WO2010024027A1 - 仮想サーバシステム及び物理サーバ選択方法 - Google Patents
仮想サーバシステム及び物理サーバ選択方法 Download PDFInfo
- Publication number
- WO2010024027A1 WO2010024027A1 PCT/JP2009/061602 JP2009061602W WO2010024027A1 WO 2010024027 A1 WO2010024027 A1 WO 2010024027A1 JP 2009061602 W JP2009061602 W JP 2009061602W WO 2010024027 A1 WO2010024027 A1 WO 2010024027A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- server
- physical
- servers
- virtual
- group
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
Definitions
- the present invention relates to a virtual server system capable of simultaneously operating a plurality of virtual servers on a plurality of physical servers, and a method for selecting a physical server as a migration destination of the virtual servers.
- Such a virtual server system in which a plurality of virtual servers operate on a plurality of physical servers is used in a large-scale computing environment such as a data center or a corporate information system.
- the management server that manages the virtual server system acquires load data of each physical server and each virtual server, and distributes the virtual server to each physical server by moving it to another physical server.
- a technology is described that reduces the power consumption of the entire virtual server system by aggregating the unused resources and turning off the power of the surplus physical servers.
- Non-Patent Document 1 when a virtual server is moved to another physical server and there are many physical servers that are destination candidates, the processing time required to determine the optimum destination However, there is a problem that the increase of Therefore, the movement timing of the virtual server is delayed, resource shortage and useless free resources are generated, and the operating efficiency of the virtual server system is lowered.
- an object of the present invention is to provide a virtual server system and a physical server selection method that can determine an optimal destination physical server of a virtual server in a shorter time even when there are many physical servers.
- a virtual server system of the present invention includes a plurality of physical servers capable of operating a plurality of virtual servers,
- the physical servers are classified into a plurality of groups based on a predetermined index, a number of physical servers determined in advance for each group are selected as representative servers, and the migration target virtual server is selected from each representative server for each group.
- An autonomous control server that determines a destination physical server and moves the virtual server to be moved to the determined destination physical server; Have
- the autonomous control server classifies a plurality of physical servers capable of operating a plurality of virtual servers into a plurality of groups based on a predetermined index, and represents a predetermined number of physical servers for each group.
- a list generation unit that selects a server, and includes a representative server for each group, and generates a candidate list indicating a physical server that is a migration destination candidate of the migration target virtual server;
- a destination determination unit that determines a destination physical server of the virtual server from the candidate list;
- a virtual server moving unit that moves the virtual server to be moved to a physical server that is the moving destination determined by the moving destination determining unit; Have
- the physical server selection method of the present invention is a physical server selection method for selecting a physical server that is a migration destination of a required virtual server among a plurality of physical servers capable of operating a plurality of virtual servers, Computer Classifying the physical servers into a plurality of groups based on a predetermined index; Select a predetermined number of physical servers for each group as representative servers, Determine the physical server of the migration destination virtual server from each representative server for each group, In this method, the virtual server to be moved is moved to the determined physical server.
- FIG. 1 is a block diagram illustrating a configuration example of a virtual server system according to the present invention.
- FIG. 2 is a block diagram showing a configuration of a list generation unit provided in the autonomous control server shown in FIG.
- FIG. 3 is a block diagram illustrating a configuration of a list generation unit provided in the autonomous control server of the first embodiment.
- FIG. 4 is a flowchart illustrating a processing procedure of the autonomous control server included in the virtual server system according to the first embodiment.
- FIG. 5 is a block diagram illustrating a configuration of a list generation unit provided in the autonomous control server of the second embodiment.
- FIG. 6 is a flowchart showing a processing procedure of the autonomous control server provided in the virtual server system of the second embodiment.
- FIG. 1 is a block diagram illustrating a configuration example of a virtual server system according to the present invention.
- FIG. 2 is a block diagram showing a configuration of a list generation unit provided in the autonomous control server shown in FIG.
- FIG. 3 is a block diagram
- FIG. 7 is a block diagram showing the configuration of the virtual server system of the third embodiment.
- FIG. 8 is a block diagram showing a state of the physical server accommodated in the server rack shown in FIG.
- FIG. 9 is a block diagram illustrating a configuration of a list generation unit provided in the autonomous control server of the third embodiment.
- FIG. 10 is a flowchart illustrating a processing procedure of the autonomous control server included in the virtual server system according to the third embodiment.
- FIG. 1 is a block diagram showing a configuration example of the virtual server system of the present invention
- FIG. 2 is a block diagram showing a configuration example of a list generation unit provided in the autonomous control server shown in FIG.
- the virtual server system of the present invention includes a plurality of physical servers 31 to 3m (m is a positive integer), and an autonomous control server 2 that manages each virtual server running on the physical servers 31 to 3m. It is the structure which has.
- the autonomous control server 2 and the physical servers 31 to 3m are connected via the communication network 1.
- codes 311 to 31n are assigned to virtual servers operating on the physical server 31, and codes 3m1 to 3mn are assigned to virtual servers operating on the physical server 3m (n is a positive integer).
- the physical servers 31 to 3m and the autonomous control server 2 are realized by a known computer that executes processing according to a program stored in a recording medium (magnetic disk, semiconductor memory, optical disk, etc.), for example.
