方式を選択し、
前記複数の計算機に、前記選択されたインスタンスの実現方式に基づいて、前記インスタンスに前記物理リソースを提供させる
管理システム。 - 複数の計算機の物理的なリソースである物理リソースを管理するリソース管理方法であって、
前記複数の計算機を管理する管理システムが、
前記複数の計算機が物理的なリソースである物理リソースが割り当てられることによって実現されるインスタンスを管理し、
前記インスタンスを構成するための設定を含む情報であって、前記インスタンスの前記物理リソースの占有の要否の設定を含むインスタンス構成情報に基づいて、複数の異なるインスタンスの実現方式の中から、前記インスタンス構成情報に対応するインスタンスの実現方式を選択し、
前記複数の計算機が、
前記選択されたインスタンスの実現方式に基づいて、前記インスタンスに前記物理リソースを提供する
リソース管理方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016568201A JP6480955B2 (ja) | 2015-01-07 | 2015-01-07 | 計算機システム、管理システム、及び、リソース管理方法 |
PCT/JP2015/050236 WO2016110950A1 (ja) | 2015-01-07 | 2015-01-07 | 計算機システム、管理システム、及び、リソース管理方法 |
US15/541,132 US10459768B2 (en) | 2015-01-07 | 2015-01-07 | Computer system, management system, and resource management method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2015/050236 WO2016110950A1 (ja) | 2015-01-07 | 2015-01-07 | 計算機システム、管理システム、及び、リソース管理方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016110950A1 true WO2016110950A1 (ja) | 2016-07-14 |
Family
ID=56355675
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/050236 WO2016110950A1 (ja) | 2015-01-07 | 2015-01-07 | 計算機システム、管理システム、及び、リソース管理方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US10459768B2 (ja) |
JP (1) | JP6480955B2 (ja) |
WO (1) | WO2016110950A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019003286A1 (ja) * | 2017-06-26 | 2019-01-03 | 株式会社日立製作所 | 仮想マシンの管理装置及び管理方法 |
CN111213121A (zh) * | 2017-10-05 | 2020-05-29 | 惠普发展公司,有限责任合伙企业 | 依据基于应用使用聚类选择计算机配置 |
US11601378B2 (en) | 2020-07-20 | 2023-03-07 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for allocating server resource, electronic device and storage medium |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2957584A1 (en) * | 2016-02-12 | 2017-08-12 | Coho Data, Inc. | Methods, systems, and devices for adaptive data resource assignment and placement in distributed data storage systems |
US11195100B2 (en) | 2016-08-17 | 2021-12-07 | International Business Machines Corporation | Determining applications based on interactions of different electronic devices |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010016104A1 (ja) * | 2008-08-04 | 2010-02-11 | 富士通株式会社 | マルチプロセッサシステム,マルチプロセッサシステム用管理装置およびマルチプロセッサシステム用管理プログラムを記録したコンピュータ読取可能な記録媒体 |
JP2010205209A (ja) * | 2009-03-06 | 2010-09-16 | Hitachi Ltd | 管理計算機、計算機システム、物理リソース割り当て方法 |
WO2013078258A1 (en) * | 2011-11-25 | 2013-05-30 | International Business Machines Corporation | System, method and program product for cost-aware selection of templates for provisioning shared resources |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4119239B2 (ja) | 2002-12-20 | 2008-07-16 | 株式会社日立製作所 | 計算機資源割当方法、それを実行するための資源管理サーバおよび計算機システム |
US8484654B2 (en) * | 2010-11-23 | 2013-07-09 | International Business Machines Corporation | Determining suitable network interface for partition deployment/re-deployment in a cloud environment |
-
2015
- 2015-01-07 US US15/541,132 patent/US10459768B2/en active Active
- 2015-01-07 JP JP2016568201A patent/JP6480955B2/ja active Active
- 2015-01-07 WO PCT/JP2015/050236 patent/WO2016110950A1/ja active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010016104A1 (ja) * | 2008-08-04 | 2010-02-11 | 富士通株式会社 | マルチプロセッサシステム,マルチプロセッサシステム用管理装置およびマルチプロセッサシステム用管理プログラムを記録したコンピュータ読取可能な記録媒体 |
JP2010205209A (ja) * | 2009-03-06 | 2010-09-16 | Hitachi Ltd | 管理計算機、計算機システム、物理リソース割り当て方法 |
WO2013078258A1 (en) * | 2011-11-25 | 2013-05-30 | International Business Machines Corporation | System, method and program product for cost-aware selection of templates for provisioning shared resources |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019003286A1 (ja) * | 2017-06-26 | 2019-01-03 | 株式会社日立製作所 | 仮想マシンの管理装置及び管理方法 |
JPWO2019003286A1 (ja) * | 2017-06-26 | 2019-11-07 | 株式会社日立製作所 | 仮想マシンの管理装置及び管理方法 |
CN111213121A (zh) * | 2017-10-05 | 2020-05-29 | 惠普发展公司,有限责任合伙企业 | 依据基于应用使用聚类选择计算机配置 |
US11601378B2 (en) | 2020-07-20 | 2023-03-07 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for allocating server resource, electronic device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
JPWO2016110950A1 (ja) | 2017-09-21 |
US10459768B2 (en) | 2019-10-29 |
US20170351553A1 (en) | 2017-12-07 |
JP6480955B2 (ja) | 2019-03-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10140066B2 (en) | Smart partitioning of storage access paths in shared storage services | |
WO2017072933A1 (ja) | 計算機システムの管理システム及び管理方法 | |
US11936731B2 (en) | Traffic priority based creation of a storage volume within a cluster of storage nodes | |
JP6480955B2 (ja) | 計算機システム、管理システム、及び、リソース管理方法 | |
US10248460B2 (en) | Storage management computer | |
WO2013160944A1 (en) | Provisioning of resources like cpu or virtual machines or virtualized storage in a cloud computing environment | |
TW201943241A (zh) | 光纖盒、分配系統資源的方法與系統 | |
JP2015001828A (ja) | 割当プログラム、割当装置および割当方法 | |
JP2015520876A (ja) | 情報記憶システム及び情報記憶システムの制御方法 | |
US11307802B2 (en) | NVMe queue management multi-tier storage systems | |
US10983708B2 (en) | Sharing alias addresses among logical devices | |
JP5996098B2 (ja) | ストレージデバイスの高速アクセス及びデータ保護を実現する計算機、計算機システム、及びi/o要求処理方法 | |
JP6448779B2 (ja) | サーバストレージシステムを含んだ計算機システム | |
WO2015056332A1 (ja) | ストレージ装置及び制御方法 | |
US9690610B2 (en) | Computer system and management computer controlling method | |
JP6244496B2 (ja) | サーバストレージシステムの管理システム及び管理方法 | |
WO2018173300A1 (ja) | I/o制御方法およびi/o制御システム | |
JP6231700B2 (ja) | サーバストレージシステムと管理システムを有する計算機システム |
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: 15876828 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016568201 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15541132 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15876828 Country of ref document: EP Kind code of ref document: A1 |