JP6234759B2 - Information system - Google Patents

Description

  The present invention relates to an information system for checking the soundness of a system configuration in order to prevent a failure from occurring in an environment in which a virtualization infrastructure is constructed on a plurality of physical servers and the system is used by a plurality of users.

  In recent years, virtualization has been attracting attention as a trump card for improving server operation efficiency, such as the reduction in the number of servers and the integration of servers with low operating rates. The development of virtualization technology for computer systems and management of IT systems in companies The philosophy requires not only system ownership but also system usage.

  With such social changes, the use of virtual hosting services and public cloud services that use computer resources as necessary is increasing. The public cloud service is a cloud service provided to general users among services operated by cloud computing. This public cloud is widely provided for a large number of unspecified users such as various companies, organizations, and individuals.

  On the other hand, a form in which a company constructs a cloud computing system in-house and provides cloud services to departments and group companies in the company is called a private cloud. In addition, a dedicated virtualization infrastructure server is prepared in-house and operated in combination with a conventional physical device such as a user terminal, a management server, or a storage device.

In an environment using the above-described virtualization technology, a CPU (Central
It is possible to flexibly change resources such as (Processing Unit), memory, and disk, and to increase or decrease the number of servers according to the load. Companies use these technologies to efficiently use resources according to server load conditions to reduce costs for IT systems.

  The above-mentioned IT system is required to operate normally for 24 hours on 365 days. There is soundness in the scale for judging normal operation. As a technique for preventing a failure from occurring according to the degree of soundness (health degree), there is one described in Patent Document 1. This technology relates to an apparatus that estimates the usage state of each node in a distributed server system that performs processing by arranging a plurality of nodes on a network.

  Specifically, in a distributed server system having a plurality of nodes, a management node that manages the driving state of each node is provided, and the management node periodically receives the driving state of each node as a plurality of parameters. A node health level determination unit that calculates a health level based on a predetermined formula from each node parameter, a health level history management unit that records the calculated health level as history information, and a calculated current health level and health level A state estimation device is configured by including a node state determination unit that determines a node state based on a history.

Japanese Published Patent No. 2012-8935

  In the prior art described in Patent Document 1, the soundness of physical resources (load, fan speed, power supply, HDD capacity, etc.) in physical servers constituting a distributed file system is mentioned. There is no mention of soundness judgment and risks that may occur due to the addition of virtual servers and the addition of virtualization functions. For this reason, problems in the virtualization infrastructure in the private cloud and public cloud described in the background art cannot be detected in advance to prevent failure.

  In order to solve the above-described problems, in the information system of the present invention, the information system includes a plurality of physical servers, a management server, and a plurality of user terminals, and operates a virtual machine that operates the physical server on the physical server and operates a virtual machine. The management server is configured to build a virtualization infrastructure on the virtual server, the management server controls the entire apparatus, a memory unit that temporarily stores collected data and analysis results, and a storage device unit that stores collected data and analysis results And the health item master data for determining the health of the virtualization infrastructure in the memory unit or the storage device unit, the pattern master data for classifying the health evaluation results of multiple times into a specific pattern, and the configuration data in the specific pattern The pattern data that increases or decreases the collection frequency of the virtual server is stored, and the CPU of the management server collects the configuration data from each virtual server and collects the collected configuration data. Calculate the scale of the virtualization platform from the data, extract the items for judging the soundness based on the calculated scale and the evaluation criteria from the soundness master data, generate the reference data, and generate the reference data and the configuration data of the virtual server. Compare to determine soundness.

  In the information system of the present invention, an item to be checked for soundness is extracted from the scale of the system that constitutes the virtualization infrastructure, a predetermined number of checks are performed on the item, and a risk that is considered to occur from the tendency of the result is identified it can. Therefore, the occurrence of a failure can be prevented in advance, so that the reliability and availability of the virtualization infrastructure can be improved. Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

FIG. 1 is a diagram showing an outline of the present invention. FIG. 2 is a block diagram showing the overall configuration of the information system to which the present invention is applied. FIG. 3 is a block diagram showing the internal configuration of the physical server. FIG. 4 is a block diagram showing an internal configuration of the user terminal. FIG. 5 is a diagram illustrating a configuration example of configuration data. FIG. 6 is a diagram illustrating a configuration example of soundness item master data. FIG. 7 is a diagram illustrating a configuration example of the reference data. FIG. 8 is a diagram illustrating a configuration example of the first analysis result data (analysis result 1). FIG. 9 is a diagram illustrating a configuration example of the health trend data. FIG. 10 is a diagram illustrating a configuration example of the second analysis result data (analysis result 2). FIG. 11 is a diagram illustrating a configuration example of pattern master data for classifying the tendency of the configuration check result. FIG. 12 is a diagram illustrating a configuration example of pattern data for determining increase / decrease in configuration data collection frequency with the adapted pattern data. FIG. 13 is a flowchart showing the overall processing of the virtualization infrastructure soundness check. FIG. 14 is a flowchart showing the configuration data collection process. FIG. 15 is a flowchart showing configuration data analysis processing. FIG. 16 is a flowchart showing the configuration data comparison process. FIG. 17 is a flowchart showing soundness determination processing. FIG. 18 is a flowchart showing soundness item / check frequency update processing.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, various types of information may be described using an expression such as “management table”, but the various types of information may be expressed using a data structure other than a table. Further, the “management table” can be referred to as “management information” to indicate that it does not depend on the data structure.