- a recording medium magnetic disk, semiconductor memory, optical disk, etc.
- the autonomous control server 2 includes a resource information collection unit 21, a resource information storage unit 22, a list generation unit 23, a movement destination determination unit 24, and a virtual server movement unit 25.
- the resource information collection unit 21 receives, from the physical servers 31 to 3m and the virtual servers 311 to 3mn, respective loads (CPU usage rate, memory usage rate, etc.), input / output performance values (for example, input / output for the maximum data transfer rate) Data transfer rate ratio), input / output performance values (transmission rate, bandwidth used, etc.) of the communication control device are collected as resource information.
- loads CPU usage rate, memory usage rate, etc.
- input / output performance values for example, input / output for the maximum data transfer rate
- Data transfer rate ratio Data transfer rate ratio
- input / output performance values transmission rate, bandwidth used, etc.
- the resource information storage unit 22 holds the resource information for each physical server 31 to 3m collected by the resource information collection unit 21.
- Physical server resource information includes physical server 31 to 3m loads (CPU, memory, network I / O, hard disk I / O usage, etc.) collected at regular time intervals, pre-measurements or other Information such as the maximum processing performance of each physical server 31 to 3m and the storage location of each physical server 31 to 3m collected by the method (for example, measured by processing by benchmark software or the like) is included.
- the list generation unit 23 includes a group generation unit 231, a representative server determination unit 232, and a physical server list generation unit 233.
- the group generation unit 231 acquires the resource information of the physical servers 31 to 3m from the resource information storage unit 22, and classifies the physical servers 31 to 3m into a plurality of groups according to a predetermined index using the resource information.
- a grouping method for example, any value of the load (usage rate of the CPU, memory, network I / O, hard disk I / O, etc.) of the physical servers 31 to 3m is used, and each value in a predetermined range.
- a reference value is set for the usage rate of each resource (CPU, memory, network I / O, hard disk I / O, etc.), and each physical server is based on the reference value.
- the grouping method is not limited to the above-described method, and any method may be used as long as each physical server can be classified into a plurality of groups based on appropriately set indexes. However, it is assumed that no matter which group of physical servers is selected as the migration destination of the virtual server, the influence on the load balance of all the physical servers is small.
- the representative server determining unit 232 selects a predetermined number of physical servers as the representative server for each group.
- the number of representative servers in any group selected at this time may be zero.
- the physical server list generation unit 233 generates a list (candidate list) of physical servers composed of representative servers of each group.
- the migration destination determination unit 24 determines the migration destination of the virtual server by using the physical server included in the candidate list generated by the list creation unit 23 as the migration destination candidate of the virtual server.
- the migration destination of the virtual server may be determined using the method described in Section 4.2 and 4.3 of Non-Patent Document 1, for example.
- the method of determining the migration destination of the virtual server is not limited to the method described in Non-Patent Document 1, and any method may be used as long as the optimum physical server of the virtual server can be determined from the migration destination candidates. .
- the virtual server migration unit 25 moves the migration target virtual server to the migration destination physical server selected by the migration destination determination unit 24.
- the resources of the physical servers 31 to 3m are stored from the resource information storage unit 22 by the group generation unit 231 provided in the list generation unit 23 of the autonomous control server 2.
- Information is acquired, and physical servers 31 to 3m are classified into a plurality of groups based on resource information according to a predetermined index.
- the representative server determination unit 232 of the list generation unit 23 selects a predetermined number of physical servers as representative servers for each group, and sets each representative server for each group as a destination candidate to be a virtual server to be moved. Determine the destination.
- Non-Patent Document 1 in the virtual server system, when free resources distributed on each physical server are aggregated, the destination of the virtual server is not only the usage rate of the CPU, for example, An evaluation value is calculated in consideration of the usage amount or the usage rate (throughput) of the network I / O or hard disk I / O, and is determined based on the calculated evaluation value. Therefore, for example, even if the physical servers 31 to 3m are classified into a plurality of groups for each predetermined load and each free resource amount, the physical server in the group with the largest load or the physical server in the group with the smallest free resource amount It is not always the optimal destination of the virtual server.
- a large number of physical servers are classified into a plurality of groups according to a predetermined index, and then representative servers selected from each group based on the same index as the grouping or different indices are selected. It is assumed that the virtual server is a migration destination candidate. As described above, even if the migration destination of the virtual server is determined from each representative server for each group, the virtual server is compared with the process of determining the migration destination of the virtual server from all the physical servers except the migration source of the virtual server. The operational efficiency of the virtual server system after server movement hardly deteriorates.
- a plurality of physical servers are classified into a plurality of groups based on an index such as a load, and a predetermined number of representative servers are selected for each group as a movement destination candidate. Since the number of physical servers that are candidates for the migration destination can be reduced, the processing time required to determine the physical server that is the migration destination of the virtual server can be shortened.
- the virtual server system according to the first embodiment is an example in which the number of candidates is reduced by generating a virtual server migration destination candidate list based on the load information of each physical server.
- FIG. 3 is a block diagram showing a configuration of a list generation unit provided in the autonomous control server of the first embodiment
- FIG. 4 is a flowchart showing a processing procedure of the autonomous control server provided in the virtual server system of the first embodiment.
