JP2009282999A - Method for management of work resource and execution system thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for managing resources according to an operation condition of each work application. <P>SOLUTION: This work resource management method in a work management system having a plurality of public domains 1 composed of a public domain management server 21 and virtual resources, and a plurality of private domains 2 composed of a private domain management server 11, a real work server 10 for executing work, and a real storage 13 for storing data to be used in work, the private domain management server 11 notifies the public domains 1 of the existence of a surplus resource where the surplus resource exists, and the public domain management server 21 associates the surplus resource as a virtual resource, receives a resource providing request from the private domain management server 11, and notifies that the surplus resource corresponding to the virtual resource is lent to the private domain 2 being a requesting source. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a program placement technique for placing a program on a resource suitable for its operating conditions.

  2. Description of the Related Art Conventionally, business application programs (also referred to as business programs, hereinafter referred to as APs) that perform various business processes are executed by a plurality of server processing devices, and when providing the business services to client processing devices, The business AP is executed by allocating the business AP and data used in the processing to resources such as various server processing devices having a memory capacity and various storage devices having different storage capacities.

  Also, according to the operating status of the server processing device or storage device executing the business AP, for example, a predetermined operation definition such as when the CPU usage rate of a server increases or the remaining capacity of the storage device decreases. Accordingly, the server processing apparatus that executes the business AP and the storage apparatus that stores the data are also changed.

  In the above prior art, performance / capacity monitoring of servers that execute various business APs, storage device performance / capacity monitoring, or monitoring of a cluster composed of a plurality of servers, etc. are performed. Re-arranged to provide business services, but each monitoring function is operating independently, and monitoring was performed regardless of the business being executed, so depending on the monitoring results In other words, there is a problem in that it is impossible to perform arrangement suitable for each business AP.

  For example, when the CPU usage rate or the remaining storage capacity of a server reaches a predetermined value, there is a possibility that the value is normal for a certain business AP but is abnormal for other business APs. However, since the conventional technology performs monitoring regardless of the business AP, when the load increases, relocation unnecessary for the business AP is performed, or relocation necessary for the business AP is performed. There were times when I wasn't.

  An object of the present invention is to provide a technique capable of solving the above problems and managing resources according to the operating conditions of each business application.

  The present invention generates a business application information in which a business application configuration management information and a resource configuration management information are associated with each other in a program allocation system that allocates a program to a resource suitable for the operating conditions, and uses the business application information for each business application. It manages resources.

  In the program placement system of the present invention, configuration management information for storing the operating status, operating conditions, and operating history of each business AP is generated, and the management information is collected at the time of execution of each business AP. A process for maintaining and managing the contents of the AP configuration management information is performed.

  In addition, processing for generating configuration management information for storing the status related to the performance and capacity of each resource, collecting the management information during operation of each resource, and maintaining the contents of the configuration management information of the resource I do.

  By reading the identification information of the resource used by each business AP from the configuration management information of the business AP, and reading the configuration management information of the resource identified by the identification information from the configuration management information of each resource, Business AP information in which the AP configuration management information is associated with the searched resource configuration management information is generated.

  Next, referring to the policy management rule for each business AP for determining the status of each business AP, the resource configuration management information in the business AP information is a rule for detecting a resource abnormality in the corresponding business AP. When it is determined whether or not a resource abnormality has occurred for the business AP, information indicating that a resource abnormality has occurred in the business AP is output.

  Further, in the present invention, when a business AP is started or when a resource abnormality is detected in a business AP as described above while the business AP is operating, resources suitable for the operating conditions of the business AP are set. Clustering is performed by execution control that searches from the resource configuration management information and places or moves the business AP to the searched resource to execute business processing. At that time, the access path to the data used in the business process is controlled at the same time so that the data can be accessed with the arranged or moved resources.

  In the present invention, clustering and data access paths are expressed using the terms “cluster management program” and “data path management program”, respectively. The “cluster management program” is a program that operates on a plurality of computers in order to stably operate the business AP, and is a program that manages the status monitoring and activation / termination of the business AP. The “data path management program” is a program for managing a communication path between a computer and a storage or a computer and a network and a storage so that a business AP operating on the server can refer to, add, change, and delete the business data. It is. The data path management program manages establishment and termination of the communication path, monitoring of the state of the communication path, and routing of the communication path. In the present invention, in order to operate the business AP, the “cluster management program” and the “data path management program” are used to control the business data used by the business AP and the business AP. An example of the operation description of the “cluster management program” and the “data path management program” is shown in FIG.

  Furthermore, in the present invention, an area where each resource is actually installed is a private domain, and an area where a resource is virtually installed is a public domain, and the configuration management information of each resource installed in a plurality of private domains is referred to. If there are surplus resources, such as those that have not been operating for a predetermined time, and if there are surplus resources, the configuration management information is provided from the private domain to the public domain, the resources are virtually If a resource that meets the operating conditions of the business AP is not found when the business AP is started or when a resource abnormality is detected, surplus resources that meet the operating conditions from the public domain are managed. Lending configuration management information to the private domain Accordingly, the available surplus resources which is installed in a certain private domain from another private domain.

  The terms “private domain” and “public domain” used in the present invention will be described. A "private domain" is a logical grouping that is based on factors such as whether multiple servers and storage are connected via a network, and the same administrator manages resources or distributes resources to the same location. Say the unit. For example, in a computer center where resources of a plurality of companies are installed, a resource for company A can be managed as one private domain, and a resource for company B can be managed as another private domain. In one company, the resources of the C office can be managed as one private domain, and the resources of the D office can be managed as another private domain. On the other hand, the “public domain” is a virtual computer system that virtually exchanges surplus resources with the private domain. For example, when there are abundant resources at the head office of company E, the head office can install both public domain and private domain, and lend surplus resources to C office and D office.

  In the present invention, policy management rules including a rule for detecting a resource abnormality in the business AP and a rule for detecting the surplus resource are centrally managed, and when those rules are changed, Distribute the changed rules to each private domain to manage the policy in an integrated manner.

  As described above, in the present invention, management of resources used for each business AP is performed by associating the resources used by the business AP with the business AP information. Therefore, the present invention is suitable for the business such as detection of resource abnormality for each business AP. Management of resources, execution control suitable for the operating conditions of each business, provision of surplus resources, and integrated management of policy management rules for performing them.

  As described above, according to the program arrangement system of the present invention, business application information in which business application configuration management information and resource configuration management information are associated with each other is generated and used resources are managed in units of business applications. It is possible to manage resources according to the operating conditions of business applications.

  According to the present invention, it is possible to manage resources according to the operating conditions of business applications.