  Further, the process may be described with “program” as the subject. The program is executed by a processor, for example, an MP (Micro Processor) or a CPU (Central Processing Unit), and performs a predetermined process. Note that the processing subject may be a processor because the storage resource (for example, a memory) and a communication interface device (for example, a communication port) are used as appropriate. The processor may have dedicated hardware in addition to the CPU. The computer program may be installed on each computer from a program source. The program source may be provided by, for example, a program distribution server or a storage medium.

  Each element, for example, the controller can be identified by a number or the like, but other types of identification information such as a name may be used as long as the information can be identified. In the drawings and description of the present invention, the same reference numerals are given to the same parts, but the present invention is not limited to the present embodiment, and any application examples that meet the idea of the present invention are technical. Included in the range. Further, unless specifically limited, each component may be plural or singular.

<Invention Summary>
FIG. 1 is a diagram showing an outline of the present invention. In the present invention, the management server 16 collects the configuration data of the virtual server 12 and performs analysis / comparison processing on the collected configuration data in order to quickly grasp the soundness of the virtualization infrastructure and reduce the risk of failure occurrence. The soundness is judged. Furthermore, the collection frequency of configuration data is updated with the determined soundness to quickly grasp the deterioration of soundness. Therefore, the management server 16 executes the following processes (a1) to (a5).

(A1) Configuration data collection (S101): Collect configuration data of each virtual server and store difference data in the configuration data DB 21.
(A2) Configuration data analysis (S102): Calculate the scale amount (the scale of the virtual server (for example, the number, the number of virtual CPUs / frequency, virtual memory / storage capacity) constituting the virtualization infrastructure) from the configuration data Extract and obtain soundness check items.
(A3) Configuration data comparison (S103): A configuration check process is performed based on the check items (a soundness level (evaluation value) is calculated based on an allowable use period, a priority level, and an importance level).
(A4) Soundness determination (S104): Soundness determination is performed based on the check items. The tendency of the soundness level is grasped by the soundness level (evaluation value) calculated in the past and the newly calculated soundness level (evaluation value). The system manager and maintenance center are notified of the risks that occur.
(A5) Soundness item & check frequency update (S105): Update (addition, correction, deletion) of the soundness check item and the importance level based on the analysis result in the configuration check process. In addition, the collection frequency has been updated from the above-mentioned tendency.

  Details of the outline will be described below. The management subject is the management server 16, but it may be a CPU (not shown) of the management server 16 or a program operating on the CPU, for example, a program for monitoring and managing a virtual environment such as a virtual environment monitoring program.

  The management server 16 is a group of virtual servers, and a new virtual server (for example, virtual server 11 12a) is added or a function that operates on the virtual server 12 (for example, “Auto Deployment function” in FIG. 1 is added). In order to determine whether or not the virtual server configuration and the functional configuration on the virtual server are periodically collected (the configuration information and the functional information are collectively referred to as configuration data). The configuration data 50 is stored in the configuration data DB 21. In this storage, all collected configuration data 50 may be stored together with the collection date and time, but in this embodiment, for efficient use of the storage area of the management server 16. Generation management (generation number and collection date / time) is performed only on data that is different from the previously collected configuration data. If it is “fourth generation”, the difference data of the configuration data collected this time is “fifth generation”.

  Next, in step S102, the current configuration data is constructed from the difference data up to the fifth generation in which the configuration data collected in step S101 is stored directly or in the configuration data DB 21, and the configuration data is acquired. From the acquired configuration data, the scale amount is calculated from the scale of the virtual server constituting the virtualization platform, for example, the number of operating virtual servers, the number of virtual CPUs / frequency, the virtual memory / storage capacity, and the like. For example, if the number of virtual servers operating on the virtualization infrastructure is 10 or less, it is classified as “small”, 11 or more and 30 or less are classified as “medium”, and 31 or more are classified as “large”.

  In FIG. 1, since the number of virtual servers is 10 (from virtual server 1 to virtual server 10) at first, the management server 16 determines and identifies “small”. With the addition of the virtual server 1112a, the scale amount of the virtualization infrastructure changes from “small” to “medium”, and the management server 16 identifies “medium”. The management server 16 uses this scale amount (“medium scale”) to extract and obtain the health check item (also referred to as reference data) from the health item master data DB 22. The soundness item master data DB 22 holds the evaluation items and evaluation criteria of the configuration data that serve as a reference for evaluating the soundness described later.

  Next, in process S103, the management server 16 acquires the soundness check item extracted in process S102. The management server 16 performs a configuration check process for comparing the acquired soundness check item and the configuration data 50 acquired in S101 on each virtual server. Then, the management server 16 stores the comparison results (or referred to as analysis results, which will be referred to as analysis results hereinafter) in each virtual server performed in S103 in the analysis result data DB 23. The analysis result data DB 23 holds past analysis results (evaluation items, evaluation results for each item, evaluation values obtained by quantifying the evaluation results, generation information, etc.) of each virtual server.

  Next, in process S104, the management server 16 acquires the analysis result from the analysis result data DB 23 or directly from the process S103, and determines the soundness of each virtual server or the entire virtualization infrastructure. Specifically, the management server 16 obtains the degree of soundness (health degree) from the evaluation result and the evaluation value in the analysis result, and generates soundness trend data (health tendency). Then, possible risk contents (performance deterioration, data backup impossibility, increase in storage capacity, etc.) are specified by the health trend data and the evaluation result and evaluation value for each evaluation item. A countermeasure for the identified risk is acquired from the analysis result data DB 23. And it reports to a system administrator with the above-mentioned soundness degree and its tendency.

