WO2011138879A1

WO2011138879A1 - Operation management device and operation management method of information processing system

Info

Publication number: WO2011138879A1
Application number: PCT/JP2011/053044
Authority: WO
Inventors: 裕和内
Original assignee: 株式会社日立製作所
Priority date: 2010-05-06
Filing date: 2011-02-14
Publication date: 2011-11-10
Also published as: JP2013178592A

Abstract

System operation management is needed such that, for an information processing system, failure is unlikely to occur, and in particular, failure imparting a large effect on business operation is unlikely to occur. An operation management device connects to an information processing system for which the configuration is to be managed. The operation management device acquires, from an information processing system in operation, configuration information of the information processing system. From the acquired configuration information, if failure has occurred in a resource constituting the information processing system, the operation management device calculates a degree of effect indicating an effect imparted to business operations of the information processing system for each resource. If there is a resource such that the calculated degree of effect exceeds a predefined upper bound of degree of effect, the operation management device creates configuration plans of the information processing system such that the degree of effect for the resource falls under the upper bound. The configuration plans of the information processing system which are to be created are configuration modification plans of a decentralized configuration and a redundancy configuration of the information processing system such that the degree of effect for the resource falls under the upper bound.

Description

Operation management apparatus and operation management method for information processing system

The present invention relates to prevention of failures in an information processing system, and more particularly to a technique for preventing in advance failures that may occur based on configuration information of the information processing system.

With the increase in the size of information processing systems, operations of services and services by execution of business programs (application programs) in information processing systems and efficient operations of information processing system resources are separated. Business operation management for the former executes business programs according to the operation schedule and load so that processing requests and service requests from users of the information processing system can be met with a predetermined performance (for example, response time). It controls the selection of resources such as computers, storage devices, and network interfaces, and start / stop of business programs. System operation management for the latter manages the resource configuration of the information processing system, monitors the operating status, monitors the occurrence of failures due to overloads and failures, and determines the extent and extent of the impact when a failure occurs. Control suppression. System operation management must be implemented under the constraints on the performance and number of resources associated with the cost performance and operation plan of the information processing system.

Patent Document 1 discloses a technique related to control when a failure occurs in an information processing system. Patent Document 1 shows that the degree of influence of a resource on a service is calculated in consideration of the state of the system, and the priority of the resource is calculated based on the importance of the service and the calculated degree of influence. Yes. As a result, when a failure occurs in a resource used by a service with high importance, it becomes possible to allocate a resource used by a service with low importance to a service with high importance.

JP 2008-217285 A

In recent years, with the advancement of virtualization technologies such as computers and storage devices, the scope of control for both business operation management and system operation management (targets of management, configuration and procedure options for controlling targets, etc.) has expanded. ing. In an information processing system in which business operation management and system operation management are separately operated, managers for the information processing system are also divided. Here, a business operation management administrator is referred to as a business administrator, and a system operation management administrator is referred to as a system administrator.

Under such circumstances, the result of the business operation management of the information processing system by the business administrator may easily cause a failure in the information processing system, or may cause a system state that has a large effect when a failure occurs. . Patent Document 1 discloses a technique for dealing with a case where a failure occurs. However, a system operation management that is unlikely to cause a failure and particularly has a large influence on business operations is desired.

The disclosed operation management device is connected to an information processing system whose configuration is to be managed. Connect to the management client. The operation management device receives a configuration information acquisition unit that acquires configuration information of an information processing system from an operating information processing system, and processes information when a failure occurs in a resource that constitutes the information processing system from the acquired configuration information. The impact of a resource when there is an impact calculator that calculates the impact that indicates the impact on the performance of the system for each resource, and when the calculated impact exceeds a predetermined upper limit of impact. A configuration generation unit that generates a configuration plan of the information processing system whose degree is lower than the upper limit value;

Another desirable mode of the disclosed operation management apparatus is a configuration change plan of a distributed configuration and a multiplexed configuration of an information processing system in which the influence degree of resources falls below the upper limit value.