It is a figure which shows the outline | summary of the program arrangement | positioning system of this embodiment. It is a figure which shows the function hierarchy of the program arrangement | positioning system of this embodiment. It is a figure which shows the layer structure of the monitoring target object managed and controlled of this embodiment. It is a figure which shows the structural example of the configuration management information of this embodiment. It is a figure which shows the structural example of the management information of the business AP of this embodiment. It is a figure which shows the structural example of AP operation information of this embodiment. It is a figure which shows the structural example of AP operating condition of this embodiment. It is a figure which shows the structural example of AP operation log | history of this embodiment. It is a figure which shows the structural example of the server management information of this embodiment. It is a figure which shows the structural example of the server hardware information of this embodiment. It is a figure which shows the structural example of OS information of this embodiment. It is a figure which shows the structural example of DB information of this embodiment. It is a figure which shows the structural example of the cluster / data path information of this embodiment. It is a figure which shows the structural example of the storage management information of this embodiment. It is a figure which shows the outline | summary of the correlation of the management information of this embodiment. It is a figure which shows the processing content of the correlation process of the management information of this embodiment. It is a figure which shows the processing content of OS information search processing of this embodiment. It is a figure which shows the processing content of the logical host name search process of this embodiment. It is a figure which shows the processing content of DB information search processing of this embodiment. It is a figure which shows the processing content of the cluster / data path information search process of this embodiment. It is a figure which shows the processing content of the storage management information search process of this embodiment. It is a figure which shows the example of the business AP information linked | related of this embodiment. It is a figure which shows the problem of the business AP of FIG. 22 of this embodiment, and the example of a countermeasure. It is a figure which shows the example of the policy management rule of this embodiment. It is a figure which shows the processing content of the distribution process of the policy management rule of this embodiment. It is a figure which shows the processing content of the execution management process at the time of business AP starting of this embodiment. It is a figure which shows the image of the schedule reservation of the resource of this embodiment. It is a figure which shows the processing content of the space scheduler 44 of the private domain 2 of this embodiment. It is a figure which shows the processing content of the confirmation process of the business AP execution server of this embodiment. It is a figure which shows the processing content of the change process of the business execution server of this embodiment. It is a figure which shows the processing content of the confirmation process of the logical host of this embodiment. It is a figure which shows the processing content of the resource schedule reservation process of this embodiment. It is a figure which shows the processing content which specifies business AP of this embodiment, and copes with it. It is a figure which shows the processing content of the clustering process of business AP of this embodiment. It is a figure which shows the processing content of the confirmation process of the clustering server name of this embodiment. It is a figure which shows the processing content of the determination process of the surplus server of this embodiment. It is a figure which shows the processing content of the confirmation process of the resource reservation of this embodiment. It is a figure which shows the processing content of the configuration management process part 41 of the private domain 2 of this embodiment. It is a figure which shows the processing content of the management information request | requirement reception process of this embodiment. It is a figure which shows the processing content of the configuration management process part 31 of the public domain 1 of this embodiment. It is a figure which shows the processing content of the space scheduler 34 of the public domain 1 of this embodiment. It is a figure which shows the processing content of the performance and capacity management process part 47 of the private domain 2 of this embodiment. It is a figure which shows the processing content of the cluster / data path management process part 46 of the private domain 2 of this embodiment. It is a figure which shows the processing content of DB change processing of this embodiment. It is a figure which shows the processing content of the data path change process of this embodiment. It is a figure which shows the processing content of the policy management process in which the prediction result of this embodiment is reflected. It is a figure which shows the resource usage status grasp | ascertainment example of the machine of this embodiment, and business AP unit. It is a figure which shows the example of the resource tolerance | permissible_range of the machine of this embodiment, and business AP unit. It is a figure which shows the example of correction | amendment of the resource use upper limit of the machine of this embodiment, and business AP unit. It is a figure which shows the processing content of the resource use upper limit correction process of this embodiment. It is a figure which shows the processing content of the resource usage tendency prediction process of this embodiment. It is a figure which shows the processing content of the predicted value reflection process of this embodiment. It is a figure of operation | movement description of the cluster management program and data path management program of this embodiment.

  A program placement system according to an embodiment that places business APs on resources suitable for the operating conditions will be described below.

  FIG. 1 is a diagram showing an outline of the program placement system of this embodiment. As shown in FIG. 1, the program arrangement system according to the present embodiment includes a plurality of private domains 2 and a public domain 1, and these domains are connected by a network 3.

  In this embodiment, a domain in which a customer operates a business is a private domain 2. The private domain 2 has a real business server group 10 representing a plurality of servers that execute actual business, a plurality of storage enclosures, a real storage 13 that stores data used in actual business, A private domain management server 11 that performs management in the domain and private domain configuration information 12 that stores configuration management information of each business AP and resource are provided.

  The private domain management server 11 manages the real business server group 10 and the real storage 13 and manages their configuration information and status information as private domain configuration information 12.

  The public domain 1 is managed by the public domain management server 21 and includes a virtual business server group 20, a virtual storage 23, and public domain configuration information 22. The public domain configuration information 22 is configuration management information that is managed when the real business server group 10 or the real storage 13 surplus in the private domain 2 is temporarily or permanently borrowed from the private domain configuration information 12. . The public domain configuration information 22 is used for the private domain configuration information 12 when the borrowed real business server group 10 or the real storage 13 is returned. In other words, the entity of the virtual business server group 20 corresponds to the real business server group 10, and the entity of the virtual storage 23 corresponds to the real storage 13.

  When resources are insufficient in an arbitrary private domain 2, resources are borrowed from the public domain 1, and the arbitrary private domain 2 can also substitute the public domain 1.

  FIG. 2 is a diagram showing a functional hierarchy of the program placement system of the present embodiment. As shown in FIG. 2, the public domain management server 21 of this embodiment includes an operation monitoring processing unit 30, a configuration management processing unit 31, a policy management processing unit 32, and a space scheduler 34.

  The operation monitoring processing unit 30 is a processing unit that monitors the states of the configuration management processing unit 31 to the space scheduler 34 that are other functional components in the public domain 1. The configuration management processing unit 31 receives the configuration management information of the resource corresponding to the rule for detecting surplus resources from the configuration management processing unit 41 on the private domain 2 side, and the configuration management information of the virtual resource 38 in the public domain 1 It is a processing unit that maintains and manages.

  The policy management processing unit 32 is a processing unit that manages policy management rules in the public domain 1. The space scheduler 34 receives a resource provision request transmitted from the space scheduler 44 on the private domain 2 side when a resource having configuration management information that satisfies the operating conditions of the business AP 48 is not found, and performs the virtual management that is maintained and managed. It is a processing unit that transmits configuration management information that satisfies the operating conditions of the business AP 48 in the resource configuration management information.

  A program for causing the public domain management server 21 to function as the operation monitoring processing unit 30, the configuration management processing unit 31, the policy management processing unit 32, and the space scheduler 34 is recorded on a recording medium such as a CD-ROM and stored on a magnetic disk or the like. Then, it is assumed that it is loaded into a memory and executed. The recording medium for recording the program may be a recording medium other than the CD-ROM. The program may be used by installing it from the recording medium into the information processing apparatus, or the program may be used by accessing the recording medium through a network.

  The private domain management server 11 includes an operation monitoring processing unit 40, a configuration management processing unit 41, a policy management processing unit 42, a distribution management processing unit 43, a space scheduler 44, an execution management processing unit 45, and a cluster / data. A path management processing unit 46 and a performance / capacity management processing unit 47 are provided.

  The operation monitoring processing unit 40 is a processing unit that monitors the statuses of the configuration management processing unit 41 to the performance / capacity management processing unit 47 that are other functional components in the private domain 2. The configuration management processing unit 41 maintains and manages the configuration management information of each resource by referring to the management information collected by each processing unit, and refers to the configuration management information including the identification information of the resource used by each business AP 48. Then, by reading out the configuration management information of the resource identified by the identification information from the configuration management information of each resource, the business AP that associates the configuration management information of the business AP 48 with the configuration management information of the searched resource Information is generated, and when the resource configuration management information in the business AP information corresponds to a rule for detecting a resource abnormality in the corresponding business AP 48, the resource abnormality has occurred in the business AP 48. It is a processing unit that outputs the indicated information.

  The policy management processing unit 42 is a processing unit that centrally manages policy management rules including a rule for detecting a resource abnormality in the business AP 48 and a rule for detecting surplus resources. The distribution management processing unit 43 is a processing unit that distributes programs such as the business AP 48 and the DB management system from the private domain management server 11 to any real business server 49.