  In this way, the management server 16 can grasp the soundness of each virtual server at the present time. Also, comprehensively analyze the analysis results of each virtual server (evaluation of soundness by the cumulative value of evaluation values of each virtual server, identification of virtual servers that require improvement through statistical analysis, and prioritization of improvements) As a result, it is possible to grasp the soundness of the entire virtualization infrastructure.

  Furthermore, the management server 16 obtains a health trend from the evaluation value for each generation in each virtual server, and how the health of the virtual server itself and the entire virtualization infrastructure changes. You can see how it will go. In addition, the management server 16 generates health trend data (health tendency) with the evaluation value for each evaluation item of each virtual server, and which evaluation item of which virtual server affects the health of the virtualization infrastructure. Can be determined.

  For example, if the evaluation value tends to decrease, it can be judged that the soundness of the virtualization infrastructure including the virtual server has deteriorated. If the evaluation value tends to increase, the soundness that has deteriorated has been improved. Can be judged. In addition, if the high evaluation value continues, it can be determined that the virtualization infrastructure maintains high soundness, and if the evaluation value vibrates as high or low, the entire virtualization infrastructure is determined to be unstable. It is possible to identify the virtual server that is the cause by the tendency of the evaluation value for each virtual server.

  Further, the management server 16 can report soundness, possible risk contents, and countermeasures to the system administrator and the maintenance center 6. The soundness determination results described above are classified for each generation of configuration data collection, and are stored in the pattern master data DB 24 by the management server 16.

  Finally, in step S105, the management server 16 updates evaluation items, importance levels, configuration data collection frequency, and the like based on the analysis results. For example, the previous (fourth generation) configuration data does not have an Auto Deploy function, and if it is newly added at this time (fifth generation), it is necessary to add an evaluation item and make a soundness judgment after the next time. . Therefore, the management server 16 updates the contents of the soundness item master data DB 22. Similarly, when the function is deleted or a virtual server is added, the management server 16 updates the content of the soundness item master data DB 22.

  Further, the number of collections of the configuration data (collection frequency) per predetermined period (for example, one day) is changed based on the analysis results of the past several generations. That is, the collection frequency is changed by increasing or decreasing the number of collections according to the evaluation result in the analysis result, the number of continuous OK or NG, or the number of OKs in a predetermined period. For example, at the current collection frequency of once / day, if the evaluation result is NG three or more consecutive times, increase the number of collections per day by 1 to 2 times / day, and judge the soundness in a short period of time. Reduce the risk of occurrence and identify risk factors at an early stage. Therefore, the analysis result is analyzed, and the management server 16 updates the evaluation item, the importance, the collection frequency of the configuration data, and the like, and stores the update result in the soundness item master data DB 22.

  As described above, in order to quickly grasp the soundness of the virtualization infrastructure and reduce the risk of failure occurrence, the management server 16 collects the configuration data of the virtual server 12, and analyzes and compares the collected configuration data. Run to determine health. In addition, it is possible to quickly grasp the deterioration of the soundness by updating the collection frequency of the configuration data with the determined soundness. Furthermore, it is possible to present a countermeasure for a risk that may occur to the system administrator. This enables stable operation and efficient management of the entire virtualization infrastructure.

<Overall configuration>
FIG. 2 is a block diagram showing the overall configuration of the information system to which the present invention is applied. The information system includes a maintenance center 6 connected to the data centers 1, 3, 4, 5 via a network 7.

  The data center 1 includes two or more physical servers 11, two or more virtual servers 12, a VM group 13 having a plurality of virtual machines (hereinafter referred to as VMs) 14, a shared storage 15, a management server 16, and a user terminal 17. The devices are connected to each other via a network 18a and a network 18b.

  The virtual server 12 may operate on one physical server 11, may operate by sharing two physical servers 11, or may share one physical server 11 with two virtual servers 12. May operate as well. On the virtual server 12, a plurality of VMs 13 having a guest OS (Operating System) and an AP (Application Program) operating on the guest OS operate. The user terminal 17 uses this VM 13 to receive predetermined work execution and service provision.

  Various virtual functions are installed in the virtual server 12, and for example, as shown by the virtual / physical server 12b in FIG. 1, the performance of the guest OS needs to be optimized and the function that requires cooperation between the guest OS and the virtual server is realized. Virtual tool management program that realizes efficiency of management and management, virtual data center management that provides functions such as automation of operation and management, optimization of physical resources, high availability, and centralized management of VMs and virtual servers, etc. There are various programs.

  The shared storage 15 is connected to the physical server 11 via the network 18a, and is shared and used by the physical server 11 to the virtual server 12 and the VM 14. The user uses the virtual infrastructure of the virtual server 12 and the VM group 13 constructed on the physical resources of the physical server 11 and the shared storage 15 by using the virtual data center management described above from the user terminal 17.

  The management server 16 is a server that monitors the operating state of the physical server 11 and the shared storage 15 that are the physical resources described above, and the usage status and soundness of the virtualization infrastructure. The management server may be provided in each data center, or a management server 61 that comprehensively manages the management server 16 of each data center as shown in FIG. 1 is provided in the maintenance center 6 to monitor the operation status of each data center. -It is possible to manage and resolve the failure remotely when a failure occurs.

<Internal configuration>
FIG. 3 is a block diagram showing the internal configuration of the physical server. The physical server 11 includes a CPU 111, a memory 112, a storage unit 113, an input unit 114, an output unit 115, a display unit 116, and a communication unit 117. The management server 16 and the management server 61 of the maintenance center 6 also have the same internal configuration as the physical server 11.

