JP5675471B2 - Data center system management method, data center system, and management apparatus - Google Patents

Description

  The present invention relates to a data center system management method, a data center system, and a management apparatus.

  In recent years, in order to achieve effective use of computing resources, cost reduction, and on-demand use of resources in enterprise and organization data centers, physical resources are aggregated by using virtualization technology and logical Construction of a cloud computing platform (hereinafter simply referred to as cloud) is progressing. In the cloud, for example, a business unit of an organization such as a company (hereinafter referred to as a cloud user) accesses a portal screen called a self-service portal and applies for allocation of a logical resource such as a virtual machine. Application systems are constructed and managed using logical resources (hereinafter referred to as tenant management). On the other hand, for example, an IT department of an organization such as a company or a vendor (hereinafter referred to as a cloud administrator) performs construction and operation management of a cloud infrastructure, allocation of logical resources to a cloud user, etc. (hereinafter referred to as cloud management). Called). In this way, by separating cloud operation management from logical resource use and operation management, cloud users can concentrate on logical resource use and operation management without being aware of the physical location and status of logical resources. Cloud administrators are expected to be able to concentrate on cloud operation management without knowing the details of application systems built and operated using logical resources. In the cloud, a virtual machine can be moved from one physical host computer to another physical host computer by virtualization technology. As a result, the cloud administrator can flexibly level the usage of resources and loads in the cloud. Furthermore, as described in Non-Patent Document 1, a technique for automatically moving a virtual machine according to a situation such as a load on a host computer has been developed. Patent Document 1 also proposes a technique for moving a plurality of virtual machines.

  On the other hand, when the physical resources that can be arranged in one data center are limited, the cloud can be configured to be constructed across a plurality of data centers. An actual application system generally includes elements for controlling communication such as a load balancer, a firewall, and an intrusion prevention system in addition to a plurality of virtual machines and physical host computers (hereinafter referred to as a middle box). ). The middle box is not necessarily virtualized and may be shared between application systems in the data center. Under these circumstances, when cloud users build application systems on the cloud, depending on the availability of data center resources, the individual components that make up the application system must be located in different data centers. There can be. Further, even when the application system is initially configured in the same data center, the methods described in documents such as Non-Patent Document 1 and Patent Document 1 take into account the relationship between the application system or virtual machine that the virtual machine configures. In other words, a situation may occur in which individual virtual machines constituting an application system are relocated to different data centers. When data centers are located geographically away from each other, communication between individual virtual machines or host computers goes back and forth between networks with relatively large delays. There is a problem that the performance of the system deteriorates or data can be transferred to an area not permitted by laws and regulations. In addition, when multiple application systems share a middle box for each data center, the cloud user cannot determine which middle box should be used in an application system that spans multiple data centers. There is a problem that operation may be complicated.

  In order to respond to the performance, compliance issues, and middlebox usage issues caused by the installation of individual virtual machines that make up the application system across multiple data centers. Have been proposed. For example, a method of allowing a cloud user to specify a location (a data center or a region) where a virtual machine is arranged and a middle box to be used is actually used. For example, Non-Patent Document 2 describes that a cloud user can specify the location of a virtual machine. Non-Patent Document 3 describes that a cloud user can specify which load distribution device of which data center is used. As another method, a method of moving a virtual machine in consideration of an application state has been proposed. For example, in Patent Literature 2 and Patent Literature 3, when an application is migrated, a data center suitable for executing the application is automatically found according to the server resource, network resource, or storage resource required by the application. A method is described. Patent Document 4 describes a method for optimizing a virtual machine arrangement in consideration of a policy specific to each application in addition to a migration policy for the entire system.

JP 2009-116859 A JP 2009-134687 A US Patent No. 7801994 JP 2009-116852 A

"VMware vSphere TM", http://www.vmware.com/products/drs/ "Amazon Elastic Compute Cloud (Amazon EC2)", http://aws.amazon.com/ec/ec2/ "F5 Load Balancer User Manual", http://wiki.gogrid.com/wiki/index.php/F5_Load_Balancer_User_Manual

  In the method disclosed in Non-Patent Document 2 or Non-Patent Document 3, a cloud user can specify a specific virtual machine placement location or a specific location of a middle box. It does not provide means for optimizing the layout of the entire application system across multiple data centers based on the relationship with the components. For this reason, it is necessary to specify the location of all the components constituting the application system in order to optimize the entire system, which is complicated.

  Similarly, in the methods disclosed in Patent Document 2 and Patent Document 3, it is possible to find a data center as a migration destination of a single application, but a plurality of virtual machines, a plurality of physical host computers, It does not take into account a plurality of applications that operate, and an application system composed of middle boxes, and a virtual machine cannot be moved based on the relationship between individual components.

  In the method disclosed in Patent Document 4, application movement between sites is described as an application example of an application-specific policy, but whether to move between sites is determined only by the presence or absence of cooperation between applications. Therefore, an application system in which a plurality of applications cooperate cannot be configured across data centers.

  From the above, in the conventional data center linkage management, there has been a problem that the configuration of the application system cannot be appropriately optimized across a plurality of data centers.

In order to solve the above-described problems and other problems, an embodiment of the present invention includes a first data center in which an application system connected to an external device and providing a data processing service to the external device is constructed. the second and the data center a management method of a data center system comprising at least, the application system a plurality of components that are communicatively connected to each other with a configuration equivalent to the first data center And selecting a first component to be a movement candidate as a candidate to move to the second data center from among the plurality of components of the application system of the first data center When, an index indicating the requirements of communication between the plurality of components included in the application system That the coupling degree, and setting for each of the plurality of components, wherein said coupling degree that is set for said first component is a movement candidates, and the first data center the second A step of comparing and evaluating communication line information between data centers, which is information relating to a communication state of a communication line between data centers, and the result of the comparison and evaluation, the degree of coupling is superior to the communication line information between data centers. Determining the second component that is the component that needs to be moved to the second data center together with the first component, and selecting the candidate as the movement candidate, a portion of the plurality of components included in the application system of the first data center after moving to the second data center, the first Detase Detecting a change in communication line status between the data center and the second data center as the communication line information between the data centers, and regarding the communication line information between the data centers after the change, in the application system, Evaluating whether the degree of coupling set between the plurality of components in the first data center and the plurality of components in the second data center can be maintained; and Whether it is possible to move the first and second components of the application system in the second data center to the first data center, Resource conditions and previous information about the plurality of components included in the application system in a first data center If it is determined that it is impossible to move to the first data center and the data center is a data center other than the first data center and the second data center. Te, select the data center is a line situation capable of maintaining cohesion regarding the determination by said first and second components, the acceptance determination of the first and second components in the data center A data center system management method including a step of requesting implementation.

  According to an aspect of the present invention, performance requirements that a specific component in an application system should have, or between components without knowing the detailed configuration of an application system that spans multiple data centers. The component can be moved between data centers while satisfying the requirements to be maintained. Furthermore, it becomes possible to level the resource load and the utilization rate based on the actual communication status between the components.

FIG. 1 is a diagram showing a system configuration of a data center system 1000 according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a hardware and software configuration of the host computer 10. FIG. 3 is a diagram showing another hardware and software configuration of the host computer 10. FIG. 4 is a diagram illustrating a hardware and software configuration of the management server 11. FIG. 5 is a diagram illustrating a hardware and software configuration of the monitoring server 12. FIG. 6 is a diagram illustrating a hardware and software configuration of the middle box 13. FIG. 7 is a diagram illustrating a hardware and software configuration of the computer 3000. FIG. 8 is a diagram showing a logical configuration of the application system 2000. FIG. 9 is a diagram showing another logical configuration of the application system 2000. FIG. 10 is a diagram illustrating an example of the combination degree definition table 170. FIG. 11 is a diagram illustrating an example of the weight definition table 171. FIG. 12 is a diagram illustrating an example of the node management table 179. FIG. 13 is a diagram illustrating an example of the association degree management table 172. FIG. 14 is a diagram illustrating an example of the node weight information management table 173. FIG. 15 is a diagram illustrating an example of the middle box management table 174. FIG. 16 is a diagram illustrating an example of the service profile 175. FIG. 17 is a diagram illustrating an example of the node movement request profile 176. FIG. 18 is a diagram illustrating an example of the node movement confirmation screen 900. FIG. 19 is a diagram illustrating an example of a weight and coupling degree setting process procedure. FIG. 20 is a diagram illustrating an example of an overall processing procedure for node movement. FIG. 21 is a diagram illustrating an example of a procedure for verifying whether or not a node can be moved. FIG. 22 is a diagram illustrating an example of a procedure for verifying the weight and the degree of coupling. FIG. 23 is a diagram illustrating an example of an acceptance / rejection verification processing procedure. FIG. 24 is a diagram illustrating an example of a reconfiguration processing procedure of the application system 2000. FIG. 25 is a diagram illustrating an example of a coupling degree derivation process procedure according to the second embodiment. FIG. 26 is a diagram illustrating an example of the processing procedure of the association degree related setting in the second embodiment.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. In addition, this invention is not limited by this embodiment.