  The space scheduler 44 uses the resource as a resource for executing the business AP 48 when the configuration management information of the resource associated with the business AP 48 for which the reservation instruction has been issued satisfies the operating condition in the configuration management information of the business AP 48. It is a processing unit to make a reservation.

  The execution management processing unit 45 receives a request for execution of the business AP 48 and gives an instruction to reserve a resource to be used by the business AP 48, and starts the business AP 48 with the resource reserved by the instruction, When an abnormality is detected in the resource being executed, another resource having the configuration management information that satisfies the operating condition in the configuration management information of the business AP 48 in which the resource abnormality has occurred is searched, and the business using the searched resource It is a processing unit that performs processing for continuing processing of AP48.

  The cluster / data path management processing unit 46 manages the configuration and status of the class management program, and manages the configuration and status of the data path between the server 49 and the storage 50 included in the data path management program and instructions for changing the data path. Part. The performance / capacity management processing unit 47 is a processing unit that collects and manages management information indicating a state relating to the performance of each server 49, the capacity of the storage 50, and the like.

  The private domain management server 11 includes an operation monitoring processing unit 40, configuration management processing unit 41, policy management processing unit 42, distribution management processing unit 43, space scheduler 44, execution management processing unit 45, cluster / data path management processing unit 46, and performance. A program for functioning as the capacity management processing unit 47 is recorded on a recording medium such as a CD-ROM and stored in a magnetic disk or the like, and then loaded into a memory and executed. The recording medium for recording the program may be a recording medium other than the CD-ROM. The program may be used by installing it from the recording medium into the information processing apparatus, or the program may be used by accessing the recording medium through a network.

  As shown in FIG. 2, the private domain management server 11 of the private domain 2 includes an operation monitoring processing unit 40, a configuration management processing unit 41 (managing the private domain configuration information 12), a policy management processing unit 42, a distribution management processing unit 43, A space scheduler 44, an execution management processing unit 45, a cluster / data path management processing unit 46, and a performance / capacity management processing unit 47 are provided, and a plurality of business APs 48, servers 49, and storages 50 that can exist are maintained and managed.

  The public domain management server 21 of the public domain 1 includes an operation monitoring processing unit 30, a configuration management processing unit 31 (managing the public domain configuration information 22), a policy management processing unit 32, and a space scheduler 34, and the surplus of the private domain 2 The resource is maintained and managed as a virtual resource 38.

  The performance / capacity management processing unit 47 collects management information of the business AP 48, the server 49, and the storage 50 and saves the history. When the configuration management processing unit 41 requests management information, the performance / capacity management processing unit 47 displays the collected management information. respond.

  FIG. 3 is a diagram showing a layer structure of a monitoring object to be managed and controlled according to this embodiment. As shown in FIG. 3, the monitoring target managed and controlled in the present embodiment is composed of layers of the business AP 70, the server 80, and the storage 90, and the business AP 70 includes a Web application 71 and a batch job 72. The server 80 includes a DB management processing unit 81, a cluster / data path management processing unit 82, and an OS 83.

  FIG. 4 is a diagram showing a configuration example of configuration management information according to the present embodiment. As shown in FIG. 4, the configuration management information of this embodiment includes information on business APs, servers, and storages.

  FIG. 5 is a diagram illustrating a configuration example of management information of the business AP according to the present embodiment. As shown in FIG. 5, the management information of the business AP includes information indicating the AP operation status, the AP operation condition, and the AP operation history.

  FIG. 6 is a diagram illustrating a configuration example of AP operation information according to the present embodiment. As shown in FIG. 6, the AP operation status information includes business AP name, business AP execution domain name, business AP type such as batch, Web AP, and workflow, business AP execution server name, business AP execution user name, and business AP time. A business AP time indicating presence / absence of designation, business AP scheduled start time, business AP end scheduled time, business AP importance, and business AP status are stored.

  FIG. 7 is a diagram illustrating a configuration example of the AP operation condition according to the present embodiment. As shown in FIG. 7, the AP operating conditions are business characteristics (periodic job / online / others), machine usage (dedicated / shared), data path name, number of CPUs, amount of memory, number of processes, number of threads, disk capacity, A logical host name, a clustering availability flag, a clustering group name, and a used DB name are stored. In addition, a process name corresponding to the number of processes and a thread name corresponding to the number of threads are stored.

  FIG. 8 is a diagram illustrating a configuration example of an AP operation history according to the present embodiment. As shown in FIG. 8, the AP operation history is the AP operation history time, CPU usage rate, memory usage, process usage, thread usage, business AP name, disk usage capacity, CPU count, business AP start time, business AP end. Stores time, AP execution server name, and business AP end status.

  FIG. 9 is a diagram showing a configuration example of server management information according to the present embodiment. As shown in FIG. 9, the server management information stores OS information, DB information, and cluster / data path information as server hardware information.

  FIG. 10 is a diagram illustrating a configuration example of server hardware information according to the present embodiment. As shown in FIG. 10, the server hardware information stores the server type such as single, partition and blade, the number of logical servers, and the number of surplus CPUs. The server name, CPU type, and number of CPUs correspond to the number of logical servers. The server stores presence / absence of server provision to other domains, domain provision start time, domain provision end time, and OS information. In addition, a CPU number and a CPU state corresponding to the number of CPUs are stored.

  FIG. 11 is a diagram illustrating a configuration example of OS information according to the present embodiment. As shown in FIG. 11, the OS information includes the CPU number, OS type, maximum memory amount, used memory amount, remaining memory amount, maximum process number, used process number, remaining process number, maximum thread number, used thread number, remaining thread number. , Maximum disk capacity, used disk capacity, server name, number of logical hosts (k), remaining disk capacity, CPU usage rate, used memory rate, used process rate, used thread rate, user name number, logical A logical host name corresponding to the number of hosts and a user name corresponding to the number of user names are stored.

  The calculation formulas for the remaining number such as the number of remaining processes and the number of remaining threads in FIG. 11 are set by the policy management processing unit, and the default calculation formula is the maximum number−the used number = the remaining number. And the number of used are the numbers such as the maximum number of processes and the maximum number of threads, the numbers of used processes and the number of used threads, respectively.

  FIG. 12 is a diagram illustrating a configuration example of DB information according to the present embodiment. As shown in FIG. 12, the DB information has a DB name, a DB state, a business AP name, a data capacity, a data capacity history for each number of DBs, and stores a data capacity history time and a data capacity history usage as a data capacity history. is doing.

  The cluster / data path management processing unit 46 manages the configuration and status of the class management program, and the data path configuration and status between the server 49 and the storage 50 of the data path management program and instructions for changing the data path. . FIG. 13 shows cluster / data path management information.

  FIG. 13 is a diagram showing a configuration example of cluster / data path information according to the present embodiment. As shown in FIG. 13, the cluster / data path information includes cluster operation presence / absence, cluster software name, number of cluster registration business APs, data path presence / absence, number of data paths, and business AP names corresponding to the number of cluster registration business APs. Stores the business AP status, active server name, active logical host name, active data path name, and data path name, data path status, storage name, and storage logical unit name corresponding to the number of data paths.

  FIG. 14 is a diagram showing a configuration example of storage management information according to this embodiment. As shown in FIG. 14, the storage management information has a storage name, storage capacity, and number of storage logical units. The storage logical unit name, storage logical unit capacity, storage logical unit usage capacity, and storage logical unit corresponding to the number of storage logical units. Remaining capacity, storage provision to other domains, storage provision capacity to other domains, provision start time of storage to other domains, provision end time of storage to other domains, storage logical unit usage history, storage Stores history time and storage history usage.