- the configuration of the virtual server system is the same as that shown in FIG.
- the autonomous control server 2 uses the load (CPU, memory, network I / O, hard disk I / O, etc.) as resource information of the physical servers 31 to 3m and virtual servers 311 to 3mn in advance by the resource information collection unit 21. Rate) information.
- the virtual server to be moved is assumed to be a virtual server 311 operating on the physical server 31 shown in FIG.
- the list generation unit 23 of the first embodiment includes a load group generation unit 231a, a load group representative server determination unit 232a, and a physical server list generation unit 233.
- the load group generation unit 231a obtains information on the loads (usage rates of the CPU, memory, network I / O, hard disk I / O, etc.) of the physical servers 31 to 3m from the resource information storage unit 22, respectively. Based on the information, the physical servers 31 to 3m are classified into a plurality of groups.
- the load representative server determination unit 232a selects a predetermined number of physical servers as the representative server for each group generated by the load group generation unit 231a.
- the physical server list generation unit 233 generates a list (candidate list) of physical servers composed of the representative servers of each group selected by the load representative server determination unit 232a.
- the autonomous control server 2 first acquires the load information of the physical servers 31 to 3m from the resource information storage unit 22 by the load group generation unit 231a of the list generation unit 23 (step A1). .
- the load group generation unit 231a classifies the physical servers 31 to 3m into a plurality of groups based on the respective loads (step A2). At this time, the physical server 31 that is the migration source of the virtual server 311 to be migrated is excluded from the group classification elements.
- the load group generation unit 231a may classify the physical servers 32 to 3m into a plurality of groups using the usage rates of individual resources such as a CPU and a memory, and calculate values calculated based on the usage rates of the plurality of resources. You may use and classify into a plurality of groups. Furthermore, the load group generation unit 231a may classify the physical servers 32 to 3m into a plurality of groups for each usage rate of a predetermined range of resources, or may classify into groups by other methods.
- the autonomous control server 2 selects a predetermined number of physical servers as representative servers for each group by the load group representative server determination means 232a.
- the load group representative server determination unit 232a sorts, for example, the physical servers in the group in order of decreasing load (resource usage rate is small), and determines the top three physical servers with the lowest loads as representative servers.
- the load group representative server determining unit 232a randomly selects a plurality of (for example, two) physical servers from the group without sorting the physical servers, and selects them as representative servers.
- the load group representative server determining unit 232a may determine a representative server for each group by other methods. At this time, the number of representative servers may be zero as a result of the selection process (step A3).
- the autonomous control server 2 generates a candidate list composed of representative servers of each group by the physical server list generation means 233 (step A4).
- the autonomous control server 2 uses the movement destination determination unit 24 as a movement destination candidate for the physical server included in the candidate list, and determines the movement destination physical server (for example, the physical server 3m) of the virtual server 311 that is the movement target. Determine (step A5).
- the destination physical server of the virtual server 311 may be determined using the method described in Non-Patent Document 1 (sections 4.2 and 4.3) as described above.
- the autonomous control server 2 causes the virtual server moving unit 25 to move the virtual server 311 to be moved to the destination physical server 3m determined by the destination determining unit 24 in Step A5 (Step A6).
- each physical server is classified into a plurality of groups based on the respective loads, and a predetermined number of representative servers are selected for each group as a destination candidate, The number of physical servers that are candidates for the migration destination of the virtual server can be reduced. Therefore, it is possible to reduce the time required for determining the physical server to which the virtual server is moved.
- FIG. 5 is a block diagram showing a configuration of a list generation unit provided in the autonomous control server of the second embodiment
- FIG. 6 is a flowchart showing a processing procedure of the autonomous control server provided in the virtual server system of the second embodiment.
- the virtual server system according to the second embodiment is an example in which the number of candidates is reduced by generating a virtual server migration destination candidate list based on the amount of free resources of each physical server, not the load information of each physical server. .
- the configuration of the virtual server system is the same as that shown in FIG.
- the autonomous control server 2 uses the resource information collection unit 21 in advance as resource information of the physical servers 31 to 3m and the virtual servers 311 to 3mn, and loads of the physical server and the virtual server (CPU, memory, Information on network I / O, hard disk I / O usage, etc.) and maximum processing performance of physical servers 31 to 3m collected by prior measurement or other methods (measured by benchmark software etc.) Suppose you are. Note that the virtual server to be moved is the virtual server 311 running on the physical server 31.
- the list generation unit of the second example is replaced with the load group generation unit 231a and the load group representative server determination unit 232a included in the list generation unit 23 of the first example shown in FIG.
- the configuration includes a free resource group generation unit 231b and a free resource group representative server determination unit 232b.
- the free resource group generation unit 231b obtains information on the load (usage rate of the CPU, memory, network I / O, hard disk I / O, etc.) and the maximum processing performance from the resource information storage unit 22 respectively. Then, the amount of free resources of the physical servers 31 to 3m is calculated using the load and processing performance values, and the physical servers 31 to 3m are classified into a plurality of groups based on the calculated amount of free resources.
- the free resource group representative server determining unit 232b selects a predetermined number of physical servers as representative servers for each group generated by the free resource group generating unit 231b.