  FIG. 1 shows a system configuration example of a data center system 1000 according to an embodiment of the present invention. In this embodiment, the data center system 1000 includes a data center 1, a data center 2, a cloud usage terminal 3, a cloud management terminal 4, a client terminal 5 and a communication network 6. The terminal 3, the cloud management terminal 4, and the client terminal 5 are connected via a communication network 6. Each of the data center 1 and the data center 2 includes a host computer 10, a management server 11, a monitoring server 12, a middle box 13, a routing device 14, and a communication network 15, and includes the host computer 10, the management server 11, and the monitoring server 12. The middle box 13 and the routing device 14 are connected via a communication network 15. The cloud usage terminal 3, the cloud management terminal 4, and the client terminal 5 may be connected to the communication network 15 in the data center 1 or the data center 2.

  The cloud user accesses the management server 11 of the data center 1 or the data center 2 using the cloud usage terminal 3. The cloud administrator accesses the management server 11 of the data center 1 or the data center 2 using the cloud management terminal 4. The end user accesses the virtual machine built on the host computer 10 or the host computer 10 of the data center 1 or the data center 2 using the client terminal 3 which is an external device, and the application on the host computer 10 or the virtual machine. Is used.

  The communication network 6 and the communication network 15 can be realized by a public network, the Internet, ISDN, a dedicated line, a wired network such as a LAN, or a wireless network using facilities such as a mobile communication base station and a communication artificial satellite. In the communication network 4 and the communication network 15, each device is identified by identification information given in advance to each device, whereby each device is connected to another device for communication.

  The middle box 13 (communication service device) is installed to apply a communication control policy and a security policy to communication performed between devices such as a load balancer, a firewall, an intrusion detection system, and an intrusion prevention system. It is a device. The middle box 13 may be implemented as separate hardware, or may be implemented as software and operated by the host computer 10 or a virtual machine constructed on the host computer 10. The routing device 14 is for controlling communication by connecting communication networks. In this embodiment, the routing device 14 connects the communication network 6 and the communication network 15.

  Next, the host computer 10 of this embodiment will be described. FIG. 2 shows an example of the hardware and software configuration of the host computer 10. The host computer 10 includes a control unit 100 including a processor such as a CPU (Central Processing Unit) and an MPU (MicroProcessing Unit), a storage unit 101, a network interface unit 102 for connecting to the communication network 15, a display unit 103, and an input unit 104. , And a data bus 105 connecting them.

  The storage unit 101 is a volatile storage device including a semiconductor memory such as a RAM (Random Access Memory), or a readable / writable nonvolatile storage device such as a hard disk drive (hereinafter referred to as “HDD”) or an SSD (Solid State Drive), or a magneto-optical device. It can be composed of a read-only nonvolatile storage device such as a medium. The display unit 103 can be configured by a display device such as a CRT display or a liquid crystal display, and the input unit 104 can be configured by an input device such as a keyboard, a mouse, or a pen tablet.

  In the host computer 10, for example, arithmetic processing associated with execution of software is executed by the control unit 100. The program executed by the control unit 100 and the data used by the program may be stored in the storage unit 101, or may be transmitted from the communication network 15 or another device via a carrier wave or a digital signal propagating on the communication network 15. May be introduced. Further, the host computer 10 may be configured such that the display unit 33 or the input unit 34 is omitted.

  The virtual machine monitor program 160 stored in the storage unit 101 is a virtualization program that is implemented to operate one or more virtual machines 161 on the host computer 10. The virtual machine monitor program 160, for example, logically divides the control unit 100, the storage unit 101, and the network interface unit 102, so that one or more virtual machines 161 that behave in the same manner as the host computer 10 on the host computer 10. It is possible to operate. Further, the virtual machine monitor program 160 receives an instruction to move the virtual machine 161 from the management server 11 and performs a process of moving the virtual machine 161 with the virtual machine monitor program 160 on the destination host computer 10. For the configuration and operation of the virtual machine monitor program 160 and the movement process of the virtual machine 161, known techniques described in the above-mentioned prior art documents can be used.

  The storage unit 101 also stores an operating system (OS) that provides basic functions such as general computer data input / output processing and memory management, and is executed by the control unit 100. This configuration is the same for a management server 11, a monitoring server 12, a middle box 13, a cloud use terminal 3, a cloud management terminal 4, and a client terminal 6 described later.

  The application program 162 is a program for providing some data processing function to the client terminal 5, and can take the form of a program for realizing each function such as a web server, a mail server, an application server, and a database. is there.

  FIG. 3 shows another configuration example of the hardware and software of the host computer 10. The host computer 10 illustrated in FIG. 3 includes a control unit 100 including a CPU, a storage unit 101, a network interface unit 102 for connecting to the communication network 15, a display unit 103, and an input unit, as in the configuration example of FIG. 104 and a data bus 105 connecting them. The host computer 10 may have a configuration in which the display unit 103 or the input unit 104 is omitted. Unlike the configuration example of FIG. 2, the host computer 10 according to this configuration executes the application program 162 without using the virtual machine monitor program 160.

  In the present embodiment, the virtual machine 161, the host computer 10, or the middle box 13 to be described later is referred to as “node” (described later) unless there is a particular difference in configuration. These are collectively referred to as “components of the application system 2000”. In this embodiment, the cloud user can construct an application system including a plurality of nodes in the data center 1 or 2.

  Next, the management server 11 of this embodiment will be described. FIG. 4 shows a configuration example of hardware and software of the management server 11. Similar to the host computer 10 illustrated in FIG. 2 or 3, the management server 11 includes a control unit 110 including a CPU, a storage unit 111, a network interface unit 112 for connecting to a communication network 114, a display unit 113, an input The unit 114 and the data bus 115 connecting them can be configured. Further, the management server 11 may be configured such that the display unit 113 or the input unit 114 is omitted.

  The storage unit 111 stores programs for realizing the functions as the management server 11 of the present embodiment and tables used by these programs.

  First, the tables will be described. A configuration example of each table will be described later. The coupling degree definition table 170 is a table for defining the coupling degree between nodes, and stores the correspondence between the coupling degree and related information associated with the coupling degree. The degree of coupling is an index representing the strength of connection between nodes. In this embodiment, the degree of coupling is defined by, for example, a bandwidth and response speed required for communication executed between nodes. The quantitative correspondence between the degree of coupling, the bandwidth, and the response speed may be determined by the cloud user in advance, or the cloud user may use what is determined in advance by the cloud administrator. In this embodiment, the cloud user accesses the later-described degree-of-association setting program 177 via the cloud-use terminal 3, and the degree of association between the individual nodes of the application system is determined as the degree of association defined in the degree-of-association definition table 170. Use to specify.

  The weight definition table 171 is a table for defining weight information assigned to each node. The weight is an index representing the strength of the connection of the node to the place where the node is currently installed. The definition of the weight may be determined by the cloud user, or the cloud user may use what is determined in advance by the cloud administrator. In this embodiment, the weight is expressed as a threshold value for determining whether or not a node can move from the current location or a moving range. For example, as the weight becomes smaller, the movement range of nodes spreads within the same rack, within the same data center, within the same region, and so on. On the other hand, when the node is the physical host computer 10, the weight of the node becomes very large because it cannot be moved easily like the virtual machine 161. Also, depending on the type of data contained in the node, the movement of the node to another location can be regulated by laws and regulations applied at the installation location, so the weight of the node is also very large.

  The node management table 179 stores the correspondence relationship between the cloud user and the application system, the node included in the application system, the location where the node was originally installed, and the current installation location of the node. The location where the node was originally installed indicates the location where the node was originally placed. When a node is moved, the current location of the node indicates the destination.

  The coupling degree management table 172 includes a cloud user and an application system, a node included in the application system, a coupling degree between the node and another node, and requirements for the middle box 13 when the other node is the middle box 13. Are stored. The requirements that the node requests to the middle box 13 are defined as a service profile 175 described later. The contents of this table are set by the degree-of-association setting program 177 based on the result registered by the cloud user in the degree-of-association setting program 177 described later via the cloud use terminal 3. In this embodiment, the cloud management program 178 determines whether or not a node can be moved based on the degree of connection described in this table or the requirements for the middle box 13 as described later.