Still another desirable mode of the disclosed operation management apparatus is that the impact calculation unit, when a failure occurs in a resource constituting the information processing system, affects other resource impacts that the resource affects and the resource The sum of importance is calculated as the influence of the resource.

Still another desirable mode of the disclosed operation management apparatus is a multiplicity calculation unit that corrects the calculated influence degree with respect to resources that are multiplexed in the resources for which the influence degree calculation unit has calculated the influence degree. It has further.

Still another desirable mode of the disclosed operation management apparatus includes an alert policy setting unit that sets an upper limit value of a predetermined influence level input from a management client, and an influence level calculated for each resource has a predetermined influence level. An alert detection unit that determines whether or not the upper limit value is exceeded, and an alert output unit that outputs an alert to the management client when the calculated influence degree exceeds a predetermined upper limit value of the influence degree.
Other problems disclosed by the present application and the solutions thereof will be clarified by the embodiments of the present invention and the drawings.

According to the present invention, it is possible to perform system operation management in which a failure is unlikely to occur in an information processing system, and particularly a failure that has a large influence on business operations is unlikely to occur.

It is a system configuration example of an embodiment. It is a figure which shows each part and each table structure of the system of embodiment. It is a processing flow of a structure control part. It is an example of a structure information table. It is an example of a structure information table. It is an example of a structure information table. It is an example of a resource related information table. It is an example of a structure confirmation screen. It is an example of an importance definition table. It is an example of an influence information table. It is an example of a dependency relationship definition table. It is a processing flow of an influence degree calculation part. It is a processing flow of an influence degree calculation part. It is an example of an influence confirmation screen. It is an example of a multiplicity definition table. It is an example of an alert policy table. It is a processing flow of alert determination and issue. It is a processing flow of a recommendation composition generation part. It is a flowchart of a dispersion | distribution structure production | generation process. It is a flowchart of the resource search process which can be allocated. It is a flowchart of a multiplicity structure production | generation process. It is an example of a decentralized configuration table. It is an example of a multiplexing structure table. It is an example of a recommended composition display screen.

FIG. 1 shows a system configuration example of the embodiment. The system according to this embodiment includes a management target device 1 including one or more server devices 2 and one or more switches (network switches) 4, a configuration management server 5, and a management client 6 connected via a communication network. The configuration management system manages and controls the configuration of the management target device 1 as an information processing system managed by the business manager.

The configuration management server 5 (operation management device) is connected to the management target device 1 which is a target for managing the configuration. The configuration management server 5 acquires the configuration information of the managed device 1 from the managed device 1 that is in operation, and manages the management target when a failure occurs in the resource that configures the managed device 1 from the acquired configuration information. The degree of influence indicating the influence on the business execution of the device 1 is calculated for each resource, and when there is a resource whose calculated degree of influence exceeds the upper limit value of the predetermined influence degree, the influence degree of the resource is set to the upper limit value. A configuration plan of the management target apparatus 1 below is created.

The configuration plan of the managed device 1 to be created is a configuration change plan of the distributed configuration and the multiplexed configuration of the managed device 1 whose resource influence level is lower than the upper limit value. The configuration management server 5 calculates, as a resource influence degree, the sum of the influence degree of other resources that the resource affects and the importance of the resource when a failure occurs in the resource that constitutes the managed device 1. . The configuration management server 5 corrects the calculated influence level with respect to the resource having the multiplexed configuration among the resources for which the influence degree is calculated. Further, the configuration management server 5 sets a predetermined upper limit value of influence input from the management client 6 and determines whether or not the calculated influence degree for each resource exceeds the predetermined upper limit value of the influence degree. When the calculated influence degree exceeds the predetermined upper limit value of the influence degree, an alert is output to the management client.

FIG. 2 shows each part and each table necessary for the description of the present embodiment of the configuration management system including the management target device 1, the configuration management server 5, and the management client 6. The configuration information providing unit 16 stored in the memory of the management target device 1 is executed by the CPU. Each unit stored in the memory of the configuration management server 5 is executed by the CPU.