  The configuration management processing unit 41 of this embodiment maintains configuration information by referring to the management information collected by the performance / capacity management processing unit 47 and the cluster / data path management processing unit 46, and Accepts management information reference and update requests from functional components. In particular, the configuration management information of the resources that the space scheduler 44 determines to be surplus is temporarily or permanently provided to the configuration management processing unit 31 of the public domain 1.

  The distribution management processing unit 43 performs processing for distributing programs such as the business AP 70 and the DB management processing unit 81 from the private domain management server 11 to any real business server.

  The space scheduler 44 monitors the schedule of the business AP and the operation status of the server 49 and the storage 50 through the configuration management processing unit 41, and if a surplus resource occurs, the space scheduler 44 notifies the configuration management processing unit 41 to that effect. Process to notify.

  The execution management processing unit 45 manages execution and termination of the business AP 70, and management information on the business AP start time 258 and the business AP end time 259 in FIG. 8 is notified to the configuration management processing unit 41 by the execution management processing unit 45. The

  The policy management processing unit 42 has a function of setting and changing operation standards of the configuration management processing unit 41, the performance / capacity management processing unit 47, the cluster / data path management processing unit 46, and the space scheduler 44, and the configuration management processing unit 41. Is provided to the configuration management processing unit 31 of the public domain 1, an access control list (ACL) for determining whether the corresponding resource is permitted the reference authority from the configuration management processing unit 31 have.

  It is assumed that the performance / capacity management processing unit 47 can change the calculation formula for calculating the remaining memory amount 354, the remaining process number 357, the remaining thread number 360, and the remaining disk capacity 365 in FIG. For example, the default calculation formula for the remaining memory amount 354 is (remaining memory amount 354) = (maximum memory amount 352) − (used memory amount 353), but when the safety coefficient α is set to the used memory amount 353 by policy management. , (Remaining memory amount 354) = (maximum memory amount 352) − (used memory amount 353) × (safety factor α), the remaining memory amount can be changed. Similarly, it is assumed that the space scheduler 44 can change the determination of surplus resources by adding a safety factor.

The cluster / data path management processing unit 46 changes the selection of the business AP 70 to be clustered by changing the business AP importance 228 in FIG. 6 according to the time zone, or changes the clustering destination server depending on the remaining amount of any resource. It can be changed.
In addition, the operation monitoring processing unit 40 performs processing for monitoring the status of other functional components of the private domain 2.

  On the other hand, the public domain 1 includes an operation monitoring processing unit 30, a configuration management processing unit 31 (which manages public domain configuration information 22), a policy management processing unit 32, and a space scheduler 34, and maintains a virtual resource 38 through configuration management. ,to manage.

  The configuration management processing unit 31 maintains and manages the resources provided from the configuration management processing unit 41 of one or more private domains 2 as virtual resources 38. Upon receiving a notification of a borrowed resource from the configuration management processing unit 41 of a private domain 2 or a notification that the use from the space scheduler 34 has ended, the configuration management processing unit 41 of the private domain 2 that provides the resource Return to

  The spatial scheduler 34 maintains and manages the virtual resource 38 existing in the configuration management processing unit 31 in consideration of its borrowing deadline, and when receiving a resource provision request, provides the virtual resource that matches the request condition. In addition, the configuration management processing unit 31 is notified of a virtual resource whose expiration date has been reached.

The policy management processing unit 32 has a function of setting and changing the operation reference of the spatial scheduler 34. When the spatial scheduler 34 receives a virtual resource provision request from a plurality of private domains 2, which private domain 2 is given priority. Can be changed.
In addition, the operation monitoring processing unit 30 performs processing for monitoring the state of other functional components of the public domain 1.

  Next, in the program placement system of the present embodiment, by managing the configuration management information of the business AP and the configuration management information of the resource, management for each resource used for each business AP is performed, and the resource abnormality for each business AP is detected. A process for realizing the stable operation of the business AP by executing resource management suitable for the business such as detection will be described.

  The configuration management processing unit 41 of the private domain management server 11 of the present embodiment maintains and manages the configuration management information of each resource by referring to the management information collected by each processing unit, and determines the resource used by each business AP 48. By referring to the configuration management information including the identification information and reading the configuration management information of the resource identified by the identification information from the configuration management information of each resource, the configuration management information of the business AP 48 and the searched resource When the business AP information associated with the configuration management information is generated and the resource configuration management information in the business AP information corresponds to a rule for detecting a resource abnormality in the corresponding business AP 48, the business AP 48 The process of outputting information indicating that a resource abnormality has occurred is performed. By taking countermeasures such as clustering to a server that satisfies the operation condition for this resource abnormality, stable operation of the business AP can be realized.

  FIG. 15 is a diagram showing an outline of management information association according to the present embodiment. As shown in FIG. 15, (1) server name match, (2) server name match, (3) logical host name corresponding to the number, (4) DB name match, (5) business AP name The management information is correlated by searching for matching information such as matching, (6) data path name matching, and (7) storage name and storage logical unit name matching.

  FIG. 16 is a diagram showing the processing contents of the management information association processing of this embodiment. In step 1601, the configuration management processing unit 41 of the private domain management server 11 performs loop control for the number indicated by the number of business APs 201.

  In step 1602, it is checked whether or not the business AP execution server name 223 matches the server name 330 of the server management information. If they match, the processing in step 1603 and subsequent steps is performed.

  In step 1603, after an area for storing the values of flag 1, flag 2, and flag 3 is secured in the memory, the value of each flag is set such that flag 1 = 0, flag 2 = 0, flag 3 = 0. Set.

  In steps 1604 to 1608, OS information research, logical host name search, DB information search, cluster / data path information search, and storage management information search are executed.

  In step 1609, the value of the flag 1 is checked. If the value is “0”, the process proceeds to step 1610. If not, the process proceeds to step 1611.

  In step 1610, business AP information in which the configuration management information of the business AP 48 is associated with the extracted resource configuration management information is generated based on the information extracted as successful extraction of related information. On the other hand, in step 1611, the fact that the association has failed is output to an output device such as a display device to notify the administrator.

  FIG. 17 is a diagram showing the processing contents of the OS information search processing of the present embodiment. In step 1701, the value of flag 1 set in the process so far is checked. If the value is “0”, the process proceeds to step 1702.

  In step 1702, it is checked whether or not there is a match between the server name 330 of the server hardware information 311 and the server name 363 of the OS information 312. If there is a match, the process proceeds to step 1703; Proceed to 1704. In step 1703, the matched OS information 312 is extracted, and in step 1704, “1” is set in the flag 1.

  FIG. 18 is a diagram showing the processing contents of the logical host name search processing of this embodiment. In step 1801, it is checked whether or not flag 1 is “0”. If flag 1 is “0”, the process proceeds to step 1802.

  In step 1802, it is checked whether a logical host name that matches the logical host name 239 in the AP operating condition 212 exists in the logical host names 370 to 372 in the OS information 312. If there is a match, the process proceeds to step 1803. If not, the process proceeds to Step 1804.

  In step 1803, the matched logical host name is extracted, and in step 1804, “1” is set in flag 1.

  FIG. 19 is a diagram showing the processing contents of the DB information search processing of this embodiment. In Step 1901, it is checked whether or not the flag 1 is “0”. If the flag 1 is “0”, the process proceeds to Step 1902.

  In step 1902, it is checked whether there is a DB name 390 and a business AP 392 that match the combination of the business AP name 220 of the AP operating condition 211 and the used DB name 242 of the AP operating condition 212. Proceed to step 1903, otherwise proceed to step 1904.