  The node weight information management table 173 stores correspondence relationships between cloud users, application systems, nodes included in application systems, and node weight information. The contents of this table are set by the coupling degree setting program 177 based on the result registered by the cloud user in the coupling degree setting program 177 described later via the cloud utilization terminal 3. In this embodiment, the cloud management program 178 determines whether or not a node can be moved or a movable range based on the weights described in this table, as will be described later.

  The middle box management table 174 stores a correspondence relationship between the middle box 13 and external requirements supported by the middle box 13. External requirements supported by the middle box 13 are defined as a service profile 175 described later. The contents of this table are set in advance by the cloud administrator via the cloud management terminal 4. The contents of the service profile 175 in this table may be set in advance by the cloud administrator via the cloud management terminal 4 or may be provided from the vendor of the middle box 13.

  The service profile 175 (service requirement information) stores information such as functions supported by the middle box 13 included in the application system, performance, and interfaces for cooperation with other nodes. The service profile 175 is used for a cloud user to define requirements for the middle box 13 or to define functions supported by the middle box 13. The service profile 175 can be described using a known technique such as XML.

  The node migration request profile 176 includes the number of migration candidate virtual machines 161, the amount of resources required for each virtual machine 161, the degree of connectivity required with other nodes, and the case where the other nodes are the middle box 13. Stores information such as requirements required by the node in the middle box 13. The requirements for the middle box 13 are defined as a service profile 175 described for the middle box 13. When the cloud administrator moves the virtual machine 161 between the data centers 1 and 2, the cloud management program 178 of the migration source data center 1 or 2 moves to the cloud management program 178 of the migration destination data center 1 or 2. The node migration request profile 176 is transmitted, and the cloud management program 178 of the migration destination data center 1 or 2 uses the node migration request profile 176 to determine whether or not the virtual machine 161 can be accepted.

  Next, a program for realizing the function of the management server will be described. First, the coupling degree setting program 177 is a program that provides an interface for setting a coupling degree between nodes and a node weight for the data center 1 or 2 to the cloud user via the cloud utilization terminal 3. In this embodiment, the connectivity setting program 177 receives the connectivity information and node weight information input by the cloud user via the cloud usage terminal 3 and stores them in the connectivity management table 172 and the node weight information management table 173.

  Next, the cloud management program 178 is a program for managing the movement of the virtual machine 161 between the host computers 10, and the virtual machine 161 between data centers is based on the degree of coupling between nodes and the weight of each node. Whether or not to move is determined. In determining whether or not to move, the cloud management program 178 determines the line status between the data centers, the resource availability of the destination data center, the status of the line in the destination data center, and the service of the middle box 13 in the destination data center. Also consider the level.

  Specifically, when the cloud administrator moves the virtual machine 161 from the host computer 10 of the data center 1 to the host computer 10 of the data center 2, the cloud administrator uses the cloud management terminal 4 to move the cloud of the data center 1. A virtual machine 161 that is a migration candidate is designated for the management program 178. Next, the cloud management program 178 determines whether or not the virtual machine 161 designated by the cloud administrator can be moved to the data center 2 based on the weight information set for each node. In addition, the cloud management program can move the virtual machine 161 to the data center 2 based on the degree of coupling with other virtual machines 161 belonging to the same application system as the virtual machine 161 and the line status between the data centers 1 and 2. It is determined whether there is a virtual machine 161 that needs to move with the virtual machine 161. Further, the cloud management program 178 uses the node movement request profile 176 to inquire whether the virtual machine 161 group can be accepted to the cloud management program 178 of the data center 2 and determines whether or not the movement is possible.

  In this embodiment, the cloud management program 178 issues an instruction to the virtual machine monitor program 160 to move the virtual machine 161 to another data center with the final judgment of the cloud administrator. The cloud management program 178 may be configured to automatically issue a movement instruction to the virtual machine monitor program 160 without the judgment of the cloud administrator.

  On the other hand, the cloud management program 178 operating in the data center 2 receives an inquiry for accepting the virtual machine 161 group from the cloud management program 178 of the data center 1 as the node movement request profile 176, and the requested resource amount described in the profile 176 Alternatively, the requested coupling degree is compared with the availability of the host computer 10 in the data center 2 and the availability of the link between the host computers 10. Further, the cloud management program 178 describes a service profile in which service contents supported by the middle box 13 are described as to whether the middle box 13 of the data center 2 can satisfy the service level required by the virtual machine 161. 175 and the service profile 175 requested by the virtual machine 161 are compared, and the acceptance is returned to the cloud management program 178 of the data center 1. The data processing by the combined soil setting program 177 and the cloud management program 178 will be described later with reference to a related processing flow example.

  Next, the monitoring server 12 of this embodiment will be described. FIG. 5 shows an example of the hardware and software configuration of the monitoring server 12. As with the host computer 10 or the management server 11, the monitoring server 12 includes a control unit 120 including a CPU, a storage unit 121, a network interface unit 122 for connecting to the communication network 15, a display unit 123, an input unit 124, A data bus 125 connecting them can be used. Further, the monitoring server 12 may be configured such that the display unit 123 and the input unit 124 are omitted.

  The storage unit 121 stores a network monitoring program 180 and a node monitoring program 181 that are executed by the control unit 120 to realize the function of the monitoring server 12. The network monitoring program 180 is a program for collecting information such as communication volume, operation status, failure information, and the like from each device provided on the communication network 15 in the data center 1 or 2. The collection of information from each device can be realized using a known technique such as SNMP (Simple Network Monitoring Protocol) or an information acquisition interface unique to each device. In this embodiment, the network monitoring program 180 collects the communication status between the nodes, the communication status between the node and the middle box 13, and the bandwidth status between the data centers from the node, the middle box 13 and the routing device 14. . Further, the network monitoring program 180 transmits the collected information to the cloud management program 178 in response to an inquiry from the cloud management program 178 in the management server 11.

  The node monitoring program 181 is a program for monitoring the resource availability of each node. The collection of information from each device can be realized by using a known technique such as SNMP or an information acquisition interface unique to each device. Further, the node monitoring program 181 transmits the collected information to the cloud management program 178 in response to an inquiry from the cloud management program 178 in the management server 11.

  Next, the middle box 13 of the present embodiment will be described. FIG. 6 shows a configuration example of hardware and software of the middle box 13. Similarly to the host computer 10 or the management server 11, the middle box 13 includes a control unit 130 including a CPU, a storage unit 131, a network interface unit 132 for connecting to the communication network 15, a display unit 133, an input unit 134, and A data bus 135 connecting them can be used. Further, the middle box 13 may be configured such that the display unit 133 and the input unit 134 are omitted.

  The storage unit 131 stores a program that realizes the function of the middle box 13 by being executed by the control unit 130, for example, an intrusion detection program 182 and a firewall program 183. As the intrusion detection program 182 and the firewall program 183, programs generally used for ensuring security in data processing via the communication network 15 can be applied. Similarly to the host computer 11 illustrated in FIG. 2, the middle box 13 may include a virtual machine 161 so that a program such as a virus detection program 182 is executed on each virtual machine 161.

  Next, a computer 3000 that is suitably applied as the cloud usage terminal 3, the cloud management terminal 4, and the client terminal 5 of the present embodiment will be described. FIG. 7 shows a configuration example of hardware and software of the computer 3000. Similar to the host computer 10 or the management server 11, the computer 3000 includes a control unit 3030 including a CPU, a storage unit 3031, a network interface unit 3032 for connecting to the communication network 15, a display unit 3033, an input unit 3034, and these Can be configured by a data bus 3035 for connecting the two.

  The storage unit 3031 stores a program that implements the functions of the cloud use terminal 3, the cloud management terminal 4, or the client terminal 5 by being executed by the control unit 3030, such as a user interface program 184 and an application program 1854. . The user interface program 184 is a program that provides a user interface for data input / output between the computer 3000 functioning as the cloud usage terminal 3, the cloud management terminal 4, or the client terminal 5 and other nodes, for example, a WWW browser program It is. The application program 185 is a program for providing other functions as the cloud usage terminal 3, the cloud management terminal 4, or the client terminal 5.