The configuration information providing unit 16 of the management target device 1 transmits the configuration information of the management target device 1 in response to a request from the configuration management server 5. The configuration management server 5 functions by controlling the execution of each unit by the configuration information control unit 35. Each information table stored in the auxiliary storage device is used for execution of each unit. A description will be given below of each unit including the configuration information control unit 35 and each information table.

FIG. 3 shows a processing flow of the configuration control unit 35. An outline of the operation of the configuration management system will be described using this processing flow, and details of the operation of each part will be described later. The configuration control unit 35 executes each processing unit, and manages and controls the configuration of the management target device 1. The configuration information acquisition unit 17 collects configuration information of the management target device 1 (step 7). Based on the collected configuration information, the importance calculator 18 calculates the importance of the resource (step 8), and the influence calculator 19 calculates the influence of the resource. (Step 9). Next, the multiplicity calculation unit 20 calculates the multiplicity of the resource, and corrects the influence based on the calculated multiplicity (step 10). Thereafter, the alert policy detection unit 21 determines whether or not the obtained influence degree matches the alert policy (steps 11 and 12). When the alert policy does not match the alert policy, the process ends. The alert policy is set in the alert policy table 34 in advance by the input from the system administrator via the management client 6 accompanying the execution of the alert policy setting unit 24.

If it matches the alert policy, the alert output unit 23 issues an alert (step 13). Next, a configuration change plan recommended by the recommended configuration generation unit 22 is created (step 14), and the configuration change plan created by the recommended configuration display unit 25 is displayed on the management client 5 (step 15). Hereinafter, each processing of Step 7 to Step 14 of the configuration control unit 35 will be described.

<Collecting configuration information of managed device 1 (step 7)>
The configuration information acquisition unit 17 of the configuration management server 5 requests the configuration information providing unit 16 of the management target device 1 to transmit the configuration information, and the configuration information providing unit 16 of the management target device 1 obtains the configuration information of the management target device 1. Send back. The configuration information of the management target device 1 is associated with the resource name and identifier of the management target device 1 including virtualized ones, and the configuration is based on the system operation management such as the attribute and the relationship with other resources. Information and configuration information based on business operation management such as which resources are used by business programs (applications). From the point of view of system operation management, applications and the like can be handled as software resources, so in the following, they are called resources regardless of hardware and software.

The configuration information of the management target device 1 varies depending on the result of the business operation management of the information processing system (the management target device 1) in operation by the business manager.

The configuration information collection unit 17 stores the received configuration information in the configuration information table 26. In addition, the configuration information collection unit 17 stores related information between resources included in the configuration information in the resource related information table 27, and displays a resource list (resource identifier and resource type for each resource) of the management target device 1 as impact information. Store in table 28. The influence information table 28 will be described later.

The configuration information table 26 is composed of tables for each resource type as shown in FIGS. FIG. 4 shows an application configuration information table 100 as an example of the configuration information table 26 in which the resource type 101 is “application”, such as an identifier 102 for identifying a resource included in the collected configuration information, an application name 103, an installation path 104, and the like. Stores attribute values. FIG. 5 shows an example of the host configuration information table 105, which stores a host name, an IP address, and the like as attribute values. FIG. 6 shows an example of the processor configuration information table 106, which stores the operating frequency of the processor, the number of cores, and the like as attribute values. Although illustration is omitted, the resource type further includes resources, switches, and the like that constitute a host such as a processor and a memory.

FIG. 7 shows an example of the resource related information table 27, which stores related information between resources and the type of relationship. Associations between resources include an association source (resource identifier and resource type) 108 and an association destination (resource identifier and resource type) 109. For example, the relationship between “host-A” and “application-A” in which the association type 110 is “owned” indicates that “host-A” as the association source owns “application-A”. The relation type 110 “used” indicates that the relation source 108 uses the relation destination 109. In other words, “own” represents an inclusion relationship, and “use” represents a connection relationship.