  The CPU 111 is a processor that controls the entire physical server 11. The memory 112 is a device that temporarily stores various programs and various data described below. The storage unit 113 is a device that permanently stores various programs and various data described below. The input unit 114 is a device that accepts input, and is, for example, a keyboard. The output unit 115 is a device that outputs information and data, and is, for example, a speaker or a printer. The display unit 116 is a device that displays information and data, and is, for example, a liquid crystal display. The communication unit 117 is a device that connects the physical server 11 and another device via a network. The above components are connected to each other via an internal bus.

  The shared storage 15 has the same configuration as that of the physical server 11, but the input unit 114, the output unit 115, and the display unit 116 indicated by dotted lines are not necessarily provided, and only an interface that can connect these components is provided. Just be fine.

  FIG. 4 is a block diagram showing an internal configuration of the user terminal. The user terminal 17 includes a CPU 171, a memory 172, a storage unit 173, an input unit 174, an output unit 175, a display unit 176, and a communication unit 177.

  The CPU 171 is a processor that controls the entire user terminal 17. The memory 172 is a device that temporarily stores various programs and various data described below. The storage unit 173 is a device that permanently stores various programs and various data described below. The input unit 174 is a device that receives input, and is, for example, a keyboard. The output unit 175 is a device that outputs information and data, and is, for example, a speaker or a printer. The display unit 176 is a device that displays information and data, and is, for example, a liquid crystal display. The communication unit 177 is a device that connects the user terminal 17 to the management server 16 and other devices via a network. The above components are connected to each other via an internal bus.

  The memory 172 stores various OSs of the OS provider and various application programs (APs) operating on the OSs. The various OSs described above can operate on a virtual server as a guest OS on a virtualization platform.

<Configuration data>
FIG. 5 is a diagram illustrating a configuration example of configuration data. The management server 16 collects configuration data from each virtual server, and calculates the scale amount of the virtualization platform from the collected configuration data group. For example, if the number of virtual servers operating on the virtualization infrastructure is 10 or less, it is classified as “small”, 11 or more and 30 or less are classified as “medium”, and 31 or more are classified as “large”. The configuration data 50 includes a host name 501, an evaluation item 502, a status 503 of the evaluation item 502, a generation 504, and a configuration data acquisition date 505.

  The host name 501 is information for uniquely identifying a virtual server. The evaluation item 502 is an item for evaluating the soundness, and is an item defined for each virtual server or for the entire virtualization infrastructure. Addition / deletion of item entries according to changes in the function of the virtual server or the configuration of the virtualization infrastructure, The item content is changed.

  A state 503 indicates a state for the evaluation item 502 at the time of collection. The generation 504 manages the number of times the configuration data 50 is acquired as a generation. Note that the difference between the configuration data of the previous generation and the configuration data of the current generation is stored in the configuration data DB 21 as the difference data of the acquired generation. If the configuration data DB 21 or the shared storage 15 has a sufficient storage area, all the configuration data for each generation may be stored instead of the difference data. The acquisition date / time 505 of the configuration data is information for uniquely identifying the date / time when the configuration data was collected and acquired.

  In the example of FIG. 5, the virtual server whose host name 501 is “virtual server 01”, the evaluation item 502 is the item status from “virtualization tool” to “distributed virtual switch”, the acquisition generation 504 and the acquisition date and time 505. Acquired by the management server 16. Then, the management server 16 detects the difference by comparing with the configuration data of the previous generation, and stores the difference data in the configuration data DB 21.

<Health property master data>
FIG. 6 is a diagram illustrating a configuration example of soundness item master data. The management server 16 extracts soundness check items from the calculated scale, and generates reference data together with evaluation criteria (recommended values). The soundness item master data 60 includes an evaluation item 601, an evaluation criterion (recommended value) 602, an allowable use period (days) 603, a priority 604, an importance 605, a scale 606, a possible risk content (hereinafter, risk content) 607. Consists of.

  An evaluation item 601 indicates an item for evaluating soundness. One evaluation item 601 is provided for each virtual server 12 or for the entire virtualization platform. The evaluation standard (recommended value) 602 is a standard indicating the ideal state of the virtual server 12 or the entire virtualization infrastructure. This evaluation criterion is compared with the acquired configuration data to determine a measure of soundness. The allowable use period (days) 603 indicates a period allowed even if the evaluation standard 602 is not satisfied. If the use of the function indicated in the evaluation item or the maintenance of the configuration is continued beyond the allowable use period (days) 603, a risk described later occurs.

  Priority 604 is an item indicating the order in which items are prioritized and improved in unsatisfactory evaluation items, and is classified into three categories of “high”, “medium”, and “small”. The score is given in five stages (1 to 5 points), and the score (evaluation score) is shown in parentheses after classification. For example, when the evaluation item 601 is “virtualization tool”, the priority 604 is “high (5)”, which means that the classification is “high” and the score is “5 points”. Similarly, in the “arrangement”, since the priority 604 is “medium (3)”, the classification is “medium” and the score is “3 points”.

  The importance 605 indicates the degree of importance in the soundness item, and is classified into three categories of “high”, “medium”, and “small”, and there are 10 levels (1 to 10 points) for each category. A score is assigned (evaluation score), and the score is shown in parentheses after classification. For example, when the evaluation item 601 is “virtualization tool”, the priority 604 is “high (10)”, which means that the classification is “high” and the score is “10 points”. Similarly, in the “placement”, since the priority 604 is “medium (5)”, it means that the classification is “medium” and the score is “5 points”.