- the physical server list generation unit 233 generates a list (candidate list) of physical servers composed of the representative servers of each group selected by the free resource group representative server determination unit 232b.
- the autonomous control server 2 first acquires the load information and the processing performance of each physical server 31 to 3m from the resource information storage unit 22 from the free resource group generation unit 231b of the list generation unit 23. (Step B1).
- the free server group generation unit 231b classifies the physical servers 32 to 3m into a plurality of groups based on the amount of free resources included in the load information (step B2). At this time, the physical server 31 that is the migration source of the virtual server 311 to be migrated is excluded from the group classification elements.
- the amount of free resources of the physical servers 32 to 3m depends on not only the resource usage rate of the physical servers 31 to 3m but also the processing performance of the physical servers 31 to 3m.
- the free resource group generation unit 231b may classify the physical servers 32 to 3m into a plurality of groups using the free resource amounts of individual resources such as a CPU and a memory, and calculate based on the free resource amounts of the plurality of resources. You may classify
- the autonomous control server 2 selects a predetermined number of physical servers as representative servers for each group by the free resource group representative server determination means 232b.
- the free resource group representative server determining unit 232b sorts, for example, the physical servers in the group in descending order of the free resource amount, and determines the top three physical servers with the large free resource amount as representative servers.
- the free resource group representative server determination unit 232b randomly selects a plurality of (for example, two) physical servers from the group without sorting the physical servers, and selects them as representative servers.
- the free resource group representative server determining unit 232b may determine a representative server for each group by other methods. At this time, the number of representative servers may be zero as a result of the selection process (step B3).
- the autonomous control server 2 generates a candidate list composed of representative servers of each group by the physical server list generation means 233 (step B4).
- the autonomous control server 2 uses the movement destination determination unit 24 as a movement destination candidate for the physical server included in the candidate list, and determines the movement destination physical server (for example, the physical server 3m) of the virtual server 311 that is the movement target. Determine (step B5).
- the destination physical server of the virtual server 311 may be determined using the method described in Non-Patent Document 1 (sections 4.2 and 4.3) as described above.
- the autonomous control server 2 causes the virtual server moving unit 25 to move the virtual server 311 to be moved to the destination physical server 3m determined by the destination determining unit 24 in Step B5 (Step B6).
- each physical server is classified into a plurality of groups based on the amount of free resources, and a predetermined number of representative servers are selected for each group as destination candidates. It is possible to reduce the number of physical servers that are virtual server migration destination candidates. Therefore, it is possible to reduce the time required to determine the physical server of the movement destination when moving the virtual server. (Third embodiment)
- a virtual server system according to a third embodiment will be described with reference to the drawings.
- the virtual server system of the third embodiment operates on server racks 41 to 4k in which a plurality of physical servers are stored, and physical servers (not shown) stored in the server racks 41 to 4k.
- the configuration includes an autonomous control server 2 that manages each virtual server.
- the autonomous control server 2 and the physical servers stored in the server racks 41 to 4k are connected via the communication network 1.
- the configuration of the autonomous control server 2 is the same as that of the virtual server system shown in FIG.
- each of the server racks 41 to 4k includes a plurality of server storage units, and a plurality of virtual servers operate on each physical server stored in the server storage unit.
- reference numerals 411 to 41l are assigned to the server storage units provided in the server rack 41, and reference signs 4k1 to 4kl are assigned to the server storage units provided in the server rack 4k (k and l are positive integers).
- reference numerals 4111 to 411m are assigned to the physical servers stored in the server storage unit 411
- reference numerals 41l1 to 41lm are assigned to the physical servers stored in the server storage unit 41l, and stored in the server storage unit 4k1.
- 4k11 to 4k1m are assigned to the physical servers, and 4kl to 4klm are assigned to the physical servers stored in the server storage unit 4kl (m is a positive integer). Further, in FIG. 8, reference numerals 4klm1 to 4klmn are assigned to virtual servers running on the physical server 4klm.
- the physical servers 4111 to 4klm are realized by a known computer that executes processing according to a program stored in a recording medium (magnetic disk, semiconductor memory, optical disk, etc.), for example.
- a recording medium magnetic disk, semiconductor memory, optical disk, etc.
- the list generation unit 23 of the third embodiment includes a free resource group generation unit 231b, a server rack group representative server determination unit 232c, and a physical server list generation unit 233.
- the free resource group generation unit 231b obtains information on the load (usage rate of the CPU, memory, network I / O, hard disk I / O, etc.) and the maximum processing performance from the resource information storage unit 22 respectively. Then, the amount of free resources of the physical servers 31 to 3m is calculated using the load and processing performance values, and the physical servers 31 to 3m are classified into a plurality of groups based on the calculated amount of free resources.
- the server rack group representative server determining unit 232c selects a predetermined number of physical servers as the representative server for each group generated by the free resource group generating unit 231b.
- the physical server list generation unit 233 generates a list (candidate list) of physical servers composed of the representative servers of each group selected by the free resource group representative server determination unit 232b.
- FIG. 10 is a flowchart showing a processing procedure of the autonomous control server provided in the virtual server system of the third embodiment.
- the virtual server system groups each physical server based on the amount of available resources, and sets a physical server (representative server) that is a virtual server migration destination candidate based on the installation location in the server rack. This is an example of selection.