  In step 1903, the matched DB information 390 to 401 is extracted, and in step 1904, “1” is set to flag 1.

  FIG. 20 is a diagram showing the processing contents of the cluster / data path information search processing of this embodiment. In step 2001, it is checked whether or not flag 1 is “0”. If flag 1 is “0”, the process proceeds to step 2002.

  In Step 2002, it is checked whether or not the cluster operation presence / absence 500 is “None”. If it is “None”, the process proceeds to Step 2003. In Step 2003, a return value indicating that there is no cluster operation is set. To do.

  In step 2004, it is checked whether or not there is a business AP name that matches the business AP name 220 in the AP operation status 211. If there is a match, the process proceeds to step 2005. If not, the process proceeds to step 2006.

  In step 2005, the matched cluster information 510 to 514 is extracted, and in step 2006, “1” is set in the flag 3 as being out of cluster operation target.

  In step 2007, it is checked whether there is a data path name 520 that matches the matched operation data path name 514. If there is a match, the process proceeds to step 2008. Otherwise, the process proceeds to step 2009.

  In step 2008, the matched data path information (520 to 523) is extracted, and in step 2009, “1” is set in the flag 2 as being out of storage operation targets.

  FIG. 21 is a diagram showing the processing contents of the storage management information search processing of this embodiment. In step 2101, it is checked whether flag 1 is “0” and flag 2 is “0”. If both flags 1 and 2 are “0”, the process proceeds to step 2102.

  In step 2102, it is checked whether the storage name 611 and storage logical unit name 620 of the storage management information 610 that match the combination of the extracted storage name 522 and storage logical unit name 523 exist. Otherwise, go to Step 2104.

  In step 2103, the matched storage information 620 to 631 is extracted, and in step 2104, “1” is set to flag 1.

  FIG. 22 is a diagram showing an example of business AP information associated with the present embodiment. As shown in FIG. 22, in the business AP information, the configuration management information of each business AP and the resource configuration management information are associated with each other.

  With reference to the policy management rule for each business AP for determining the status of each business AP, the resource configuration management information in the business AP information corresponds to a rule for detecting a resource abnormality in the corresponding business AP. It is possible to determine whether or not a resource abnormality has occurred for the business AP. When a resource abnormality is determined, for example, display of * 1 to 3 in FIG. And output information indicating that a resource abnormality has occurred in each business AP.

  FIG. 23 is a diagram illustrating an example of problems and countermeasures of the business AP of FIG. 22 according to the present embodiment. FIG. 23 shows problems of the business AP shown in FIG. 22 and countermeasure examples. A policy management rule is used when detecting a resource abnormality as shown in the problem of FIG.

  FIG. 24 is a diagram showing an example of policy management rules of this embodiment. FIG. 24 shows an example of policy management rules used when generating the problems and countermeasures of FIG.

  Next, in the program placement system according to the present embodiment, processing for distributing the policy management rules as described above to each functional component and performing integrated management so that the program placement can be automatically operated will be described.

  In this embodiment, the policy management processing unit 42 of the private domain management server 11 and the policy management processing unit 32 of the public domain management server 21 detect a rule for detecting a resource abnormality in the business AP 48 and a rule for detecting a surplus resource. A process for centrally managing policy management rules including

  FIG. 25 is a diagram showing the processing contents of the policy management rule distribution processing of this embodiment. In step 2501, the policy management processing unit 42 of the private domain management server 11 checks whether there is a policy management rule that has been added, deleted, or changed.

  In step 2502, it is checked whether or not the distribution destination domain is the own domain. If it is the own domain, the process proceeds to step 2503, and if not, the process proceeds to step 2509.

In step 2503, it is checked whether or not the processing content to the policy management rule is the case of adding a policy management rule. In step 2504, the added policy management rule is distributed to the distribution destination.
In step 2505, it is checked whether or not the processing content of the policy management rule is a case of deleting the policy management rule. If it is a deletion, the process proceeds to step 2506. In step 2506, the deleted policy management rule is deleted from the policy management rule of the distribution destination.

  In step 2507, it is checked whether or not the processing content to the policy management rule is a change of the policy management rule. If the change is a change, the process proceeds to step 2508. In step 2508, the changed policy management rule is overwritten and distributed to the distribution destination (replaced with the distribution-destination policy management rule).

  In step 2509, the policy management rule is distributed to the policy management processing unit of another domain.

  Next, execution control for selecting a resource suitable for the operating condition of the business AP at the start by associating the business AP configuration management information with the resource configuration management information in the program placement system of the present embodiment will be described.

  The execution management processing unit 45 of the private domain management server 11 of the present embodiment receives an execution request from the business AP 48 and instructs reservation of resources used by the business AP 48, and uses the resource reserved by the instruction. Processing for starting the AP 48 is performed. In addition, when the configuration management information of the resource associated with the business AP 48 for which the reservation instruction has been given satisfies the operating condition in the configuration management information of the business AP 48, the space scheduler 44 executes the resource AP as a resource for executing the business AP 48. Process to reserve.

  Then, the distribution management processing unit 43 performs a process of distributing a program such as the activated business AP 48 or DB management system to the real business server 49 corresponding to the reserved resource.

  FIG. 26 is a diagram showing the processing contents of the execution management process when the business AP is activated according to this embodiment. In step 2601, the execution management processing unit 45 of the private domain management server 11 checks whether or not a business AP execution request has been received, and proceeds to step 2602 if received.

  In step 2602, resource reservation processing is performed for the spatial scheduler 44. In step 2603, it is checked whether the resource can be reserved by the processing in step 2602. If the resource can be reserved, the process proceeds to step 2604. Proceed to

  In step 2604, a business execution server name and a scheduled business AP start time are input from the space scheduler 44. In step 2605, schedule reservation information for executing the business AP at the scheduled start time is generated on the input business execution server. In 2606, the business AP start time 258 is recorded, and in step 2607, after the business AP ends, the business AP end time 259 is recorded.

  In step 2608, a response indicating that there is no business execution server capable of executing the business AP is issued, and a retry is instructed.

  FIG. 27 is a diagram illustrating an image of resource schedule reservation according to the present embodiment. As shown in FIG. 27, in this embodiment, a resource schedule is reserved.

  FIG. 28 is a diagram showing the processing contents of the spatial scheduler 44 of the private domain 2 of this embodiment. In step 2801, the configuration management processing unit 41 is checked for resource usage status in units of business APs.

  In step 2802, the contents of the received event are checked, and if a resource reservation notification is received, the process proceeds to step 2803. If a resource abnormality notification is received, the process proceeds to step 2806, and an unused server notification is received. If YES in step 2808, the flow advances to step 2808. If a policy management rule notification is received, the flow advances to step 2810.

  In step 2803, the contents of the resource reservation are received from the execution management. In step 2804, the confirmation process of the business AP execution server is executed. In step 2805, the resource schedule reservation process is executed.

  In step 2806, the contents of the resource abnormality are received from the configuration management. In step 2807, the business AP is specified and the countermeasure processing is executed.

  In step 2808, the unused server identification information is received from the configuration management. In step 2809, a surplus server determination process is executed.

  In step 2810, a policy management rule for the spatial scheduler is input and reflected.

  FIG. 29 is a diagram showing the processing contents of the confirmation processing of the business AP execution server of this embodiment. In step 2901, it is checked whether or not the business AP execution user name 224 specified in the user name 701 of the OS information 312 of the server of the specified business AP execution server name 223 exists. .