  Next, the configuration of the application system 2000 in the present embodiment will be described. FIG. 8 shows a logical configuration example of the application system 2000. For example, the application system 2000 built in the data center 1 by the cloud user can be configured by a plurality of nodes and a plurality of middle boxes 13. For example, in the case of a general web application service system, it can be configured from a node group that operates a web server, a node group that operates an application server, a node group that operates a database, and a middle box such as a firewall or a load balancer. In the configuration example illustrated in FIG. 8, the application system 2000 includes a plurality of virtual machines 202 to 207 and a plurality of host computers 208 and 209. Access from the client terminal 200 is distributed by the load balancer 201 to the virtual machines 202 to 204 serving as a web server. A virtual machine 202 and a virtual machine 203 that are web servers are connected to a virtual machine 205 and a virtual machine 206 that serve as an application server, respectively. The virtual machine 205 and the virtual machine 206 are connected to a host computer 208 and a host computer 209 that serve as a database, respectively. It is connected.

  On the other hand, the virtual machine 204 serving as a web server is connected to a virtual machine 207 serving as an application server, and the virtual machine 207 is connected to a host computer 209 serving as a database. The host computer 208 and the host computer 209 are connected to each other, and synchronization between databases is performed.

  Next, movement of a node in the application system 2000 will be described. FIG. 9 is a diagram that is disclosed in the present embodiment when the cloud administrator tries to move the virtual machine 207 in the application system 2000 having the configuration of FIG. 8 that is operating in the data center 1 to the data center 2. An example of a final logical configuration when movement is performed by the method is shown. Here, the degree of coupling between the virtual machine 207 and the virtual machine 204 specified by the cloud administrator, the degree between the virtual machine 204 and the load balancer 201 is high, and the degree of coupling between the virtual machine 207 and the host computer 209. Is determined to be lower than the line status between the data centers 1 and 2, the virtual machine 204 is moved together with the virtual machine 207. When the load balancer 210 in the data center 2 supports the same service profile 175 as the load balancer 201 used by the virtual machine 204 in the data center 1, the virtual machine 204 uses the load balancer 210 in the data center 2. It can be used. When the virtual machine 207 is moved from the data center 1 to the data center 2, as shown in FIG. 9, the application system 2000 is configured across the data center 1 and the data center 2 while maintaining the service level. Here, the distribution of access from the client terminal 200 to the load balancer 201 and the load balancer 210 can be realized using a known technique such as DNS or global load balancing.

  Next, tables used in this embodiment will be described. As described with reference to the management server 11 illustrated in FIG. 4, the tables are stored in the storage unit 111 of the management server 11. First, the coupling degree definition table 170 will be described. FIG. 8 is a diagram illustrating an example of the coupling degree definition table 170. In the coupling degree definition table 170, items of coupling degree 300, band 301, and response speed 302 are associated and stored. The degree of coupling 300 is a value indicating the degree of coupling between nodes, and is indicated by numerical indexes 1 to 5 in the example of FIG. A band 301 is a bandwidth necessary for satisfying the degree of coupling between nodes. The response speed 302 is an average response speed required to satisfy the degree of coupling between nodes.

  Next, the weight definition table 171 will be described. FIG. 11 is a diagram illustrating an example of the weight definition table 171. The weight definition table 171 stores the items of the weight 400, the movement range 401, and the data center 402 in association with each other. The weight 400 is a value indicating the strength of the connection of the node to the physical installation location of the node, and is represented by a numerical index from 0 to 5. The movement range 401 is an index indicating a range in which a node having the value of the weight 400 can move. The data center 402 is identification information for uniquely identifying a data center (DC) included in the movement range 401.

  Next, the node management table 179 will be described. FIG. 12 is a diagram illustrating an example of the node management table 179. The node management table 179 stores items of user ID 450, system ID 451, node ID 452, location 453, and current location 454 in association with each other. The user ID 450 is identification information for uniquely identifying a cloud user in the data center system 1000. The system ID 451 is identification information for uniquely identifying the application system 2000 constructed by the cloud user. The node ID 452 is identification information for uniquely identifying each node. In FIG. 12, the symbol “VM” indicates that the node is the virtual machine 161, and the symbol “Host” indicates that the node is the physical host computer 10. The location 453 is identification information for uniquely identifying the location where the node was originally installed. The current location 454 is identification information for uniquely identifying the location of the current node when the node moves to another data center by the method disclosed in the present embodiment.

  Next, the coupling degree management table 172 will be described. FIG. 13 is a diagram illustrating an example of the association degree management table 172. In the association degree management table 172, items of a user ID 500, a system ID 501, a node ID 502, a related node ID 503, an association degree 504, and a service profile 505 are recorded in association with each other. The user ID 500 is identification information for uniquely identifying a cloud user in the data center system 1000, and is the same as the user ID 450 in the node management table 179. The system ID 501 is identification information for uniquely identifying the application system 2000 constructed by the cloud user, and is the same as the system ID 451 in the node management table 179. The node ID 502 is identification information for uniquely identifying each node, and is the same as the node ID 452 in the node management table 179. The related node ID 503 is identification information for uniquely identifying another node belonging to the same application system 2000 as each node and the middle box 13 used by each node. The degree of association 504 is a value indicating the degree of association between the nodes indicated by the node ID 502 and the related node ID 503, and is the same as the degree of association 300 in the association degree definition table 170. The service profile 505 is identification information of the service profile 175 requested to the middle box 13 to which the node to be moved is related when the related node ID 503 is the middle box 13.

  Next, the node weight information management table 173 will be described. FIG. 14 is a diagram illustrating an example of the node weight information management table 173. In the node weight information management table 173, items of user ID 600, system ID 601, node ID 602, and weight 603 are recorded in association with each other. The user ID 600 is identification information for uniquely identifying a cloud user in the data center system 1000, and is the same as the user ID 450 in the node management table 179. The system ID 601 is identification information for uniquely identifying the application system 2000 constructed by the cloud user, and is the same as the system ID 451 in the node management table 179. The node ID 602 is identification information for uniquely identifying each node, and is the same as the node ID 452 in the node management table 179. The weight 603 is a value indicating the weight of each node, and is the same as the weight 400 of the weight definition table 171.

  Next, the middle box management table 174 will be described. FIG. 15 is a diagram illustrating an example of the middle box management table 174. In the middle box management table 174, a middle box ID 700 and a service profile ID 701 are recorded in association with each other. The middle box ID 700 is identification information for uniquely identifying the middle box 13 in the data center system 1000. The service profile ID 701 is identification information of the service profile 175 describing functions and the like supported by the middle box 13, and is the same as the service profile 505 of the coupling degree management table 172.

  Next, the service profile 175 will be described. FIG. 16 is a diagram illustrating an example of the service profile 175. In the present embodiment, the service profile 175 can be described using a structured language such as XML, which is a known technique. The service profile 175 describes service contents provided to the target node in association with the service profile ID 800. The service contents are described by being classified into items of function 801 and performance 802. The service profile ID 800 is identification information for uniquely identifying the service profile 175 in the data center system 1000 and is the same as the service profile ID 701 of the middle box management table 174. A function 801 is a descriptor for describing the function of the middle box 13. The example of FIG. 16 shows that the corresponding middle box 13 has a load distribution function and a health check function corresponding to layer 4. The performance 802 is a descriptor for describing the performance of the middle box 13.

  Next, the node movement request profile 176 will be described. FIG. 17 is a diagram illustrating an example of the node movement request profile 176. In the present embodiment, the node movement request profile 176 can be described using a structured language such as XML which is a publicly known technique. The node movement request profile 176 records items of a node number 811, a node ID 812, a related node number 813, a related node ID 814, a resource 815, a coupling degree 816, a band 817, a response speed 818, and a service profile ID 819.

  The number of nodes 811 is a descriptor for describing the number of nodes that are requested to be accepted by the movement destination data center in the node movement request profile 176. The node ID 812 is a descriptor for describing an identifier for uniquely identifying each node that is requested to be accepted by the destination data center. The related node number 813 is a descriptor for describing the number of nodes having some degree of connection with the node identified by the node ID 812. The related node ID 814 is a descriptor for describing the identifier of the node to which the node identified by the node ID 812 is related. The resource 815 is a descriptor for describing resource values related to an auxiliary storage device such as a CPU, memory, HDD, and network interface required by the node identified by the node ID 812. For example, the clock speed of the CPU and the memory The storage capacity of the auxiliary storage device is included. The degree of coupling 816 is a descriptor for describing the degree of coupling that should be satisfied between the node identified by the node ID 812 and the node identified by the related node ID 814 in accordance with the definition of the coupling degree definition table 170. A band 817 is a value of the band column 301 in a record in which the coupling degree column 300 matches the coupling degree 816 in the coupling degree definition table 170. The response speed 818 is a value of the response speed 302 in the record in which the combination degree column 300 matches the combination degree 816 in the combination degree definition table 170. In the node movement request profile 176, not only the degree of coupling 816 but also the corresponding bandwidth 817 and response speed 818 are described because the data center 1 and the data center 2 are not necessarily operated by the same business operator. This is because there is a possibility that the same combination degree definition table 170 is not held. The service profile ID 819 is a descriptor for describing an identifier of the service profile 175 in which requirements required for the middle box 13 are described when the node identified by the related node ID 814 is the middle box 13.