- the autonomous control server 2 is configured so that the resource information collection unit 21 previously collects resource information of the physical servers 4111 to 4klm and virtual servers 41111 to 4klmn for each physical server and virtual server collected at regular time intervals. Information on the load (CPU, memory, network I / O, hard disk I / O usage, etc.), and each physical server 4111-4klm collected by prior measurement or other methods (measured by processing with benchmark software etc.) It is assumed that the maximum processing performance is collected.
- the physical server installation locations in the server racks 41 to 4k are specified as follows.
- the installation location P of the physical servers can be defined by the following coordinates. .
- the virtual server to be moved is the virtual server 44321 on the physical server 4432, and the installation location of the physical server 4432 is P (4, 3, 2).
- the autonomous control server 2 first acquires the load information and processing performance of each physical server 4111 to 4 klm from the resource information storage unit 22 by the free resource group generation unit 231 b of the list generation unit 23. (Step C1).
- the free resource group generation unit 231b acquires the load information of each physical server 4111 to 4klm
- the physical server 4111 to 4klm is classified into a plurality of groups based on the free resource amount included in the load information (step C2).
- the movement source physical server 4432 of the virtual server 44321 to be moved is excluded from the group classification elements.
- the free resource amounts of the physical servers 4111 to 4klm are calculated in the same manner as in the second embodiment.
- the free resource group generation unit 231b may classify the physical servers 4111 to 4klm into a plurality of groups using the free resource amounts of individual resources such as a CPU and a memory, and calculate based on the free resource amounts of the plurality of resources. You may classify
- the autonomous control server 2 selects a predetermined number of physical servers as representative servers for each group by the server rack group representative server determination means 232c.
- the server rack group representative server determining unit 232c determines, for example, each physical server in the group in ascending order of the server rack number, the server storage unit number, and the physical server number that stores the physical server, that is, P (x, y, z ) In ascending order of x, y, z, and a representative server is selected from physical servers stored in a server rack having a rack number smaller than the number of the server rack storing the physical server 4432 that is the movement source.
- the server rack group representative server determining means 232c uses the number of the server storage unit in which the physical server 4432 as the movement source is installed and the number of the physical server in the same procedure as the selection procedure using the server rack number described above.
- a representative server may be selected (step C3).
- a predetermined number for example, two
- the number of representative servers selected for each group may be zero as a result of the selection process.
- the autonomous control server 2 generates a candidate list composed of representative servers of each group by the physical server list generation means 233 (step C4).
- the movement destination determination unit 24 uses the physical server included in the candidate list as the movement destination candidate, and selects the movement destination physical server (for example, the physical server 4111) of the virtual server 44321 that is the movement target. Determine (step C5).
- the destination physical server of the virtual server 44321 may be determined using the method described in Non-Patent Document 1 (sections 4.2 and 4.3) as described above.
- the autonomous control server 2 causes the virtual server moving unit 25 to move the virtual server 44321 to be moved to the destination physical server 4111 determined by the destination judging unit 24 in Step C5 (Step C6).
- each physical server is classified into a plurality of groups based on the amount of free resources.
- a plurality of physical servers are allocated based on the load on each physical server. You may categorize into groups.