In Step 2902, it is determined whether or not the resource of the data path name 232 to the use DB name 242 specified in the AP operating condition 212 of the specified business AP exists as a free resource in the specified business AP execution server name 223. If it does not exist, go to step 2903.
In step 2903, change processing of the business execution server is executed.

  In step 2904, it is checked whether or not the flag is “0”. If the flag is “0”, the process proceeds to step 2905. In step 2905, logical host confirmation processing is performed.

  FIG. 30 is a diagram showing the processing contents of the change processing of the business execution server of this embodiment. In step 3001, “1” is set to flag 1 indicating that there is no corresponding server. In step 3002, while the flag is “1”, processing for looping the number of other server names 330 is performed.

  In step 3003, it is checked whether the resources of the data path name 232 to the used DB name 242 specified in the AP operating condition 212 of the specified business AP exist in the OS information of other server names. Proceed to step 3004.

  In step 3004, it is checked whether or not the business AP execution user name 224 specified in the user name 701 of the OS information 312 of the corresponding server exists, and if it exists, the process proceeds to step 3005.

  In step 3005, “0” is set in the flag. In step 3006, the corresponding server name is set as the business AP execution server name 223, and the corresponding server is selected as the business execution server.

  In step 3007, it is checked whether or not the flag is “1”. If the flag is “1”, the process proceeds to step 3008. In step 3008, a request is made to the public domain spatial scheduler 34 to provide a server that matches the AP operating conditions of the designated business AP.

  FIG. 31 is a diagram showing the processing contents of the logical host confirmation processing of this embodiment. In step 3101, the logical host name 239 of the designated business AP is acquired.

  In step 3102, it is checked whether there is any logical host name 370 to 372 of the specified server name 330 that matches the logical host name 239. If there is no match, the process proceeds to step 3103.

  In step 3103, it is checked whether or not there is a match with the logical host name 239 in the logical host names 370 to 372 of the OS information 312 of other server names. If there is a match, the process proceeds to step 3104.

  In step 3104, the matched logical host name 239 is deleted from the OSs of other server names. In step 3105, the logical host name 239 is generated in the OS of the business execution server.

  FIG. 32 is a diagram showing the processing contents of the resource schedule reservation processing of this embodiment. In Step 3201, it is checked whether or not the flag is “0”. If the flag is “0”, the process proceeds to Step 3202.

  In Step 3202, it is checked whether or not the business AP time designation presence / absence 225 of the designated business AP is “Yes”. If “Yes”, the process proceeds to Step 3203, and if not, the process proceeds to Step 3206.

  In step 3203, it is checked whether there are any free resources in the data path 232 to the disk capacity 238 specified in the AP operation condition 212 of the business AP specified in the space scheduler 44. Otherwise, go to step 3205.

  In step 3204, resources are reserved in the space scheduler 44 by specifying the time, and in step 3205, “2” indicating that scheduling is impossible is set in the flag.

  In step 3206, resources are reserved in the space scheduler 44 for an indefinite resource reservation time.

  Next, in the program placement system of the present embodiment, execution control for performing clustering by selecting resources suitable for the operating conditions of the operating business AP by associating the business AP configuration management information with the resource configuration management information. Will be described.

  The execution management processing unit 45 of the private domain management server 11 of the present embodiment, when an abnormality is detected in the resource that is executing the business AP 48, sets the operating condition in the configuration management information of the business AP 48 in which the resource abnormality has occurred. A search is made for other resources having configuration management information that is satisfied, and processing for continuing the processing of the business AP 48 is performed using the searched resources. Then, the distribution management processing unit 43 distributes a program such as a business AP 48 or a DB management system in which a resource abnormality has occurred to the real business server 49 of the searched resource.

  FIG. 33 is a diagram showing the processing contents for identifying and dealing with the business AP of this embodiment. In step 3301, a business server in which a resource abnormality has occurred is specified. In step 3302, a list of business APs operating on the server is created.

  In step 3303, the resource abnormality is checked. If the server is abnormal, the process proceeds to step 3304. If the DB is abnormal, the process proceeds to step 3305. If the storage is abnormal, step 3308 is performed. Proceed to

In step 3304, clustering processing of business APs is performed.
In step 3305, DB recovery processing is performed. In step 3306, whether or not the DB abnormality has been removed by the recovery processing is checked. If it cannot be removed, business AP clustering processing is performed in step 3307.
In step 3308, the capacity of the storage logical unit is added.

FIG. 34 is a diagram showing the processing contents of the clustering processing of the business AP of this embodiment. In step 3401, “0” is set in the flag, and in step 3402, business APs for which the clustering availability flag 240 is allowed are extracted from the business AP list, and a list is created in order of decreasing business AP importance.
In step 3403, while the flag is “0”, processing is performed for the number of extracted business APs, and in step 3404, processing is performed for the number of business AP names 510 while the flag is “0”. Do.

  In step 3405, it is checked whether there is a business AP name that matches the extracted business AP. If there is a match, the process proceeds to step 3406. In step 3406, a clustering server name confirmation process is executed.

  FIG. 35 is a diagram showing the processing contents of the clustering server name confirmation processing of the present embodiment. In step 3501, it is checked whether or not there is an operating server name that does not match the corresponding server, and if there is, the process proceeds to step 3502.

  In step 3502, it is checked whether or not there is a vacancy in the AP operation condition 212 of the extracted business AP in the resource of the OS information 312 of the operating server name that does not match. If there is a vacancy, the process proceeds to step 3503.

  In step 3503, “1” is set in the flag 1, and in step 3504, an instruction to perform clustering to the corresponding operation server is requested to the cluster / data path management.

  Next, in the program arrangement system of this embodiment, a process for managing surplus resources by the private domain and the public domain and providing surplus resources suitable for operating conditions when a business AP is started or when a resource is abnormal will be described.

  FIG. 36 is a diagram illustrating processing contents of the surplus server determination processing according to the present embodiment. As shown in FIG. 36, the space scheduler 44 of the private domain management server 11 performs processing for determining surplus resources.

  In step 3601, an unused server is input, and in step 3602, a spatial scheduler policy is input.

  In step 3603, the unused server checks whether the machine usage form 231 is a business server on which a dedicated business AP operates. If the unused server is a business server on which a dedicated business AP operates, the process proceeds to step 3604, and the server is made redundant. Return to the caller without being a server.

  In step 3605, the contents of the resource reservation are read. In step 3606, it is checked whether there is a resource reservation for the server. If there is a resource reservation, the process proceeds to step 3607. If there is no resource reservation, the process proceeds to step 3608. .

  In step 3607, resource reservation confirmation processing is executed. In step 3608, the server is determined to be a surplus server. In step 3609, the configuration management information is provided to the public domain space scheduler as surplus server information.

  FIG. 37 is a diagram showing the processing contents of the resource reservation confirmation processing of this embodiment. In step 3701, it is checked whether or not there is one hour or more until the resource reservation. If there is one or more hours, the process proceeds to step 3702.

  In step 3702, the server is determined to be a surplus server. In step 3703, the configuration management processing unit 41 is provided as a surplus server until the resource reservation time.

  In step 3704, it is checked whether or not the resource reservation time is indefinite. If it is indefinite, the process proceeds to step 3705.

  In step 3705, the AP operation history of the business AP is referred to. In step 3706, it is checked whether the business AP end time of the business AP is recorded. If the business AP end time is recorded, step 3707 is performed. Proceed to In step 3707, the resource reservation of the business AP is canceled.

  FIG. 38 is a diagram showing the processing contents of the configuration management processing unit 41 of the private domain 2 of this embodiment. In step 3801, management information association processing is executed.