  Next, a user interface provided to the cloud administrator in this embodiment will be described. FIG. 18 is a diagram for illustrating an example of an interface for node movement management provided to the cloud administrator by the cloud management program 178, and is an example of a screen for requesting the cloud administrator to finally confirm node movement. This user interface is provided to the user through the display unit 1033 of the cloud management terminal 4 by the cloud management program 178 running on the management server 11. The screen 900 includes a display area 901 of a candidate node specified by the cloud administrator, a display area 902 of a candidate movement node determined by the cloud management program 178 using the method described in the present embodiment, and a cloud management program 178. The display part 903 of the change content of the middle box 13 determined by the method described in the embodiment is included. Note that display content other than the display content illustrated in FIG. 18 may be displayed on the screen 900. The screen 900 may have a screen configuration other than the screen configuration example illustrated in FIG.

  Next, data processing executed in the data center system 1000 of the present embodiment having the above configuration will be described. First, the data processing flow executed for setting the weight and the degree of coupling that are the premise of data processing will be described. FIG. 19 is a flowchart illustrating an example of a procedure for setting the weights of nodes constituting the application system 2000 and the degree of coupling between nodes in the present embodiment. In the flowcharts referred to hereinafter, the symbol “S” included in the reference numerals attached to the respective processing steps indicates “steps”.

  First, using the cloud usage terminal 3, a command for notifying the start of setting is transmitted to the coupling degree setting program 177 (S1001). Specifically, a setting start command is input by the cloud user from the cloud usage terminal 3 to the coupling degree setting program 177 of the management server 11 through the user interface.

  Next, the weight information of each node included in the application system 2000 is input to the coupling degree setting program 177 via the interface provided by the coupling degree setting program 177 (S1002).

  Next, the coupling degree between individual nodes included in the application system 2000 is input to the coupling degree setting program 177 via the interface provided by the coupling degree setting program 177 (S1003).

  Next, requirements required for the middle box 13 are input as a service profile 175 to the coupling degree setting program 177 via the interface provided by the coupling degree setting program 177 (S1004).

  The connectivity setting program 177 stores the node weight information in the node weight information management table 173 based on the information set by the cloud user through the cloud use terminal 3 in S1002. Also, the degree-of-association setting program 177 stores the degree-of-joint information between nodes in the degree-of-association management table 172 based on the information specified by the cloud user through the cloud-use terminal 3 in S1003. When the related node is the middle box 13, the identification information of the service profile 175 designated by the cloud user through the cloud use terminal 3 in S1004 is stored in the association degree management table 172 (S1005).

  Through the above-described weight and coupling degree setting processing flow, data used for determining whether or not a node included in the application system 200 can be moved is collected.

  Next, the entire node movement process executed when moving any node included in the application system 2000 will be described. FIG. 20 is a flowchart illustrating an example of the entire node movement procedure in the present embodiment.

  First, the cloud management program 178 receives from the cloud management terminal 4 designation of a node as a movement candidate and a movement destination (S1101). The reason for the cloud administrator instructing the start of the movement procedure from the cloud management terminal 4 may be, for example, an event in which the monitoring server 12 issues a warning to the cloud administrator when the load on a certain node increases.

  The cloud management program 178 executes a node movement availability verification flow starting from the node specified in step 1101. This flow will be described later (S1102).

  The cloud management program 178 evaluates the result of the node movement availability verification flow performed in step 1102 (S1103). If it is determined that there is a movable node (S1103, Yes), the cloud management program 178 notifies the cloud management terminal 4 that the node designated in step 1101 can be moved (S1104). If another node that needs to be moved together with the node is found in step 1102, the cloud management program 178 also presents the other node to the cloud administrator through the cloud management terminal 4. When it is determined in step 1102 that the middle box 13 needs to be changed as the node moves, the cloud management program 178 also presents the change to the cloud administrator through the cloud management terminal 4. An example of the screen is as shown in FIG.

  When the cloud administrator determines that the node has been moved and inputs that fact through the cloud management terminal 4 (Yes in S1105), in S1106, the cloud management program 178 causes the virtual of the host computer 10 on which the virtual machine 161 is operating. The machine monitor program 160 is instructed to perform the migration procedure of the virtual machine 161 with the virtual machine monitor program 160 of the host computer 10 of the destination data center 2, and a completion notification is received (S1106).

  Next, the cloud management program 178 updates the current location 454 column to the destination data center 2 for the record whose node ID column 452 matches the node whose movement has been completed in the node management table 179 and ends the processing ( S1107).

  On the other hand, if it is determined in S1103 that there is no movable node (S1103, No), the cloud management program 178 cannot move the node specified by the cloud administrator in S1101 to the cloud administrator through the cloud management terminal 4. (S1108), the screen returns to S1101, and a screen for selecting a different destination is presented to the cloud administrator through the different migration candidate virtual machines or the cloud management terminal 4.

  In step 1105, when the cloud administrator notifies that the movement of the node is canceled through the cloud management terminal 4 (S1105, No), the cloud management program 178 returns to S1101 and selects a different migration candidate virtual machine or a different migration destination. A screen to be selected is presented to the cloud administrator through the cloud management terminal 4.

  According to the overall node movement flow described above, the node movement process between the data centers 1 and 2 in consideration of various constraint conditions can be executed.

  Next, the node movement availability verification process in the entire node movement process will be described. FIG. 21 is a flowchart illustrating an example of a detailed processing procedure of node movement availability verification processing in the present embodiment.

  First, the cloud management program 178 inquires the monitoring server 12 about the line status between the data center 1 and the data center 2. The monitoring server 12 notifies the cloud management program 178 of the bandwidth between the data centers 1 and 2 and the response speed collected by the network monitoring program 180 (S1201).

  The cloud management program 178 creates an empty verified node list and a movement candidate node list (S1202). These lists are temporary lists used in the node movement availability verification flow. The verified node list and the movement candidate node list may have a general known data structure such as a queue or a list.

  Next, when this step is executed for the first time since the start of the entire node movement flow, the cloud management program 178 determines whether the node selected by the cloud administrator in S1101 A flow for verifying the degree of coupling is executed (S1203). This flow will be described later. When this step is executed for the second and subsequent times after starting the entire node movement flow, the node weight and the connection degree verification flow with the related node are executed for the node selected in S1207, which will be described later.

  The cloud management program 178 determines whether or not the node can be moved based on the weight and the execution result of the connectivity verification flow in S1203 (S1204). If it is determined that the node is movable as a result of the weight and coupling degree verification (S1204, Yes), the cloud management program 178 adds the node to the verified node list (S1205). If it is determined that the node cannot be moved as a result of the weight and connection degree verification (No in S1205), the cloud management program 178 advances the process to S1206.

  The cloud management program 178 compares the movement candidate node list with the verified node list, and the presence / absence of a node included in the movement candidate node list but not included in the verified node list, that is, an unverified movement candidate node The presence or absence is determined (S1206). When there is an unverified movement candidate node (S1206, Yes), the cloud management program 178 advances the processing to S1207. When there is no unverified movement candidate node (S1206, No), the cloud management program 178 advances the processing to S1208.

  In S1207, the cloud management program 178 selects one movement candidate node not included in the verified node list from the movement candidate node list, and repeatedly executes the processing of S1203 to S1206.

  Next, when it is determined in S1206 that there is no unverified movement candidate node (S1206, No), the cloud management program 178 refers to the movement candidate node list and evaluates whether there is a movement candidate (S1208). . When it is determined that there is a movement candidate (S1208, Yes), the cloud management program 178 advances the processing to S1209. If it is determined that there is no movement candidate (S1208, No), the cloud management program 178 ends this processing and returns to the node movement overall flow.