The configuration information collection unit 17 displays the configuration confirmation screen 111 shown in FIG. 8 on the display of the management client 6 based on the related information stored in the resource related information table 27. The example of the configuration confirmation screen 111 shown in FIG. 8 indicates that, for example, “Host-A” owns “Application-A” corresponding to the resource related information table 27 shown in FIG.

<Calculation of resource importance (step 8)>
The resources included in the collected configuration information have attributes for each resource type. In the application configuration information table 100 of FIG. 4, the application name 103, the installation path 104, and the like are attributes. The importance level calculation unit 18 refers to the importance level 203 that matches the condition 202 regarding the resource type 201 and its attributes defined in the importance level definition table 31 shown in FIG. 9, and calculates the importance level for each resource. The importance of a resource that does not match the condition of the resource type 201 is set to 1, which is the minimum value. The obtained importance is stored in the importance column of the influence information table 28.

The definition contents of the resource type 201, the condition 202, and the importance 203 in the importance definition table 31 are input and updated by the system administrator through the management client 6 according to the configuration change of the management target device 1. Note that the value of the importance 203 that matches the condition 202 regarding the resource type 201 and its attributes is defined according to the importance of the resource from the viewpoint of business operation management, and the importance of the resource related to business operation management such as an application. The degree is obtained from the business manager in advance.

FIG. 10 is an example of the impact information table 28. In the influence level information table 28, the resource identifier 301 and the resource type 302 are already stored by the configuration information collection unit 17. Since the resource identifier “application-A” hits the importance level definition table 31 and the importance level 203 is 5, 5 is stored in the importance level 303 of the resource identifier 301 “application-A” in the influence level information table 28. . Similarly, importance is obtained and stored for all records (all resources of the management target device 1) in the influence information table 28.

<Calculation of resource influence (step 9)>
The degree of influence for each resource is obtained by the following formula.
Resource impact = (sum of impacts of all affected resources) + (resource importance)
The degree of influence indicates the degree of influence on the business execution of the managed device 1 (information processing system) when a failure occurs in a certain resource. As shown in the above equation, when a failure occurs in a certain resource, Define the sum of the impact level of the resource affected by the resource failure and the importance level of the resource.

The influence degree calculation unit 19 uses the dependency definition table 30 shown in FIG. 11 in order to obtain the target resource for which the influence degree of the above equation is obtained (the sum of the influence degrees of all the affecting resources). The dependency relationship definition table 30 stores an influence source resource type 401 that is an influence source, an influence destination resource type 402 that is affected, and an influence source relation type 403. The influence source relation type 403 indicates which of the relation source resource and the relation destination resource is stored in the resource relation information table 27.

The influence degree calculation unit 19 obtains the influence degree of each resource by the above formula and stores it in the influence degree 304 of the influence degree information table 28. The processing flow of the influence calculation unit 19 is shown in FIGS.

It is searched whether the resource type 302 corresponding to the resource identifier 301 of the influence degree information table 28 is in the influence source resource type 401 of the dependency relationship definition table 30 (step 501). The processing flow shows processing related to one of the resource identifiers 301 of the influence degree information table 28, but is repeated until the influence degree 304 is stored for all of the resource identifiers 301 of the influence degree information table 28.

As a result of the search, if there is a resource type, step 503 is repeated for each dependency, and if not, the process branches to step 505 (step 502). If there is no resource type, the importance degree 303 of the resource in the influence degree information table 28 is set as the influence degree 304 (step 505). For example, when “dependence definition table 30 is searched for“ host ”, which is resource type 302 having resource identifier 301 of“ host-A ”in impact information table 28, resource type“ host ”is the influence source of dependency definition table 30. Since the resource type 401 exists, step 503 is executed to calculate the degree of influence for each dependency relationship.