  The scale 606 is set as a condition for selecting the evaluation item 601 with the calculated scale amount, and “common” means that the evaluation item 601 is selected regardless of the scale amount, and “medium scale”. Indicates the scale of 11 to 30 virtual servers 12, and “large scale” indicates the scale of 31 or more virtual servers 12.

  The risk content 607 indicates a risk content that is considered to occur when each evaluation item is not sound (e.g., the evaluation standard 602 is not satisfied, or the allowable use period (days) 603 is exceeded). For example, when the evaluation item 601 is “virtualization tool”, the risk content may cause performance degradation (degradation of processing and response performance). In “SnapShot”, there is a possibility that back-up is not possible or there is a risk of an increase in disk usage capacity.

<Reference data>
FIG. 7 is a diagram illustrating a configuration example of the reference data. The reference data is data for checking the health of each virtual server and the entire virtualization infrastructure in comparison with the collected configuration data. The reference data 70 includes a host name 701, an evaluation item 702, an evaluation reference 703, a collection frequency upper limit (times / day) 704, and a collection frequency (times / day) 705 for a predetermined period (per day). Since the host name 701 is the same as the host name 501 of the configuration data 50, and the evaluation item 702 and the evaluation criterion 703 are the same as the evaluation item 601 and the evaluation criterion 602 of the configuration data of the soundness item master data 60, description thereof will be omitted.

  The collection frequency upper limit (times / day) 704 is an upper limit of the number of times configuration data is collected per day, and by setting this upper limit, performance degradation of the entire virtualization platform due to the configuration data collection process is prevented. The collection frequency (times / day) 705 is a number that is increased or decreased according to the classification result of the evaluation results. Details of the increase and decrease will be described later.

<Analysis results>
FIG. 8 is a diagram illustrating a configuration example of the first analysis result data (analysis result 1). The analysis result 180 is a summary of the results of comparison between the configuration data 50 and the reference data 70 and analysis. The analysis result 1 80 includes a host name 801, an evaluation item 802, a status 803, a setting period (day) 804, an evaluation result 805, an evaluation value 806, an analysis number (generation) 807, and an analysis date and time 808. Since the host name 801 / evaluation item 802 / status 803 is the same as the host name 501 / evaluation item 502 / status 503 of the configuration data 50, a description thereof will be omitted.

  A setting period (day) 804 represents a period in which the function and configuration of the setting item 802 are set and used. For example, if the evaluation item 802 is “SnapShop”, the state 803 is “present” and the set period (day) 804 is “16 days”. Incidentally, when the evaluation item 702 in the reference data 70 is “SnapShop”, the evaluation criterion 703 is “No”, and even when “Yes”, the allowable use period (days) 603 is a maximum of 14 days.

  The evaluation result 805 and the result of evaluating the soundness of each evaluation item 802 are stored. For example, if the evaluation item 802 is “virtualization tool”, the status 803 is “latest” and the setting period (day) 804 is “−”. Incidentally, since the evaluation item (recommended criterion) 703 of the “virtualization tool” as the evaluation item 702 in the reference data 70 is “latest”, the evaluation result 805 of the analysis result 180 is “◯” (conformity / healthy). It becomes.

  If the evaluation item 802 is “SnapShop”, the state 803 is “present” and the set period (day) 804 is “16 days”. Incidentally, when the evaluation item 702 in the reference data 70 is “SnapShop”, the evaluation criterion 703 is “No”, and even when “Yes”, the allowable use period (days) 603 is a maximum of 14 days. Since the state 803 is also non-conforming and the set period (days) 804 has also been exceeded, the evaluation result 805 of the analysis result 180 is “x” (non-conforming / unhealthy).

  On the other hand, if the evaluation item 802 is “DVD Mount”, the state 803 is “present” and the set period (day) 804 is “7 days”. Incidentally, even when the evaluation item 702 in the reference data 70 is “DVD Mount”, the evaluation criterion 703 is “None”, and even if it is “Yes”, the allowable use period (days) is 14 days at the maximum. Therefore, even if the state 803 is nonconforming, the set period (days) 804 is within the allowable use period, and therefore the evaluation result 805 of the analysis result 180 is “◯” (conforming / healthy). In addition, the management server 16 exceeds the usage status message if the set period (day) 804 is a period less than half of the allowable use period (day) 603, and exceeds the alert message if it is more than half and less than the upper limit. If there is a problem, an improvement instruction or an emergency warning message is displayed on the display unit 116 and reported to the system administrator to indicate the risk of failure.

  The evaluation value 806 is obtained by multiplying the evaluation score at the priority level 604 and the evaluation score at the importance level 605 in each evaluation item 802, and the calculation result is plus (+) when the evaluation result 805 is “◯”. In the case of “x”, a minus (−) is assumed. For example, if the evaluation item 802 is “virtualization tool”, the priority 604 is “high (5)”, the importance 605 is “high (10)”, and the evaluation result 805 is “◯”, so the final calculation result is +50. . If the evaluation item 802 is “SnapShop”, 4 * 9 * (− 1) = − 36, and if the evaluation item 802 is “DVD Mount”, 4 * 8 * (+ 1) = + 32. .

  The cumulative value obtained by adding the evaluation values in each evaluation item is the evaluation value in the virtual server whose host name 801 is “virtual server 01”. Further, a value obtained by adding and accumulating evaluation values in each virtual server may be used as an evaluation value for the entire virtualization infrastructure.