  In step S1209, the cloud management program 178 determines the requested service level for the middle box 13. Specifically, the cloud management program 178 uses the service in the record out of all the records whose node ID column 502 matches the ID of the node included in the movement candidate node list from the connectivity management table 172 of the management server 11. Records including values in the profile column 505 are acquired. Further, the cloud management program 178 acquires a service profile 175 having, in the service profile ID descriptor 800, a service profile ID 701 that matches a value included in the service profile column 505 of the acquired record.

  Next, in step 1210, the cloud management program 178 makes the following inquiry to the management server 11 provided in the data center 2 that is the movement destination (S1210). Specifically, the cloud management program 178 acquires all the records in which the node ID column 502 matches the node ID included in the migration candidate node list from the connectivity management table 172 of the management server 11, and moves the node. A request profile 176 is created. In creating the node movement request profile 175, the cloud management program 178 acquires the bandwidth 301 and the response speed 302 corresponding to the coupling degree 504 from the coupling degree definition table 170 of the management server 11. Further, the cloud management program 178 transmits the node movement request profile 176 and the service profile 175 acquired in S1209 to the cloud management program 178 running on the management server 11 of the data center 2. For the transmission of the node movement request profile 176 and the service profile 175, a known technique relating to a communication protocol such as SOAP or HTTP can be used.

  Next, the cloud management program 178 running on the management server 11 of the data center 2 executes an acceptance / rejection determination flow (S1211). This flow will be described later.

  Next, the cloud management program 178 evaluates the execution result of the acceptability determination flow in step 1210 (S1212). As a result of the acceptability determination by the cloud management program 178 of the data center 2, when it is determined that the acceptance is possible (S1212, Yes), the cloud management program 178 determines the created migration candidate node list and returns to the overall node migration flow. (S1213). As a result of acceptability determination, if it is determined that acceptance is not possible (No in S1212), the cloud management program 178 determines that there is no migration candidate and returns to the overall node migration flow.

  According to the above node moveability verification process, node candidates that can move between data centers can be selected in consideration of the weight of each node and the relevance between other nodes.

  Next, the weight and connection degree verification process executed in the node movement availability verification process will be described. FIG. 22 is a flowchart illustrating an example of a detailed processing procedure of node weight and coupling degree verification in the present embodiment.

  The cloud management program 178 acquires, from the node weight information management table 173, the value of the record weight column 603 in which the node ID column 602 matches the node ID specified in S1203 of the node movement availability verification processing flow. Further, the cloud management program 178 acquires the value of the data center column 402 of the record whose weight column 400 matches the weight value from the weight definition table 171 (S1301).

  Next, the cloud management program 178 compares and evaluates whether the destination specified in S1203 is included in the value acquired in S1301 in order to verify the validity of the node specified in S1203 and its destination. (S1302). If it is determined as a result of the movement destination verification that the movement destination is appropriate (S1302, Yes), the cloud management program 178 adds the node specified in S1203 to the movement candidate node list (S1303). As a result of the verification of the movement destination, when it is determined that the movement destination is not appropriate (S1302, No), the cloud management program 178 determines that the node cannot be moved, and returns to the node movement availability verification flow.

  In S1304, the cloud management program 178 acquires the value of the related node column 503 of all the records in which the node ID column 502 matches the ID of the node specified in S1203 from the association degree management table 172, and stores the related node list. create.

  Next, the cloud management program 178 compares the related node list with the verified node list, selects nodes that are not included in the verified node list one by one from the related node list, and repeats subsequent steps S1306 to S1309. Process (S1305).

  Next, the cloud management program 178 obtains a record in which the node ID column 502 and the related node ID column 503 match the node specified in S1203 and the related node selected in S1305 from the coupling degree management table 172, and combines them. The value of the degree column 504 is acquired (S1306).

  Next, the cloud management program 178 acquires a record in which the value of the coupling degree column 300 matches the coupling degree value acquired in S1306 from the coupling degree definition table 170, and acquires the values of the band column 301 and the response speed column 302. (1307).

  Next, the cloud management program 178 compares whether the bandwidth value and response speed value acquired in S1307 exceed both the bandwidth and response speed between the data centers 1 and 2 acquired in S1201 (S1308). As a result of the comparison, when it is determined that the degree of coupling is equal to or exceeding the inter-data center line status (S1308, Yes), the cloud management program 178 advances the process to S1309. As a result of the comparison, when it is determined that the degree of coupling is below the inter-data center line status (S1308, No), the cloud management program 178 advances the process to S1310.

  Next, the cloud management program 178 acquires a record in which the related node ID column 503 matches the ID of the related node selected in S1305 from the association degree management table 172, and whether the record has a value in the service profile column 505. By determining whether or not the related node selected in S1305 is the middle box 13. If the related node selected in S1305 is not the middle box 13 and is not included in the movement candidate node list, the related node is added to the movement candidate node list (S1309).

  In S1310, the cloud management program 178 evaluates whether or not the processing of S1306 to S1309 has been performed on all nodes that are not included in the verified node list in the related node list created in S1304. If it is determined that the process has been performed (S1310, Yes), the process returns to the node movement availability verification flow. When it is determined that there is a related node that has not been implemented yet (S1310, No), the cloud management program 178 and the process are returned to S1305.

  According to the weight and coupling degree verification process described above, it is possible to verify whether each node can be a candidate for movement to another data center in terms of the weight of the node and the relevance with another node.

  Next, the acceptance / rejection verification process executed in the node movement availability verification process will be described. FIG. 23 is a flowchart illustrating an example of a detailed processing procedure for verifying whether a migration candidate node can be accepted by the cloud management program 178 running on the management server 11 of the migration destination data center 2 in the present embodiment.

  First, the cloud management program 178 of the data center 2 receives a movement request from the cloud management program 178 of the data center 1 (S1401). The movement request includes a node movement request profile 176 and a service profile 175 required for the middle box 13.

  The cloud management program 178 of the data center 2 makes an inquiry to the network monitoring program 180 and the node monitoring program 181 of the monitoring server 12 of the data center 2, and in the free resources of the host computer 10 of the data center 2 and data communication between the host computers 10. The bandwidth and response speed are acquired (S1402).

  Next, the cloud management program 178 of the data center 2 indicates that the resource availability of the host computer 10 acquired in S1402, the bandwidth in the data communication between the host computers 10 and the response speed in the node movement request profile 176 received in S1401. By comparing whether or not the values described in the resource descriptor 815, the bandwidth descriptor 817, and the response speed descriptor 818 are satisfied, the movement candidate node can maintain the same degree of coupling as the data center 1 in the data center 2. It is determined whether or not (S1403). As a result of the determination, when it is determined that the combination degree of the movement candidate nodes can be maintained (S1403, Yes), the cloud management program 178 advances the process to S1404. As a result of the determination, when it is determined that the degree of coupling cannot be maintained (S1403, No), the cloud management program 178 advances the processing to S1408 as being unacceptable.

  The cloud management program 178 of the data center 2 refers to the service profile column 701 in the middle box management table 174, and acquires a list of service profiles 175 supported by the existing middle box 13 in the data center 2 (S1404). ).

  Next, the cloud management program 178 of the data center 2 receives the requested service profile received from the management server 11 of the data center 1 in S1401 from the service profiles 175 indicated by the service profile ID 800 in the list of service profiles 175 acquired in S1404. It is determined by comparing the values of the individual descriptors whether or not there are those satisfying 175 (S1405). As a result, the cloud management program 178 determines whether or not the middle box 13 of the data center 2 can provide the same service level as the middle box 13 of the data center 1 to the migration candidate node (S1406).

  If the cloud management program 178 of the data center 2 determines that a candidate service profile has been found in S1405 (S1406, Yes), the process proceeds to S1407. If it is determined that a candidate service profile 175 is not found (No in S1406), the cloud management program 178 advances the processing to S1408 as being unacceptable.

  In S1407, the cloud management program 178 of the data center 2 reserves resources of the host computer 10 in the data center 2 in order to secure resources for accepting the migration candidate nodes. The resource reservation management may be performed by any of the virtual machine monitor program 160, the cloud management program 178, and an external management program.

  Next, the cloud management program 178 of the data center 2 notifies the cloud management program 178 of the data center 1 of the acceptability result (S1408). The cloud management program 178 of the data center 1 returns the processing to the node movement availability verification flow with the result.

  According to the acceptability verification process described above, it is possible to determine whether or not the node that is the migration target can be accepted in terms of the resources provided by the host computer 10 and the service provided by the middle box 13 in the destination data center. it can.