- each physical server is classified into a plurality of groups based on the amount of free resources, and a predetermined number of representative servers are selected for each group based on the installation location of the physical server. Similar to the first embodiment and the second embodiment, the number of physical servers that are virtual server migration destination candidates can be reduced, and therefore the time required to determine the physical server that is the migration destination when the virtual server is migrated can be shortened.
- the resources of the physical server can be aggregated. Can be suppressed.
- the power consumption of the virtual server system can be reduced because the power supply of the physical server in which the number of operating virtual servers becomes zero, the server storage unit storing them, or the server rack can be turned off.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Transfer Between Computers (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
前記物理サーバを所定の指標に基づいて複数のグループに分類し、前記グループ毎に予め決められた台数の物理サーバを代表サーバとして選択し、前記グループ毎の各代表サーバから移動対象の仮想サーバの移動先の物理サーバを決定し、該決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する自律制御サーバと、
を有する。
前記候補リストから前記仮想サーバの移動先の物理サーバを決定する移動先判断部と、
前記移動先判断部で決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する仮想サーバ移動部と、
を有する。
コンピュータが、
前記物理サーバを所定の指標に基づいて複数のグループに分類し、
前記グループ毎に予め決められた台数の物理サーバを代表サーバとして選択し、
前記グループ毎の各代表サーバから移動対象の仮想サーバの移動先の物理サーバを決定し、
該決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する方法である。
次に第1実施例の仮想サーバシステムについて図面を用いて説明する。
(第2実施例)
次に第2実施例の仮想サーバシステムについて図面を用いて説明する。
ここで、Rは物理サーバの空きリソース量、Uは物理サーバのリソースの使用率、Pは物理サーバの最大の処理性能を示している。空きリソースグループ生成手段231bは、CPU、メモリ等の個別のリソースの空きリソース量を用いて物理サーバ32~3mを複数のグループに分類してもよく、複数のリソースの空きリソース量に基づいて算出した値を用いて複数のグループに分類してもよい。さらに、空きリソースグループ生成手段231bは、所定範囲の空きリソース量毎に物理サーバ32~3mを複数のグループに分類してもよく、その他の手法でグループに分類してもよい。
(第3実施例)
次に第3実施例の仮想サーバシステムについて図面を用いて説明する。
なお、移動対象となる仮想サーバは、物理サーバ4432上の仮想サーバ44321とし、物理サーバ4432の設置場所はP(4,3,2)とする。
Claims (12)
- 複数の仮想サーバが稼動可能な複数の物理サーバと、
前記物理サーバを所定の指標に基づいて複数のグループに分類し、前記グループ毎に予め決められた台数の物理サーバを代表サーバとして選択し、前記グループ毎の各代表サーバから移動対象の仮想サーバの移動先の物理サーバを決定し、該決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する自律制御サーバと、
を有する仮想サーバシステム。 - 前記所定の指標は、
前記物理サーバの負荷である請求項1記載の仮想サーバシステム。 - 前記所定の指標は、
前記物理サーバの空きリソース量である請求項1記載の仮想サーバシステム。 - 前記自律制御サーバは、
前記物理サーバの設置場所に基づいて前記グループ毎の代表サーバを選択する請求項1から3のいずれか1項記載の仮想サーバシステム。 - 複数の仮想サーバが稼動可能な複数の物理サーバを所定の指標に基づいて複数のグループに分類し、前記グループ毎に予め決められた台数の物理サーバを代表サーバとして選択し、前記グループ毎の各代表サーバから成る、移動対象の仮想サーバの移動先候補の物理サーバを示す候補リストを生成するリスト生成部と、
前記候補リストから前記仮想サーバの移動先の物理サーバを決定する移動先判断部と、
前記移動先判断部で決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する仮想サーバ移動部と、
を有する自律制御サーバ。 - 前記リスト生成部は、
前記物理サーバを、それぞれの負荷に基づいて複数のグループに分類する請求項5記載の自律制御サーバ。 - 前記リスト生成部は、
前記物理サーバを、それぞれの空きリソース量に基づいて複数のグループに分類する請求項5記載の自律制御サーバ。 - 前記リスト生成部は、
前記物理サーバの設置場所に基づいて前記グループ毎の代表サーバを選択する請求項5から7のいずれか1項記載の自律制御サーバ。 - 複数の仮想サーバが稼動可能な複数の物理サーバのうち、所要の仮想サーバの移動先となる物理サーバを選択するための物理サーバ選択方法であって、
コンピュータが、
前記物理サーバを所定の指標に基づいて複数のグループに分類し、
前記グループ毎に予め決められた台数の物理サーバを代表サーバとして選択し、
前記グループ毎の各代表サーバから移動対象の仮想サーバの移動先の物理サーバを決定し、
該決定した移動先の物理サーバへ前記移動対象の仮想サーバを移動する物理サーバ選択方法。 - 前記所定の指標は、
前記物理サーバの負荷情報である請求項9記載の物理サーバ選択方法。 - 前記所定の指標は、
前記物理サーバの空きリソース量である請求項9記載の物理サーバ選択方法。 - 前記物理サーバの設置場所に基づいて前記グループ毎の代表サーバを選択する請求項9から11のいずれか1項記載の物理サーバ選択方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/061,440 US8966038B2 (en) | 2008-08-28 | 2009-06-25 | Virtual server system and physical server selection method |
JP2010526609A JPWO2010024027A1 (ja) | 2008-08-28 | 2009-06-25 | 仮想サーバシステム及び物理サーバ選択方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008219871 | 2008-08-28 | ||
JP2008-219871 | 2008-08-28 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010024027A1 true WO2010024027A1 (ja) | 2010-03-04 |
Family
ID=41721200
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/061602 WO2010024027A1 (ja) | 2008-08-28 | 2009-06-25 | 仮想サーバシステム及び物理サーバ選択方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US8966038B2 (ja) |
JP (1) | JPWO2010024027A1 (ja) |
WO (1) | WO2010024027A1 (ja) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011232916A (ja) * | 2010-04-27 | 2011-11-17 | Hitachi Ltd | 計算機システム及び管理計算機 |
JP2012108655A (ja) * | 2010-11-16 | 2012-06-07 | Hitachi Ltd | シンクライアント制御システム |
WO2013042269A1 (ja) * | 2011-09-22 | 2013-03-28 | 富士通株式会社 | 電源管理装置、電源管理方法および電源管理プログラム |
WO2013084332A1 (ja) * | 2011-12-08 | 2013-06-13 | 株式会社日立製作所 | 仮想計算機の制御方法及び仮想計算機システム |
US20130227144A1 (en) * | 2010-09-24 | 2013-08-29 | Hitachi, Ltd. | Operation managing method for computer system, computer system and computer-readable storage medium having program thereon |