The influence degree calculation process for each dependency will be described with reference to FIG. It is determined whether or not the relation type 403 of the influence source resource type 401 in the dependency relation definition table 30 is the relation source (step 601). If the relation type is the relation source, the corresponding resource is the relation source, and the dependency relation definition table 30 The resource relation information table 27 is searched for the relation of the relation destination resource type equal to the influence destination resource type 402 (step 602). If it is not the association source, the relevant resource is the association destination, and the association source resource type equal to the influence source resource type 401 of the dependency relationship definition table 30 is searched from the resource association information table 27 (step 605). For example, when the influence source resource type 401 is “host”, since “related source” is defined as the related type in the dependency definition table 30, “host-A” is the related source, and the related destination is When a relationship that is an application of the affected resource type 402 in the dependency relationship definition table 30 is searched from the resource related information table 27, “application-A”, “application-B”, and “application-C” are found as related resources. . This is the resource that “Host-A” affects. This is stored in the impact information table 28 as the affected resource 305.

Next, for each resource found as a related resource, the degree of influence is obtained (step 603). For example, since “application”, which is the resource type of “application-A”, is not included in the influence source resource type 401 of the dependency definition table 30, there is no resource that is affected by “application-A”. As a result, the “application-A” (the sum of the influence levels of all the affected resources) is 0. Further, the importance level of “application-A” is obtained as 5 from the influence degree table 28. As a result, the influence degree of “application-A” is obtained as 0 + 5 = ， 5, and this is stored in the influence degree 304 of the row “application-A” of the influence degree information table 28. Similarly, the degree of influence of “application-B” and “application-C” is both 1.

Next, the sum of the influence level of all the dependencies of the corresponding resource and the importance level of the corresponding resource is set as the influence level of the corresponding resource (step 504). When the corresponding resource is “Host-A”, the “Sum of the impact of all affected resources” of “Host-A” is (the impact of “Application-A”) + (the impact of “Application-B”) ) + (Influence of “Application-C 度”) = 5 + 1 + 1 = 7. Also, since the importance level of “host-A” is 1 in the impact level information table 28, the impact level of “host-A” is calculated as 7 + 1 = 8. This is stored in the influence degree 304 of “host-A” in the influence degree information table 28.

Further, the influence degree calculation unit 19 displays an influence degree confirmation screen 112 shown in FIG. 14 representing the calculated influence degree on the display of the management client 6. The influence degree confirmation screen 112 is a screen in which the influence degree of each resource is superimposed on the configuration information confirmation screen 111 shown in FIG.

<Calculation of resource multiplicity (step 10)>
When resources of the managed device 1 are multiplexed as in a cluster configuration or a hardware redundant configuration, the influence on the managed device 1 when a failure occurs in the resource is an unmultiplexed resource Smaller than. Therefore, the resource multiplicity calculation unit 20 searches for multiplexed resources and corrects the influence.

FIG. 15 is an example of the multiplicity definition table 29 in which a resource type 701 to be multiplexed and a condition 702 for multiplexing are defined. The resource multiplicity calculation unit 20 searches the multiplicity definition table 29 for each collected resource to find a set of multiplexed resources. The discovered resource set is handled as one resource in the impact information table 28, and the impact is corrected according to the multiplicity. The influence degree of the duplicated resource is corrected to 1 / (multiplicity) so that the influence degree of the resource is 1/2, and the influence degree 304 of the influence degree information table 28 is updated.

<Alert policy judgment and issue (steps 11 to 13)>
The alert detection unit 21 refers to the alert policy 34 to determine whether to issue an alert. When issuing an alert, the alert output unit 23 issues an alert. The alert output unit 23 notifies the user of an alert by at least one of GUI display, mail transmission, SNMP (Simple Network Management Protocol), etc. on the display of the management client 6.

FIG. 16 is an example of the alert policy table 34, in which an excluded resource type is defined as an upper limit value (threshold value) 801 of the degree of influence, a threshold value 802 of the number of resources exceeding the upper limit value, and an exclusion condition 803. A resource whose impact level exceeds the upper limit value 801 and whose number (the number of resources) is not less than the threshold value 802 for the number of resources is not subject to the exclusion condition 803.