  Next, the reconstruction process will be described. FIG. 24 is a flowchart illustrating an example of a detailed processing procedure of application system reconfiguration after moving a node in the present embodiment. Here, after the node movement from the data center 1 to the data center 2, the node of the application system 2000 across the data center 1 and the data center 2 is re-established as the line status between the data centers 1 and 2 changes. An example of arrangement will be shown.

  First, the cloud management program 178 of the data center 1 recognizes a change in the line status between the data center 1 and the data center 2 (S1501). Specifically, the cloud management program 178 may periodically make an inquiry to the network monitoring program 180 of the monitoring server 12, or the network monitoring program 180 may notify the cloud management program 178 periodically or when a change occurs. .

  Next, the cloud management program 178 of the data center 1 acquires the value of the node ID column 452 of the record in which the current location 454 indicates the data center 2 among the records in which the location 453 indicates the data center 1 in the node management table 179. Next, the cloud management program 178 acquires the value of the combination degree column 504 of the record that matches the value acquired by the node ID column 502 or the related node ID column 503 of the combination degree management table 172 (S1502). Furthermore, the cloud management program 178 determines whether or not the line status between the data centers 1 and 2 recognized in S1501 exceeds the acquired coupling degree value (S1503). The determination method is the same as S1307 to S1308 in the weight and coupling degree verification flow illustrated in FIG. As a result of the determination, if it is determined that the line status is below the coupling degree value (S1503, No), the process proceeds to S1504. If it is determined that the degree of coupling can be maintained (S1503, Yes), this flow ends with no need for reconfiguration.

  In S1504, the cloud management program 178 of the data center 1 inquires of the network monitoring program 180 and the node monitoring program 181 of the monitoring server 12 of the data center 1, and acquires free resources of nodes, bandwidth between nodes, and response speed (S1504). ).

  Next, the cloud management program 178 of the data center 1 determines the node resource status and the node status between nodes acquired in S1504 between the nodes of the data center 1 and the data center 2 constituting the application system 2000. It is determined whether or not the node moved to the data center 2 can be returned to the data center 1 by comparing with the degree of coupling (S1505). As a result of the determination, when it is determined that the node of the data center 2 can be returned to the data center 1 (S1505, Yes), the cloud management program 178 advances the processing to S1506. As a result of the determination, when it is determined that the node of the data center 2 cannot be returned to the data center 1 (S1505, No), the cloud management program 178 advances the processing to S1508 as being unacceptable.

  In S1506, the cloud management program 178 of the data center 1 reserves resources of the host computer 10 in the data center 1 in order to secure resources for returning the nodes of the data center 2.

  Next, the cloud management program 178 of the data center 1 notifies the cloud management program 178 of the data center 2 of the movement for returning the node (S1507).

  The cloud management program 178 of the data center 1 and the cloud management program 178 of the data center 2 execute processing by instructing the migration to the virtual machine monitor program 160 on which the node to be moved and the destination virtual machine monitor program 160 are moved. The process ends (S1508).

  On the other hand, if it is determined that the node of the data center 2 cannot be returned to the data center 1, in S1509, the cloud management program 178 of the data center 1 is a data center other than the data center 2 and the application system 2000 A data center having a line status that can maintain the degree of coupling between the nodes is selected, and the cloud management program 178 of the data center is inquired about migration. The subsequent processing is the same as that after S1210 in the above-described node movement availability verification flow.

  According to the above processing, when the line status changes with respect to the destination node, it is possible to perform the rearrangement of the node while maintaining the degree of coupling.

  The first embodiment of the present invention has been described above. According to the first embodiment, the data center system 1000 manages the relationship between nodes with an abstracted index called the degree of coupling, and manages the relationship between the node and the middle box 13 with the index called the degree of coupling and the service profile 175. . As a result, it becomes possible for the cloud administrator to know the configuration of the application system 2000 of the cloud user, and to construct the application system 2000 across the data center as long as the service level of the application system 2000 is not affected. In addition, the application system 2000 can be reconfigured according to changes in the line status between data centers. As a result, it is possible to achieve both leveling of resource load and utilization rate and appropriate separation between cloud management and cloud usage.

  Next, a second embodiment of the data center linkage system 1000 illustrated in FIG. 1 to which the present invention is applied will be described. In the following, parts not specifically described shall have the same configuration as in the first embodiment.

  The present embodiment is different from the first embodiment in that the data center system 1000 derives the degree of coupling between nodes. With this configuration, in this embodiment, the cloud user does not need to set the degree of coupling between nodes in advance. Further, for example, by calculating the degree of coupling using the substantial line usage status between nodes, the cloud management program 178 can determine the movement of the virtual machine 161 according to the actual status of the line status.

  FIG. 25 is a flowchart illustrating an example of a detailed processing procedure for deriving the degree of coupling by the cloud management program 178 in the second embodiment. This flow may be executed when the cloud user updates the parameters described later via the interface provided by the coupling degree setting program 177, or when the cloud administrator updates the parameters. It may be implemented, or the coupling degree setting program 177 may be implemented periodically using the parameters at that time.

  First, the connection degree setting program 177 inquires of the network monitoring program 180 on the monitoring server 12 about the line status between nodes (S1601). In the present embodiment, it is assumed that the bandwidth used between nodes and the measured value of the response speed are used as the line status.

  Next, the coupling degree setting program 177 calculates the coupling degree according to the following calculation formula (S1602).

Degree of coupling between nodes mn = C1 (communication flow rate per unit time between nodes mn / average flow rate per unit time between all nodes in the application system) + C2 (average response speed between nodes mn / total in the application system) Average response speed between nodes)
Here, C1 and C2 are coefficients, and are parameters that can be used to adjust which of flow rate and response speed is important. The parameters C1 and C2 may be defined in advance by the cloud user or may be defined in advance by the cloud administrator.

  Next, the coupling degree setting program 177 stores the coupling degree calculated in S1602 in the coupling degree column 504 of the coupling degree management table 172 (S1603).

  FIG. 24 is a flowchart illustrating an example of a detailed processing procedure for setting a degree of coupling by a cloud user according to the second embodiment.

  First, according to the operation of the cloud user, the management server 11 receives an instruction from the cloud use terminal 3 (S1701).

  Next, on the basis of the instruction received from the cloud usage terminal 3, the parameter designation for calculating the degree of coupling between individual nodes included in the application system 2000 is performed via the interface provided by the degree of coupling setting program 177. Accept from the use terminal 3 (S1702). Here, the parameters are the coefficients C1 and C2 in the above calculation formula, and the service profile 175 corresponding to the requirements required for the middle box 13.

  Next, the association degree setting program 177 updates the service profile column 505 of the association degree management table 172 based on the service profile 175 designated by the cloud user in S1702 (S1703).

  Finally, the coupling degree setting program 177 executes the coupling degree derivation process flow of FIG. 25 based on the coefficient set by the cloud user in S1702 (S1704).

  The second embodiment of the present invention has been described above. According to the second embodiment, the data center system 1000 derives the degree of coupling between nodes, thereby eliminating the need for the cloud user to set the degree of coupling between nodes in advance. Further, for example, by calculating the degree of coupling using the substantial line usage status between the nodes, the resource usage can be leveled according to the actual usage.

  The embodiment of the invention described above is for facilitating the understanding of the present invention, and does not limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and equivalents thereof are also included in the present invention.

1 Data center 1
2 Data center 2
DESCRIPTION OF SYMBOLS 3 Cloud utilization terminal 4 Cloud management terminal 5 Client terminal 6 Communication network 10 Host computer 11 Management server 12 Monitoring server 13 Middle box 14 Routing apparatus 15 Communication network

Claims (10)