  An analysis number (generation) 807 is a number for managing the result of comparison between the configuration data 50 and the reference data 70 and is the same as the generation number 504 of the configuration data. The analysis date and time 808 manages the date and time when the analysis is performed.

  The above analysis results are calculated by the management server 16 every time configuration data is collected and stored in the analysis result data DB 23.

  FIG. 9 is a diagram illustrating a configuration example of the health trend data. The health trend data 90 indicates the health trend of each virtual server or the health trend of the entire virtualization infrastructure. With this health trend data, the management server 16 indicates the health of the virtual server and the entire virtualization infrastructure. Understand how it is and how it will change. Moreover, the management server 16 obtains the health trend (soundness tendency) generated with the evaluation value for each evaluation item of each virtual server, and which evaluation item of which virtual server has an effect on the health. Can be judged.

  The health trend data 90 includes an analysis number (generation) 901 and an evaluation value 902 for each generation. An analysis number (generation) 901 represents the order in which the analysis is executed, and is generation-managed. The evaluation value 902 for each generation is a value obtained by adding the evaluation values for each evaluation item and accumulating in one virtual server described with reference to FIG. 8, and the accumulated evaluation value calculated for each generation is stored.

  The health trend data 90 is provided for each virtual server 12 and grasps the influence of each virtual server 12 on the virtualization infrastructure. Alternatively, the cumulative evaluation values of the virtual servers 12 may be totaled for each analysis number (generation) 901 to grasp the health trend of the entire virtualization infrastructure.

  FIG. 10 is a diagram illustrating a configuration example of the second analysis result data (analysis result 2). The analysis result 2 100 is a summary of the analysis results for each generation in which the configuration data 50 and the reference data 70 are compared and analyzed. The analysis result 2 100 includes a host name 1001, an evaluation item 1002, a result 1 1003 to a result 5 1007, and a matching pattern 1008. Since the host name 1001 and the evaluation item 1002 are the same as the host name 801 and the evaluation item 802 of the analysis result 180, description thereof is omitted.

  Result 1 1003 to Result 5 1007 are analysis results in each generation, Result 1 1003 is the analysis result in the first generation, Result 2 1004 is the analysis result in the second generation, and Result 3 1005 is in the third generation. The result 4 1006 is the analysis result in the fourth generation, and the result 5 1007 is the analysis result in the fifth generation (latest generation). Each entry stores an evaluation result for each evaluation item. Incidentally, “◯” represents conformity / sound, and “x” represents non-conformity / unsound.

  The matching pattern 1008 stores the result of what matching pattern the analysis result matches with the result 1 1003 to the result 5 1007. That is, the management server 16 determines which pattern in FIGS. 11A to 11B the contents of the result 1 1003 to the result 5 1007 are classified, and stores the determination result in the matching pattern 1008. In the example of FIG. 10, the patterns to be classified are the patterns P1 to P4, “undefined” that does not correspond to any pattern, and “−” indicating that classification is not necessary.

<Pattern classification>
FIG. 11 is a diagram illustrating a configuration example of pattern master data for classifying the tendency of the configuration check result. The pattern master data 110 is data for discriminating and classifying which pattern the analysis result 1 1003 to the analysis result 5 1007 of the analysis result 2 100 matches. The pattern master data 110 is discriminated by master data 1 110a that is determined by the number of consecutive healthy (OK) or unhealthy (NG), and by the number of healthy (OK) or unhealthy (NG) at a predetermined number of analysis times. There are two types of master data 2 110b. Both master data 1 110 a and master data 2 110 b are composed of a pattern 1101 and a classification 1102.

(1) Master data 1
When the pattern is NG four or more times in succession, the pattern 1101 of the master data 1 110a matches the “pattern P1”. Similarly, the pattern when it becomes NG three times or more continuously matches the “pattern P2”. On the other hand, the pattern when the pattern is OK three times or more is “pattern P3”, and the pattern when the pattern is OK four or more times is “pattern P4”. In this way, pattern classification is performed based on the number of consecutive healthy (OK) or unhealthy (NG), and the collection frequency of configuration data to be described later is updated by comparing with the analysis result 2100.

(2) Master data 2
The pattern 1101 of the master data 2110b matches the “pattern P1” when the number of analysis is NG 5 times or more in a predetermined number of analyzes (for example, 5 times). Similarly, the pattern when it becomes NG three times or more matches the “pattern P1”. Conversely, the pattern when it is OK three or more times is “pattern P3”, and the pattern when it is OK five or more times is “pattern P4”. In this way, pattern classification is performed based on the number of healthy (OK) or unhealthy (NG) in a predetermined number of analysis times, and the collection frequency of configuration data described later is updated by collating with the analysis result 2100.

  FIG. 12 is a diagram illustrating a configuration example of pattern data for determining increase / decrease in configuration data collection frequency with the adapted pattern data. The pattern data 120 includes a pattern 1201, a classification (increase / decrease number) 1202, and an increase / decrease number 1203 to the collection frequency. The pattern 1201 stores the entries of the above-described patterns P1 to P4. The classification (number of increase / decrease) 1202 defines the number of collections that increase / decrease for each pattern. If the pattern 1201 is “pattern P1”, the classification (increase / decrease number) 1202 is defined as “increase twice”, and the collection frequency is increased by adding twice to the current value of the collection frequency. Similarly, “pattern P2” is increased once, “pattern P3” is maintained as it is, and “pattern P4” is decreased once.