WO2016088163A1 (ja) * | 2014-12-01 | 2016-06-09 | 株式会社日立製作所 | 計算機システム及びリソース管理方法 |
WO2018135329A1 (ja) * | 2017-01-19 | 2018-07-26 | 富士通株式会社 | 管理装置、制御方法、および管理プログラム |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5541289B2 (ja) * | 2009-10-09 | 2014-07-09 | 日本電気株式会社 | 仮想サーバシステム、その自律制御サーバ、そのデータ処理方法およびコンピュータプログラム |
WO2012117471A1 (ja) * | 2011-03-01 | 2012-09-07 | 日本電気株式会社 | 仮想サーバシステム、管理サーバ装置及びシステム管理方法 |
DE102012217202B4 (de) | 2011-10-12 | 2020-06-18 | International Business Machines Corporation | Verfahren und System zum Optimieren des Platzierens virtueller Maschinen in Cloud-Computing-Umgebungen |
US9477529B2 (en) * | 2012-06-20 | 2016-10-25 | International Business Machines Corporation | Job distributed within a grid environment using mega-host groupings of execution hosts based on resource attributes |
US9130880B2 (en) * | 2012-07-11 | 2015-09-08 | Hitachi, Ltd. | Management system and information acquisition method |
US10013261B2 (en) * | 2012-09-10 | 2018-07-03 | Intel Corporation | Techniques for managing or controlling computing devices |
US10083051B1 (en) * | 2014-04-11 | 2018-09-25 | VCE IP Holding Company LLC | System, method, and code for classifying resources of a virtual computing environment |
US9946564B2 (en) * | 2015-06-23 | 2018-04-17 | International Business Machines Corporation | Adjusting virtual machine migration plans based on alert conditions related to future migrations |
US20170344397A1 (en) * | 2016-05-27 | 2017-11-30 | International Business Machines Corporation | Distributing work in a streaming application to computer systems according to system resources |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002259409A (ja) * | 2001-03-01 | 2002-09-13 | Nippon Telegr & Teleph Corp <Ntt> | 情報抽出方法、情報抽出装置、コンピュータ読取可能な記憶媒体およびコンピュータ・プログラム |
WO2008062864A1 (en) * | 2006-11-24 | 2008-05-29 | Nec Corporation | Virtual machine locating system, virtual machine locating method, program, virtual machine management device and server |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3320665B2 (ja) | 1998-12-03 | 2002-09-03 | 住友軽金属工業株式会社 | 熱間圧延における圧延順決定方法および圧延順決定装置 |
US7143170B2 (en) * | 2003-04-30 | 2006-11-28 | Akamai Technologies, Inc. | Automatic migration of data via a distributed computer network |
US20060004909A1 (en) * | 2004-04-30 | 2006-01-05 | Shinya Takuwa | Server system and a server arrangement method |
US20060069761A1 (en) * | 2004-09-14 | 2006-03-30 | Dell Products L.P. | System and method for load balancing virtual machines in a computer network |
WO2006043320A1 (ja) * | 2004-10-20 | 2006-04-27 | Fujitsu Limited | アプリケーション管理プログラム、アプリケーション管理方法、およびアプリケーション管理装置 |
JP2006154991A (ja) | 2004-11-26 | 2006-06-15 | Fujitsu Ltd | 情報処理システム、情報処理システムの制御方法、監視装置、監視プログラム、保守管理プログラム |
US7587492B2 (en) * | 2005-07-29 | 2009-09-08 | Hewlett-Packard Development Company, L.P. | Dynamic performance management for virtual servers |
US20070204266A1 (en) * | 2006-02-28 | 2007-08-30 | International Business Machines Corporation | Systems and methods for dynamically managing virtual machines |
JP4519098B2 (ja) | 2006-03-30 | 2010-08-04 | 株式会社日立製作所 | 計算機の管理方法、計算機システム、及び管理プログラム |
US7640023B2 (en) * | 2006-05-03 | 2009-12-29 | Cisco Technology, Inc. | System and method for server farm resource allocation |
US20070271560A1 (en) * | 2006-05-18 | 2007-11-22 | Microsoft Corporation | Deploying virtual machine to host based on workload characterizations |
JP4800837B2 (ja) * | 2006-05-22 | 2011-10-26 | 株式会社日立製作所 | 計算機システム、その消費電力低減方法、及びそのプログラム |
JP4240062B2 (ja) * | 2006-05-31 | 2009-03-18 | 日本電気株式会社 | 計算機システムおよび性能計測方法ならびに管理サーバ装置 |
JP4609380B2 (ja) | 2006-05-31 | 2011-01-12 | 日本電気株式会社 | 仮想サーバ管理システムおよびその方法ならびに管理サーバ装置 |
JP2008140240A (ja) | 2006-12-04 | 2008-06-19 | Hitachi Ltd | 仮想サーバ分散配置方法 |
US8019812B2 (en) * | 2007-04-13 | 2011-09-13 | Microsoft Corporation | Extensible and programmable multi-tenant service architecture |
US7941510B1 (en) * | 2007-10-11 | 2011-05-10 | Parallels Holdings, Ltd. | Management of virtual and physical servers using central console |
-
2009
- 2009-06-25 JP JP2010526609A patent/JPWO2010024027A1/ja active Pending
- 2009-06-25 WO PCT/JP2009/061602 patent/WO2010024027A1/ja active Application Filing
- 2009-06-25 US US13/061,440 patent/US8966038B2/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002259409A (ja) * | 2001-03-01 | 2002-09-13 | Nippon Telegr & Teleph Corp <Ntt> | 情報抽出方法、情報抽出装置、コンピュータ読取可能な記憶媒体およびコンピュータ・プログラム |