At least a first data center in which an application system connected to an external device and providing data processing services to the external device is constructed, and a second data center having a configuration equivalent to the first data center A data center system management method comprising:
The application system has a plurality of components that are communicably connected to each other,
Selecting a first component to be a movement candidate as a candidate to move to the second data center from among the plurality of components of the application system of the first data center;
Setting a degree of coupling for each of the plurality of components , which is an index indicating a necessary condition regarding communication between the plurality of components included in the application system;
Between data centers that are information on the degree of coupling set for the first component that is the movement candidate and the communication status of the communication line between the first data center and the second data center Comparing and evaluating communication line information;
As a result of the comparative evaluation, when it is determined that the degree of coupling is superior to the communication line information between the data centers, the component that needs to move to the second data center together with the first component Determining a second component that is and selecting as a new candidate for movement;
After moving some of the plurality of components of the application system of the first data center to the second data center, between the first data center and the second data center Detecting a change in communication line status as communication line information between the data centers;
With respect to said data center communication line information after the change, in the application system, are set between the plurality of components that the plurality of components in said first data center to the second data center Evaluating whether the degree of coupling can be maintained;
Whether it is possible to move the first and second components of the application system in the second data center to the first data center when it is determined that the degree of coupling cannot be maintained. Determining using the resource condition and the degree of coupling for the plurality of components included in the application system in the first data center;
When it is determined that it is difficult to move to the first data center, a said first data center and other data centers other than the second data center, said first and is the determination select data center is a line situation capable of maintaining cohesion regarding second component, comprising the steps of requesting implementation of acceptance determination of the in the data center first and second components,
Management method of data center system including   The data center system management method according to claim 1, wherein the coupling degree includes a communication band and a response speed for the communication line. The step of setting the plurality of components and, said first data center and the second data center the weight information indicating correspondence to each of the range that can move the plurality of components of the application system When,
For the first and second components that are the movement candidates, to the second data center of the first and second components based on the weight information set for the first data center The method of managing a data center system according to claim 1, further comprising: Wherein the coupling degree between one or more of the components included in the moving candidate, the second in a data center communications line information about the communication status is a component communication line between each of said plurality of components A step of comparing and evaluating information;
As a result of the comparative evaluation, the communication line information between the plurality of components in the second data center is superior to the coupling degree set between the first and second components that are the movement candidates. The data center system management method according to claim 1, further comprising a step of evaluating whether or not. Of the plurality of components of the application system, service requirement information indicating requirements regarding service contents to be provided by a communication service device that provides a service for communication performed in the application system is set for the communication service device. Steps,
The first and second components that are the movement candidates have the service requirement information having contents equivalent to the service contents received at the first data center in the second data center. The data center system management method according to claim 1, further comprising a step of evaluating whether the communication service device is available.   6. The service content of the communication service device includes a function of applying a communication control policy or a security policy to communication, including a load distribution function, a firewall function, or an intrusion detection function. Data center system management method. Inter-component communication line information, which is information related to the communication status of communication lines between the plurality of components in the second data center , is set for the first and second components that are the movement candidates. The service requirement information having a content equivalent to the service content received at the first data center by the first and second constituent elements that are superior to the combined degree that is the movement candidate. Determining that the first and second components that are the movement candidates are acceptable to the second data center when it is determined that the communication service apparatus having
Notifying the first data center of the determination result that the data is acceptable from the second data center;
The data center system management method according to claim 5. 2. The step of calculating the degree of coupling set between the plurality of components of the application system based on an actual communication state in a communication line between the plurality of components. Data center system management method. A first data center in which an application system connected to an external device and providing a data processing service to the external device is constructed; and a second data center having a configuration equivalent to the first data center. Data center system,
The application system has a plurality of components that are communicably connected to each other,
A management device that is communicably connected to each of the plurality of components in each of the data centers and manages the application system,
From among the plurality of components by the application system has the first data center, comprising: selecting a first component comprising a first migration candidate as a candidate to be moved to the second data center ,
A step of coupling degree is an index showing the requirements of communication between the plurality of components included in the application system, set for each of the plurality of components,
Setting weight information indicating the correspondence between the plurality of components of the application system and the range in which the plurality of components can be moved, for the first data center and the second data center,
Of the plurality of components of the application system, service requirement information indicating requirements regarding service contents to be provided by a communication service device that provides a service for communication performed in the application system is set for the communication service device. ,
Information on the degree of coupling set for the first component that is the first movement candidate and the communication status of the communication line between the first data center and the second data center Compare and evaluate communication line information between data centers,
As a result of the comparative evaluation, when it is determined that the degree of coupling is superior to the communication line information between the data centers, the component that needs to move to the second data center together with the first component Is determined as a second candidate for movement,
The first and second components that are the first and second movement candidates are moved to the second data center based on the weight information set for the first data center. determining the first and second components can as a third movement candidates,
And the degree of coupling between the third one or more of the components included in the candidate travel, between said second component which is information related to a communication environment of the communication line between each of said components in the data center communications Compare and evaluate the line information,
As a result of the comparative evaluation, evaluate whether the inter-component communication line information in the second data center is superior to the coupling degree set for the third movement candidate,
The service requirement information in which the first and second components that are the third movement candidates have contents equivalent to the service contents received at the first data center in the second data center. Evaluating whether the communication service device having
The inter-component communication line information, which is information relating to the communication status of the communication lines between the plurality of components in the second data center , is superior to the coupling degree set for the third movement candidate. And the communication service apparatus having the service requirement information having contents equivalent to the service contents received by the first data center in the first and second constituent elements that are the third movement candidates. Is determined to be usable, the first and second components that are the third movement candidates are determined to be acceptable to the second data center, and the third movement is performed. Moving the candidate first and second components from the first data center to the second data center;
After moving some of the plurality of components of the application system of the first data center to the second data center, between the first data center and the second data center Detecting a change in communication line status as communication line information between the data centers,
With respect to said data center communication line information after the change, in the application system, are set between the plurality of components that the plurality of components in said first data center to the second data center And evaluate whether the degree of coupling can be maintained,
Whether it is possible to move the first and second components of the application system in the second data center to the first data center when it is determined that the degree of coupling cannot be maintained. , determined using the resource condition and the coupling degree of the plurality of components included in the application system at the first data center,
When it is determined that it is difficult to move to the first data center, a said first data center and other data centers other than the second data center, said first and is the determination select data center is a line situation capable of maintaining cohesion on two components, request the implementation of acceptance determination of the first and second components are the said to the data center the third migration candidate To
Data center system. A first data center having a plurality of components connected so as to be communicable with each other, an application system connected to an external device and providing a data processing service to the external device; and the first data center A data center system comprising a second data center having a configuration equivalent to the data center, and a management device for managing the application system connected to each component in the data center in a communicable manner,
The management device is
From among the plurality of components by the application system has the first data center, comprising: selecting a first component comprising a first migration candidate as a candidate to be moved to the second data center ,
Setting a degree of coupling for each of the plurality of components , which is an index indicating a necessary condition regarding communication between the plurality of components included in the application system;
And said component of said application system, the settings for the first data center and the second data center the weight information indicating correspondence to each of the range that can move the plurality of components,
Of the plurality of components of the application system, service requirement information indicating requirements regarding service contents to be provided by a communication service device that provides a service for communication performed in the application system is set for the communication service device. ,
Information on the degree of coupling set for the first component that is the first movement candidate and the communication status of the communication line between the first data center and the second data center Compare and evaluate communication line information between data centers,
As a result of the comparative evaluation, when it is determined that the degree of coupling is superior to the communication line information between the data centers, the component that needs to move to the second data center together with the first component determining a second component is, then selected as the second movement candidates,
The first and second components that are the first and second movement candidates are moved to the second data center based on the weight information set for the first data center. determining the first and second components can as a third movement candidates,
Wherein the coupling degree between one or more of the components included in the third migration candidate components is information related to a communication environment of the communication line between each of said plurality of components in said second data center Compare and evaluate communication line information between
As a result of the comparative evaluation, evaluate whether the inter-component communication line information in the second data center is superior to the coupling degree set for the third movement candidate,
The service requirement information in which the first and second components that are the third movement candidates have contents equivalent to the service contents received at the first data center in the second data center. Evaluating whether the communication service device having
The inter-component communication line information, which is information relating to the communication status of the communication lines between the plurality of components in the second data center , is superior to the coupling degree set for the third movement candidate. And the communication service apparatus having the service requirement information having contents equivalent to the service contents received by the first data center in the first and second constituent elements that are the third movement candidates. Is determined to be usable, the first and second components that are the third movement candidates are determined to be acceptable to the second data center, and the third movement is performed. Directing the candidate first and second components to move from the first data center to the second data center;
After moving some of the plurality of components of the application system of the first data center to the second data center, between the first data center and the second data center Detecting a change in communication line status as communication line information between the data centers,
Regarding the communication line information between the data centers after the change, in the application system, it is set between the plurality of components in the first data center and the components in the second data center. Evaluate whether the degree of binding can be maintained,
Whether it is possible to move the first and second components of the application system in the second data center to the first data center when it is determined that the degree of coupling cannot be maintained. , determined using the resource condition and the coupling degree of the plurality of components included in the application system in the first data center,
When it is determined that it is difficult to move to the first data center, a said first data center and other data centers other than the second data center, said first and is the determination select data center is a line situation capable of maintaining cohesion on two components, request the implementation of acceptance determination of the first and second components are the said to the data center the third migration candidate To
Data center system management device.

Images