Fig. 17 shows the processing flow for alert determination and issue. When the degree of influence exceeds a threshold specified by the user, an alert is issued regarding the resource and notified to the system administrator. In the alert detection unit 21, the upper limit value of the influence degree of the alert policy table 34 is obtained from the influence degree information table 28 in which the influence degree 304 is obtained by the influence degree calculation part 19 (including correction by the multiplicity calculation part 20) for each resource. When the number of resources exceeding 801 is searched and the number of searched resources exceeds the threshold 802 for the number of resources exceeding the upper limit of the alert policy table 34 (step 900), the alert is sent to the user through the alert output unit 23. Notification is made (step 901). Instead of using the threshold value 802 for the number of resources, a simple determination may be made such that a resource exceeding the upper limit 801 of the influence degree is set as a resource to be alerted. In the example shown in FIG. 16, since the threshold value 802 for the number of resources is 1, this is substantially the same as the simple determination.

In the example shown in the impact information table 28 in FIG. 10 and the alert policy table 34 in FIG. 16, there are four resources that exceed the impact upper limit value 801 of 7 except for the network switch that satisfies the exclusion condition 803. Since 1 of the resource number threshold value 802 is also exceeded, an alert is issued.

<Recommended configuration change proposal creation (step 14)>
When the degree of influence exceeds a specified threshold, an alert is issued and a recommended configuration change proposal is provided to the user. The recommended configuration generation unit 22 creates a recommended configuration change plan, and the recommended configuration display unit 25 displays the created configuration change plan.

FIG. 18 shows a processing flow of the recommended configuration generation unit 22. The recommended configuration generation unit 22 executes the distributed configuration creation processing (step 1002) and the multiplexed configuration creation processing (step 1004), and changes the configuration of the distributed configuration and the multiplexed configuration as a new configuration plan of the management target device 1. Create a draft.

Hereinafter, the distributed configuration creation processing (step 1002) and the multiplexed configuration creation processing (step 1004) will be described.

<Distributed configuration change plan creation (step 1002)>
FIG. 19 shows a processing flow for creating a distributed configuration. In this process, the distributed recommended configuration is stored in the distributed configuration information table 32.

First, a list of affected resources 305 of the corresponding resource is acquired from the impact information table 28 (step 1105). Taking “host-A” as an example, “application-A”, “application-B”, and “application-C” are acquired from the impact information table 28. Here, the corresponding resource is a resource whose influence level exceeds a prescribed threshold value. When there are a plurality of resources, the processing shown in FIG. 19 is repeated for the plurality of resources.

Next, the resources that can be allocated are searched in the order of the resources having the greatest influence on the resource (step 1106). When the resources that affect “Host A” are arranged in order of impact, they are in the order of “Application-A”, “Application-B”, and “Application-C” (in the case of the same impact, they are in no particular order). In this case, first, a resource to which “application A” is allocated is searched.

FIG. 20 shows a process flow for searching for resources that can be allocated in step 1106. A list of resources is acquired (step 1109). The resource list is a list of resources of the management target device 1 and is stored and managed in the configuration management server 5. Note that the resource list includes resources that constitute the management target device 1 but are not used. When there is no host as an additional resource in the resource list, the only resource that can be allocated is “Host-A”, and no association is assigned to “Host-A” (for reassignment, (Assuming an initial state in which no association is assigned to “Host-A” without considering the association collected by the configuration change collection unit), “Application-A” is assigned to “Host-A”. As a result, the impact level of Host A becomes “Application A” impact level + “Host A” severity level = 5 + 1 = 6, which exceeds the alert threshold, but “Host A” is an unallocated resource. Therefore, “application A” is assigned and stored in the distributed configuration table 1100. FIG. 22 is an example of the distributed configuration information table 32, and shows that the resource identifier 1103 assigns “application-A” to the resource with the resource identifier 1101 “host A” (step 1108).