  In step 3802, an event is received. In step 3803, the content of the received event is examined. If the received event is a management information acquisition request, the process proceeds to step 3804. If the received event is a server return, step 3806 is performed. If the resource is abnormal, the process proceeds to step 3807. If it is an unused server, the process proceeds to step 3810. If it is a surplus server, the process proceeds to step 3811. If it is a policy management rule, the process proceeds to step 3813. Proceed to

In step 3804, a management information request receiving process is executed. In step 3805, the management information is returned to the request source.
In step 3806, release of the returned server configuration information to the public domain is stopped.

  In step 3807, it is checked whether or not the abnormal resource received from the performance / capacity has been provided to the public domain. If it has been provided, the process proceeds to step 3808, and if not, the process proceeds to step 3808.

  In step 3808, the resource abnormality is notified to the configuration management processing unit 31 in the public domain, and in step 3809, the resource abnormality is notified to the space scheduler 44 and the cluster / data path management.

In step 3810, the unused server is notified to the space scheduler.
In step 3811, the surplus server is notified to the public domain configuration management, and in step 3812, the surplus server management information is disclosed to the public domain configuration management processing unit 31.
In step 3813, the policy management rule is received to set the ACL and the safety factor.

  FIG. 39 is a diagram showing the processing contents of the management information request reception processing of this embodiment. In step 3901, the content of the requested management information is checked. If the requested management information is AP operating status, the process proceeds to step 3902. If the requested management information is the AP operating condition, the process proceeds to step 3903, where DB information is obtained. If YES in step 3904, the flow advances to step 3904. If it is cluster / data path information, the flow advances to step 3905. If it is other information, the flow advances to step 3907.

  In step 3902, an operator setting value is acquired for information other than the business AP state of the AP operation status 211, and in step 3903, an operator setting value is acquired for the AP operation condition 212.

  In step 3904, information is acquired from the DB. In step 3905, information is acquired from the class management program. In step 3906, data path information is acquired from the data path management program. In step 3907, performance / capacity management is performed. Get information.

  The configuration management processing unit 31 of the public domain management server 21 according to the present embodiment receives the configuration management information of the resource corresponding to the rule for detecting surplus resources from the configuration management processing unit 41 on the private domain 2 side, and the public domain 1 is maintained and managed as the configuration management information of the virtual resource 38 in 1.

  FIG. 40 is a diagram illustrating processing contents of the configuration management processing unit 31 of the public domain 1 according to the present embodiment. In step 4001, an event is received. In step 4002, the received event is examined. If the received event is a surplus server, the process proceeds to step 4003. If the received event is a server return, the process proceeds to step 4005. If it is a rule, the process proceeds to step 4008.

In step 4003, the received surplus server configuration information is acquired from the configuration management of the private domain.
In step 4004, the surplus server is notified to the spatial scheduler 34. In step 4005, a return of the server is received from the spatial scheduler 34. In step 4006, the returned server is returned to the provided private domain. In step 4007, the returned server configuration information is deleted, and in step 4008, policy management rules are received and set.

  The spatial scheduler 34 of the public domain management server 21 according to the present embodiment sends a resource provision request sent from the spatial scheduler 44 on the private domain 2 side when a resource having configuration management information that satisfies the operating conditions of the business AP 48 is not found. It receives and transmits the configuration management information that satisfies the operating conditions of the business AP 48 in the configuration management information of the virtual resource that is maintained and managed.

  FIG. 41 is a diagram showing the processing contents of the spatial scheduler 34 in the public domain 1 of this embodiment. In step 4101, a surplus server is acquired from the configuration management processing unit 31.

  In Step 4102, a server provision request is received from the private domain spatial scheduler 44. In Step 4103, it is checked whether there is a surplus server provided from the requested private domain, and if there is, proceed to Step 4104. Otherwise, go to step 4105.

  In step 4104, the corresponding surplus server is lent out. In step 4105, it is checked whether there is a surplus server provided from another private domain. If there is, the process proceeds to step 4106. If not, the process proceeds to step 4107. . In step 4106, the surplus server is lent out. In step 4107, a response indicating that there is no lent server is returned.

In step 4108, the return of the lent server is received from the space scheduler 44 of the private domain, and in step 4109, the return of the server is notified to the configuration management processing unit 31 of the public domain.
In step 4110, a policy management rule is received and set.

  FIG. 42 is a diagram illustrating processing contents of the performance / capacity management processing unit 47 of the private domain 2 according to the present embodiment. As shown in FIG. 42, the performance / capacity management processing unit 47 of the private domain management server 11 performs processing for collecting and managing management information indicating the status of each server 49, the capacity of the storage 50, and the like.

  In step 4201, an event is received. In step 4202, the received event is examined. If the received event is a management information acquisition request, the process proceeds to step 4203. If the received event is a policy management rule, the process proceeds to step 4204. If it is resource monitoring, the process proceeds to step 4205. If it is another event, the process proceeds to step 4211.

  In step 4203, the requested management information is acquired from the business AP 48, server 49, and storage 50 and responded. In step 4204, the policy management rule is received and set.

  In step 4205, the history is saved, and in step 4206, it is checked whether there is a resource whose resource usage rate exceeds 95%. If there is, the process proceeds to step 4207. This value of 95% of the resource usage rate can be changed by policy management rules.

  In step 4207, the corresponding resource is examined. If the corresponding resource is a server, the process proceeds to step 4208. If the corresponding resource is a storage, the process proceeds to step 4209. If the corresponding resource is a DB, the process proceeds to step 4210.

In step 4208, a resource abnormality of the server is notified to the configuration management processing unit 41. In step 4209, a resource abnormality of insufficient storage capacity is notified to the configuration management processing unit 41. Notify the resource error.
In step 4211, a resource monitoring event is transmitted at regular intervals.

  FIG. 43 is a diagram showing the processing contents of the cluster / data path management processing unit 46 in the private domain 2 of the present embodiment. As shown in FIG. 43, the cluster / data path management processing unit 46 of the private domain management server 11 manages the configuration and status of the class management program and the configuration of the data path between the server 49 and the storage 50 of the data path management program. And a process for managing the state and data path change instructions.

  In step 4301, an event is received. In step 4302, the received event is examined. If the received event is information acquisition, the process proceeds to step 4303. If the received event is a clustering request, the process proceeds to step 4304. If it is a rule, the process proceeds to step 4309.

In step 4303, the requested information is returned.
In step 4304, a clustering request is received from the spatial scheduler. In step 4305, the current business AP execution server name and the clustering destination business AP execution server name are acquired.

In step 4306, DB change processing is executed. In step 4307, data path change processing is executed. In step 4308, the business AP is started on the clustering destination business AP execution server.
In step 4309, the policy management rule is received and set.

  FIG. 44 is a diagram showing the processing contents of the DB change processing of this embodiment. In step 4401, it is checked whether or not the use DB used by the corresponding business AP name 220 exists in the clustering destination business AP execution server.

  In step 4402, it is checked whether or not the corresponding DB is operating. If not, the process proceeds to step 4403. In step 4403, the corresponding DB is activated.

  In step 4404, the corresponding DB is distributed from the private domain management server to the clustering destination business AP execution server. In step 4405, the corresponding DB is activated.

  FIG. 45 is a diagram showing the processing contents of the data path change processing of this embodiment. In step 4501, it is checked whether or not the storage logical unit used by the corresponding business AP name 220 can be referred (whether there is a data path) from the clustering destination business AP execution server.

  In step 4502, it is checked whether or not the storage logical unit can be referred to from the corresponding DB. In step 4503, the DB setting is changed so that the storage logical unit can be referred to from the DB.