  In this manner, the frequency of collecting the configuration data 50 is updated by counting the number of times of NG (unhealthy) or OK (healthy) continuously or during a predetermined number of times, and extracting a pattern that matches that. As a result, if it is determined that the trend of a healthy state continues, the collection data is collected to the virtualization infrastructure by the management server 16 with a low collection frequency, and the load is reduced. On the contrary, if it is determined that the unhealthy tendency continues, the configuration data 50 is collected at a high frequency, and the collected configuration data 50 is immediately analyzed, whereby the health of the virtual server 12 or the entire virtualization infrastructure is collected. It is possible to quickly and accurately extract the factors that are deteriorating and take appropriate countermeasures.

<Overall processing>
FIG. 13 is a flowchart showing the overall processing of the virtualization infrastructure soundness check. In this example, the subject of processing is the CPU 111 of the management server 16, but it may be a control program running on the CPU 111 or the management server 16 itself. This process is executed by the scheduler of the management server 16, but the management server 16 may receive and execute an execution instruction from the system administrator.

  In step S1301, the CPU 111 of the management server 16 determines whether to start a soundness check. When the check is started (Yes), the CPU 111 executes S1302, and when not (No), the entire process of the virtualization infrastructure health check is terminated. In S1302, the CPU 111 executes configuration data collection processing (FIG. 14). In step S1303, the CPU 111 executes configuration data analysis processing (FIG. 15). In step S1304, the CPU 111 executes configuration data comparison processing (FIG. 16). In step S1305, the CPU 111 executes soundness determination processing (FIG. 17).

  In step S <b> 1306, the CPU 111 determines whether to display the soundness determination result and a countermeasure corresponding to the result on the display unit 116 of the management server 16. When not displaying (No), S1308 is executed, when displaying (Yes), S1307 is executed, and the CPU 111 executes. In step S <b> 1307, the CPU 111 displays the determination result and countermeasure on the display unit 116 of the management server 16. In step S <b> 1308, the CPU 111 transmits the determination result and the countermeasure to the management server 61 of the maintenance center 6. In step S1309, the CPU 111 executes soundness item / check frequency update processing (FIG. 18). After completing the processing in S1309, the CPU 111 ends the overall processing of the virtualization infrastructure soundness check.

  Further, the above-described overall processing is performed on each data center in FIG. 2 to collect configuration data / failure information / health tendency from each data center and update the soundness item master data, so that various types of virtual It is possible to check the soundness of the infrastructure.

<Collection processing>
FIG. 14 is a flowchart showing the configuration data collection process. In S1401, the CPU 111 of the management server 16 collects the configuration data 50 of each virtual server 12 (from the virtual server 01 to the virtual server 11). In step S1402, the CPU 111 compares the configuration data acquired in the past and stores the difference data in the configuration data DB 21. In order to simplify the process, the collected configuration data itself may be stored in the configuration data DB 21.

<Analysis processing>
FIG. 15 is a flowchart showing configuration data analysis processing. In step S <b> 1501, the CPU 111 of the management server 16 calculates a virtual infrastructure scale amount from the acquired configuration data 50. As described above, this is classified into one of small / medium / large scale based on the number of virtual servers, that is, the number of acquired configuration data.

  In step S <b> 1502, the CPU 111 extracts soundness check items from the soundness item master data 60 with the calculated scale amount, and generates reference data 70. In step S1503, the CPU 111 stores the reference data 70 in the soundness item master data DB 22. Thereby, the reference data for judging the soundness according to the scale amount can be generated.

<Comparison process>
FIG. 16 is a flowchart showing the configuration data comparison process. In S <b> 1601, the CPU 111 of the management server 16 acquires the configuration data 50 of each virtual server from the configuration data DB 21. In S1602, the CPU 111 acquires the reference data 70 from the soundness item master data DB 22.

  In step S1603, the CPU 111 compares the configuration data 50 with the reference data 70. That is, the state 503 and the reference value 703 in the same evaluation item (the evaluation item 502 of the configuration data 50 and the evaluation item 702 of the reference data 70) are compared. The CPU 111 executes the comparison processing for all the evaluation items, and uses the processing result as the comparison result. In step S <b> 1604, the CPU 111 stores the comparison result as the analysis result 180 in the analysis result data DB 23. This analysis result 180 is generated for each virtual server and stored in the analysis result data DB 23 by the CPU 111.

<Health assessment>
FIG. 17 is a flowchart showing soundness determination processing. In step S <b> 1701, the CPU 111 of the management server 16 acquires the latest analysis result in each virtual server 12 from the analysis result data DB 23. In S1702, the CPU 111 generates the health trend data 90 in each virtual server from the acquired analysis result and the past analysis result. In S <b> 1703, the CPU 111 identifies a risk that is highly likely to occur from the generated health trend data 90 and the latest analysis result 180 from the risk content 607 of the health item master data 60.

  In S <b> 1704, the CPU 111 acquires a countermeasure for the identified risk from the analysis result data DB 23. For example, if the risk is performance deterioration, a response destination that extracts the evaluation item 601 to the latest version of the virtualization tool or restores the changed arrangement is extracted. By presenting the latest analysis result 180 of this countermeasure, the system administrator can grasp the soundness of the virtualization infrastructure.

<Check frequency update process>
FIG. 18 is a flowchart showing soundness item / check frequency update processing. In step S <b> 1801, the CPU 111 of the management server 16 acquires the latest analysis result from the analysis result data DB 23. For example, the CPU 111 acquires the latest analysis result in the virtual server 01 from the analysis result data DB 23. In step S <b> 1802, the CPU 111 acquires past analysis results for several generations from the analysis result data DB 23. For example, the past analysis results in the virtual server 01 are acquired by the CPU 111 from the analysis result data DB 23. Then, the CPU 111 generates an analysis result 2100 in each virtual server.