Next, search for resources to which “Application-B” is assigned. The resource to be assigned is “Host-A”, but when assigned to “Host-A”, the impact of “Host-A” is the impact of “Application-A” + The impact of “Application-B” + The importance of “Host-A” is = 5 + 1 + 1 = 7, which exceeds the upper limit of the impact level of the alert policy.

Since “Host-A” is not an unassigned resource (“Application-A” has already been assigned), the assigned resource is not found (step 1110). If a resource to be allocated cannot be found, a new resource is added (step 1107). In the example of FIG. 22, a new resource “additional host” is added and “application-B” is assigned to “additional host”. Similarly, “Application-C” is assigned to “Additional Host”. FIG. 22 shows a decentralized table (one of the forms for realizing decentralized change plan information) created for “Host-A” as described above. Also, the degree of influence 1104 of the created recommended configuration is stored in the distributed table 1100.

The above example is a case where “additional host” is in the resource list as a new resource, that is, there is a resource (host) that is a resource (host) that constitutes the managed device 1 but is not used. There may be no new resources. In this case, it is necessary to adopt a decentralized configuration, but there is no additional resource, and the system administrator is notified via the management client 6 that the decentralized configuration cannot be taken. Create a configuration proposal and notify the system administrator of a message prompting you to add resources.

<Multiplexing configuration change proposal creation (step 1004)>
FIG. 21 shows a processing flow for creating a multiplexed configuration. In this processing, the recommended multiplexing configuration is stored in the multiplexing configuration information table 33.

First, the multiplicity is calculated from the current influence level and threshold value by the following formula (step 1204). Multiplicity = Resource impact / Resource threshold 但 (however, rounded up after the decimal point) Hereafter, “Host-A” will be taken up as a resource, but if there are multiple resources whose impact exceeds the specified threshold , The processing described below is repeated for the plurality of resources.

Since the influence degree of “Host-A” is 8 from the influence degree table 28 of FIG. 10 and the influence upper limit value is 7 from the alert policy table 34 of FIG. 16, the multiplicity is obtained as [8/7] = 2. Step 1204). [] Represents an integer value rounded up after the decimal point. Thus, one “additional host” is added to the current “host-A” (step 1205), and the number of hosts is set to two. Next, an association is created between “additional host” and resources affecting “host-A” (“application-A”, “application-B”, and “application C”) (step 1206). . Furthermore, since the degree of influence of “host-A” and “additional host” is multiplexed at multiplicity 2, the degree of influence is 1 / (multiplicity), and 8/2 = 4 respectively. The created multiplexing configuration resource identifier 1201 and resource type 1202 and the calculated recommended configuration impact 1203 are stored in the multiplexing table 1200 (one of the implementation forms of multiplexing change plan information). FIG. 23 shows an example of a multiplexing configuration table.

As for “additional host”, as in the case of creating a distributed configuration change proposal, if there is no resource as an additional host, it is necessary to adopt a multiplexing configuration, but there is no additional resource and a multiplexing configuration is adopted. The system administrator is notified via the management client 6 that the system cannot perform the operation, or a multiplexing configuration plan is created on the assumption that there are additional resources, and a message prompting the addition of resources is notified to the system administrator.

<Display recommended configuration change proposal (step 15)>
The recommended configuration display unit 25 displays a recommended configuration change plan on the management client 6. FIG. 24 shows an example of a recommended configuration display screen 1300. Further, since the information necessary for the recommended configuration display screen is stored in the distributed configuration information table 32 and the multiplexed configuration information table 33, the recommended configuration display screen 1300 can be displayed even after the configuration is changed.

Therefore, the system administrator can change the configuration to the same by displaying the impact after the configuration change on the impact confirmation screen after making the configuration change according to the recommended configuration and comparing it with the recommended configuration screen 1300. Can be confirmed.

According to the present embodiment, it is possible to perform system operation management in which a failure is unlikely to occur in the information processing system, and particularly, a failure that has a large influence on business operations is unlikely to occur. Specifically, it is possible to detect in advance resources that may have a serious impact on the information system in the event of a failure, and propose a configuration change proposal corresponding to the detected content to the system administrator. Become. In addition, it is possible to verify whether the configuration can be changed according to the configuration change plan after the actual configuration change.