  In step 4504, the data path setting is changed so that the storage logical unit can be referred to from the data path management. In step 4505, it is checked whether or not the storage logical unit can be referred to from the corresponding DB. In step 4506, the DB setting is changed so that the storage logical unit can be referred to from the DB.

  Next, a process for predicting a change in the operating state of a resource and changing a policy management rule in the program placement system of this embodiment will be described.

  FIG. 46 is a diagram showing the processing contents of the policy management processing that reflects the prediction result of this embodiment. In step 4601, an initial value of the upper limit of resource use is input for each business AP, each server, and each resource.

  In step 4602, configuration management information is input. In step 4603, it is checked whether there is a business AP, server, or storage that exceeds the resource use upper limit.

  In step 4604, a resource use upper limit correction process is executed. In step 4605, it is checked whether or not there is a policy management rule for the business AP. If it exists, the process proceeds to step 4606, and if not, the process proceeds to step 4607.

  In step 4606, the upper limit value of the policy management rule for the business AP is changed. In step 4607, processing for registering the policy management rule for the business AP is performed.

  In step 4608, a resource usage trend prediction process is executed. In step 4609, a resource management rule distribution process is executed.

  FIG. 47 is a diagram showing an example of grasping the resource usage status in units of machines and business APs according to this embodiment. As shown in FIG. 47, in this embodiment, the resource usage status is displayed from the history information, and the resource usage status is grasped for each machine and business AP.

  FIG. 48 is a diagram showing an example of a resource allowable range in units of machines and business APs according to this embodiment. As shown in FIG. 48, in this embodiment, a resource allowable range is obtained from the resource usage rate for each machine and business AP, and an upper limit value is set from the allowable range.

  FIG. 49 is a diagram showing a correction example of the resource use upper limit value for each machine and business AP of this embodiment. As shown in FIG. 49, in the present embodiment, a long-term fluctuation tendency is obtained from the fluctuation of the history information as shown in the “tilt” in the figure, and is used for correcting the resource use upper limit value in units of machines and business APs.

  FIG. 50 is a diagram showing the processing contents of the resource use upper limit correction processing of this embodiment. In step 5001, the AP operation history (FIG. 8) of the corresponding business AP is input, and in step 5002, the upper limit value of stable operation is calculated (FIG. 48).

  In step 5003, a process for looping the resource items is performed. In step 5004, it is checked whether an abnormality has occurred. If it has occurred, the process proceeds to step 5005.

  In step 5005, the resource usage rate at the time of occurrence of an abnormality is input. In step 5006, it is checked whether or not the business AP termination status in the latest history of the relevant business AP at the resource usage rate at the time of occurrence of abnormality is normal. If completed, go to Step 5007.

In step 5007, correction of the upper limit value of the resource usage of the corresponding business AP (example in FIG. 48, the upper limit value of the stable operation of the memory is corrected from 45% to 65%)
FIG. 51 is a diagram showing the processing contents of the resource usage trend prediction processing of the present embodiment. In step 5101, a process for looping the server resources and storage resources is performed. In step 5102, it is checked whether there is a resource whose resource usage rate is increasing.

In step 5103, the slope of the resource usage rate with respect to the future time is calculated,
Formula (future resource usage) = (latest value) x slope x (future time-current time)
In step 5104, the reflection process of the predicted value is performed.

  FIG. 52 is a diagram showing the processing contents of the predicted value reflection processing of this embodiment. In step 5201, it is checked whether or not the resource usage rate exceeds the action boundary value at the next history time.

  In step 5202, it is checked whether or not the current resource usage rate exceeds the action boundary value. If so, the process proceeds to step 5203, and if not, the process proceeds to step 5205.

  In Step 5203, it is checked whether or not an abnormality has occurred in the business AP operating on the corresponding resource. If no abnormality has occurred, the process proceeds to Step 5204.

  In step 5204, the action boundary value of the policy management rule for the corresponding resource is changed to the current resource utilization rate. In step 5205, the action boundary value of the policy management rule for the corresponding resource is changed to the current resource utilization rate.

  Note that the action boundary value is a resource usage rate at which an abnormality occurs, which is separately obtained through experiments or simulations.

  As described above, according to the program placement system of this embodiment, business application information in which business application configuration management information and resource configuration management information are associated is generated, and resources used are managed in units of business applications. It is possible to manage resources according to the operating conditions of each business application.

  DESCRIPTION OF SYMBOLS 1 ... Public domain, 2 ... Private domain, 3 ... Network, 10 ... Real business server group, 11 ... Private domain management server, 12 ... Private domain structure information, 13 ... Real storage, 20 ... Virtual business server group, 21 ... Public Domain management server, 22 ... public domain configuration information, 23 ... virtual storage, 38 ... virtual resource, 48 ... business AP, 49 ... server, 50 ... storage, 30 ... operation monitoring processing unit, 31 ... configuration management processing unit, 32 ... Policy management processing unit 34... Spatial scheduler 40. Operation monitoring processing unit 41. Configuration management processing unit 42. Policy management processing unit 43. Distribution management processing unit 44. 46: Cluster / data path management processing unit, 47 ... Performance / capacity management Management unit, 70 ... Business AP, 71 ... Web application, 72 ... Batch job, 73 ... Workflow, 80 ... Server, 81 ... DB management processing unit, 82 ... Cluster / data path management processing unit, 83 ... OS, 90 ... Storage 228 ... Business AP importance 258 ... Business AP start time 259 ... Business AP end time 352 ... Maximum memory amount 353 ... Used memory amount 354 ... Remaining memory amount 357 ... Remaining process number 360: Remaining thread Number, 365 ... remaining disk capacity.

Claims (6)

Business in a business management system that has multiple private domains consisting of a public domain consisting of a public domain management server and virtual resources, a private domain management server, a real business server that executes business, and real storage that stores data used in the business A resource management method,
The private domain management server notifies the public domain that there is a surplus resource when there is a surplus resource,
The public domain management server associates the surplus resource as a virtual resource,
The public domain management server receives a resource provision request from the private domain management server,
Determine if there is a virtual resource available to the requesting private domain,
When there is the virtual resource, a business resource management method comprising: notifying that the surplus resource corresponding to the virtual resource is lent to the private domain that is a request source. In Claim 1, when the surplus resource is present and the surplus resource is a surplus resource provided from a private domain as a request source, the surplus resource is lent.
A business resource management method characterized by lending out a surplus resource provided from another private domain when there is no surplus resource provided from the requesting private domain.   3. The business resource management method according to claim 2, wherein the private domain management server determines whether a resource is the surplus resource based on a usage pattern of a machine and a resource reservation state. A business management system with multiple private domains consisting of a public domain consisting of a public domain management server and virtual resources, a private domain management server, a real business server that executes business, and a real storage that stores data used in the business. And
The private domain management server includes a first configuration management processing unit that notifies the public domain that there is a surplus resource when there is a surplus resource,
The public domain management server includes a second configuration management processing unit that associates the surplus resource as a virtual resource, and a virtual resource that can receive a resource provision request from the private domain management server and provide the request to the private domain And a means for notifying that the surplus resource corresponding to the virtual resource is lent to the private domain that is the request source.   5. The method according to claim 4, wherein when the surplus resource is present and the surplus resource is a surplus resource provided from a private domain as a request source, the surplus resource is lent out and the request source private A business management system characterized by lending out surplus resources provided from another private domain when there is no surplus resources provided from a domain.   6. The business management system according to claim 5, wherein the private domain management server includes means for determining whether a resource is the surplus resource based on a usage pattern of a machine and a resource reservation state.

Images