WO2008062864A1 (en) * | 2006-11-24 | 2008-05-29 | Nec Corporation | Virtual machine locating system, virtual machine locating method, program, virtual machine management device and server |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011232916A (ja) * | 2010-04-27 | 2011-11-17 | Hitachi Ltd | 計算機システム及び管理計算機 |
US9397953B2 (en) * | 2010-09-24 | 2016-07-19 | Hitachi, Ltd. | Operation managing method for computer system, computer system and computer-readable storage medium having program thereon |
US20130227144A1 (en) * | 2010-09-24 | 2013-08-29 | Hitachi, Ltd. | Operation managing method for computer system, computer system and computer-readable storage medium having program thereon |
JP2012108655A (ja) * | 2010-11-16 | 2012-06-07 | Hitachi Ltd | シンクライアント制御システム |
US9274593B2 (en) | 2011-09-22 | 2016-03-01 | Fujitsu Limited | Change the destination physical machine information such that the destination candidate of the certain virtual machine includes at least a physical machine belonging to the second physical machine group |
WO2013042269A1 (ja) * | 2011-09-22 | 2013-03-28 | 富士通株式会社 | 電源管理装置、電源管理方法および電源管理プログラム |
JPWO2013042269A1 (ja) * | 2011-09-22 | 2015-03-26 | 富士通株式会社 | 電源管理装置、電源管理方法および電源管理プログラム |
JPWO2013084332A1 (ja) * | 2011-12-08 | 2015-04-27 | 株式会社日立製作所 | 仮想計算機の制御方法及び仮想計算機システム |
US9223604B2 (en) | 2011-12-08 | 2015-12-29 | Hitachi, Ltd. | Control method of virtual machine and virtual machine system |
WO2013084332A1 (ja) * | 2011-12-08 | 2013-06-13 | 株式会社日立製作所 | 仮想計算機の制御方法及び仮想計算機システム |
WO2016088163A1 (ja) * | 2014-12-01 | 2016-06-09 | 株式会社日立製作所 | 計算機システム及びリソース管理方法 |
WO2018135329A1 (ja) * | 2017-01-19 | 2018-07-26 | 富士通株式会社 | 管理装置、制御方法、および管理プログラム |
JP2018116556A (ja) * | 2017-01-19 | 2018-07-26 | 富士通株式会社 | 管理装置、制御方法、および管理プログラム |
US11119815B2 (en) | 2017-01-19 | 2021-09-14 | Fujitsu Limited | Management apparatus, control method of calculation resources, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
JPWO2010024027A1 (ja) | 2012-01-26 |
US8966038B2 (en) | 2015-02-24 |
US20110161483A1 (en) | 2011-06-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2010024027A1 (ja) | 仮想サーバシステム及び物理サーバ選択方法 | |
CN104951425B (zh) | 一种基于深度学习的云服务性能自适应动作类型选择方法 | |
CN107734052B (zh) | 面向组件依赖的负载均衡容器调度方法 | |
US8380960B2 (en) | Data allocation and replication across distributed storage system | |
Sethi et al. | RecShard: statistical feature-based memory optimization for industry-scale neural recommendation | |
Zhang et al. | Virtual machine placement strategy using cluster-based genetic algorithm | |
CN107291539B (zh) | 基于资源重要程度的集群程序调度方法 | |
WO2021126439A1 (en) | Selecting a normalized form for conversion of a query expression | |
KR102347375B1 (ko) | 복수의 예측 모델을 기초로 가상 머신 워크 로드 예측 방법 | |
Hu et al. | Improved heuristic job scheduling method to enhance throughput for big data analytics | |
Magotra et al. | Adaptive computational solutions to energy efficiency in cloud computing environment using VM consolidation | |
Lu et al. | InSTechAH: Cost-effectively autoscaling smart computing hadoop cluster in private cloud | |
Yang et al. | An end-to-end and adaptive i/o optimization tool for modern hpc storage systems | |
CN116610731B (zh) | 一种大数据分布式存储方法、装置、电子设备及存储介质 | |
CN103324577B (zh) | 基于最小化io访问冲突和文件分条的大规模分条文件分配*** | |
CN111083201B (zh) | 一种工业物联网中数据驱动制造服务的节能资源分配方法 | |
CN116647560A (zh) | 物联网计算机集群协调优化控制方法、装置、设备及介质 | |
CN110058942B (zh) | 基于层次分析法的资源分配***及方法 | |
Mirsadeghi et al. | PTRAM: A parallel topology-and routing-aware mapping framework for large-scale HPC systems | |
Zhang et al. | NADE: nodes performance awareness and accurate distance evaluation for degraded read in heterogeneous distributed erasure code-based storage | |
Namyar et al. | Solving {Max-Min} Fair Resource Allocations Quickly on Large Graphs | |
Sha et al. | Optimizing data placement of mapreduce on ceph-based framework under load-balancing constraint | |
CN111506394B (zh) | 一种Docker Swarm容器调度方法及调度*** | |
Chen et al. | Minimizing virtual machine migration probability for cloud environments | |
WO2013128836A1 (ja) | 仮想サーバ管理装置及び仮想サーバの移動先決定方法 |
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: 09809682 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010526609 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13061440 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: 09809682 Country of ref document: EP Kind code of ref document: A1 |