1: managed device, 2: server device, 4: switch, 5: configuration management server, 6: management client, 16: configuration information providing unit, 17: configuration information acquisition unit, 18: importance calculation unit, 19: influence Degree calculation unit, 20: multiplicity calculation unit, 21: alert detection unit, 22: recommended configuration generation unit, 23: alert output unit, 24: alert policy setting unit, 25: recommended configuration display unit, 26: configuration information table, 27: Resource related information table, 28: Impact information table, 29: Multiplicity definition table, 30: Dependency definition table, 31: Importance definition table, 32: Distributed configuration information table, 33: Multiplex configuration information table , 34: Alert policy, 35: Configuration information control unit.

Claims

An operation management device connected to an information processing system whose configuration is to be managed,
A configuration information acquisition unit that connects to a management client and acquires configuration information of the information processing system from the information processing system in operation;
An influence calculation unit for calculating an influence degree indicating an influence on business execution of the information processing system when a failure occurs in the resource constituting the information processing system from the obtained configuration information; and
A configuration generation unit configured to create a configuration plan of the information processing system in which the influence degree of the resource is lower than the upper limit value when the calculated influence degree exceeds a predetermined upper limit value of the influence degree; An operation management apparatus characterized by that.
2. The operation management according to claim 1, wherein the configuration plan of the information processing system is a configuration change plan of a distributed configuration and a multiplexed configuration of the information processing system in which the influence degree of the resource is lower than the upper limit value. apparatus.
The influence calculation unit, when a failure occurs in the resource constituting the information processing system, uses the sum of the importance of the resource and the other resource influence that the resource affects as the influence degree of the resource The operation management apparatus according to claim 2, wherein the operation management apparatus is calculated.
The multiplicity calculation unit for correcting the calculated influence degree with respect to a resource configured to be multiplexed among the resources for which the influence degree calculation unit has calculated the influence degree. 2. The operation management apparatus according to 2.
An alert policy setting unit for setting the upper limit value of the predetermined influence level input from the management client, whether the influence level calculated for each resource exceeds the upper limit value of the predetermined influence level An alert detection unit for determining, and an alert output unit for outputting an alert to the management client when the calculated degree of influence exceeds an upper limit value of the predetermined degree of influence. 2. The operation management apparatus according to 2.
An operation management method by an operation management apparatus connected to an information processing system whose configuration is to be managed, wherein the operation management apparatus is connected to a management client, and the operation management apparatus is connected to the information processing system in operation, The configuration information of the information processing system is acquired, and the degree of influence indicating the influence on the performance of the information processing system when a failure occurs in a resource configuring the information processing system from each of the acquired configuration information. Calculate resources,
When there is a resource for which the calculated influence degree exceeds a predetermined upper limit value of the influence degree, a configuration plan of the information processing system in which the influence degree of the resource is lower than the upper limit value is created. Operation management method.
7. The operation management according to claim 6, wherein the configuration plan of the information processing system is a configuration change plan of a distributed configuration and a multiplexed configuration of the information processing system in which the influence degree of the resource is lower than the upper limit value. Method.
The operation management device calculates, as a resource influence degree, the sum of importance of the resource and another resource influence degree that the resource influences when a failure occurs in the resource constituting the information processing system The operation management method according to claim 7, wherein:
The operation management method according to claim 7, wherein the operation management apparatus corrects the calculated influence degree with respect to a resource configured to be multiplexed among the resources for which the influence degree is calculated.
The operation management apparatus sets an upper limit value of the predetermined influence degree input from the management client, and whether the influence degree calculated for each resource exceeds the upper limit value of the predetermined influence degree. The operation management method according to claim 7, wherein an alert is output to the management client when the calculated degree of influence exceeds an upper limit value of the predetermined degree of influence.