  In S1803, the CPU 111 analyzes which pattern is classified from the latest analysis result and the past analysis result in the virtual server 01 using the pattern master data 110a to the pattern master data 110b (stored in the pattern master data DB 24). To do. That is, the CPU 111 identifies which matching pattern is the analysis result 2 100 (any one of the patterns P1 to P4).

  In step S <b> 1804, the CPU 111 collates the specified matching pattern against the pattern data 120 and determines an increase / decrease value of the collection frequency of the configuration data 50. For example, if the evaluation item 1002 of the analysis result 2 100 is “high availability”, the matching pattern 1008 is “pattern P4”. Then, it is specified as “decrease once” from the classification 1202 of the pattern data 120, and the collection frequency of the configuration data 50 is reduced once.

  In the example of FIG. 10, the matching pattern is specified for each evaluation item, and the update of the collection frequency of the evaluation item of the configuration data is shown. As a result, it is possible to collect data at a high frequency for evaluation items that adversely affect the soundness of the virtual server, thereby enabling early extraction of problems and quick execution of countermeasures. In addition, the adaptation frequency may be specified not in the evaluation item unit but in the virtual server unit to change the collection frequency of the configuration data.

  In step S1805, the CPU 111 updates the current collection frequency with the determined increase / decrease value. That is, the CPU 111 adds and updates the determined increase / decrease value to the collection frequency 705 of the reference data 70. In step S1806, the CPU 111 first stores the reference data whose collection frequency has been updated in the soundness item master data DB 22.

  In step S1807, the CPU 111 extracts the difference between the evaluation items for each virtual server from the current analysis result and the past analysis result. If there is a difference, the reference data 70 for each virtual server is updated with the difference contents, and the soundness item master data 60 is also updated. For example, if the AutoDeploy function is added as shown in FIG. 1, an item related to the AutoDeploy function is added. In S1808, the CPU 111 stores the updated soundness item master data and the reference data 70 in the soundness item master data DB 22.

  As described above, it is possible to extract an item for checking the soundness from the scale of the system constituting the virtualization infrastructure, perform a predetermined number of checks on the item, and identify a risk that is considered to be generated from the tendency of the result. Therefore, it is possible to identify faults in advance and implement countermeasures, thereby improving the reliability and availability of the virtualization infrastructure.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. The above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function may be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD. Further, the control lines and information lines are those that are considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

1, 3, 4, 5: Data center, 6: Maintenance center, 7: Network, 11: Physical server, 12: Virtual server, 13: Virtual machine group, 14: Virtual machine, 15: Shared storage, 16: Management server 17: user terminal, 18: network, 21: configuration data DB, 22: soundness item master data DB, 23: analysis result data DB, 24: pattern master data DB, 50: configuration data, 60: reference data, 61 : Management server, 70: Soundness item master data, 80: Analysis result 1, 90: Soundness trend data, 100: Analysis result 2, 110: Pattern master data, 120: Pattern data, 111, 171: CPU, 112, 172: Memory, 113, 173: Storage device, 114, 174: Input device, 115, 175: Output device, 16,176: display device, 117,177: communication device

Claims (8)

An information system comprising a plurality of physical servers, a management server, and a plurality of user terminals,
A virtual server is operated on the physical server, and a virtualization platform on which a virtual machine used in the user terminal operates is constructed on the virtual server,
The management server
A CPU for controlling the entire apparatus, a memory unit for temporarily storing collected data and analysis results,
A storage device unit for storing collected data and analysis results,
Health item master data for determining the health of the virtualization infrastructure in the memory unit or the storage device unit, pattern master data for classifying a plurality of health evaluation results into a specific pattern, and configuration data of the specific pattern Stores pattern data that increases or decreases the number of collections.
The CPU of the management server
Collecting configuration data from each of the virtual servers;
Calculate the scale of the virtualization platform from the collected configuration data, extract the items for judging the soundness at the calculated scale and the evaluation criteria thereof from the soundness item master data, and generate the reference data,
The information system, wherein the soundness is determined by comparing the reference data with the configuration data of the virtual server. The information system according to claim 1, wherein the soundness item master data has an allowable use period, a priority, an importance level, and an occurrence risk in addition to an item for determining soundness and an evaluation standard for each item. Characteristic information system. 3. The information system according to claim 2, wherein the reference data includes items for judging soundness, evaluation criteria for each item, the number of times of collecting configuration data per unit time, and the upper limit of the number of times of collecting configuration data per unit time. An information system characterized by The information system according to claim 3, wherein in the determination of soundness, items that satisfy the evaluation criteria or are within the allowable use period are determined to be healthy, and other items are determined to be unhealthy, and each determined item is determined. A soundness evaluation value is calculated from the determination result, the priority, and the importance. The information system according to claim 4, wherein the risk of occurrence is specified by generating a tendency of the degree of soundness from the evaluation value of soundness calculated for each collection of the configuration data.   6. The information system according to claim 5, wherein the number of collections of the configuration data is increased or decreased according to the tendency of the soundness determination result.   4. The information system according to claim 3, wherein the contents of the soundness master data are updated with the soundness determination result. 6. The information system according to claim 5, wherein the information system further includes a maintenance center having a management server for a maintenance center, and a virtualization platform different from the virtualization platform by combining the physical server and the virtual server. 1 or more, and the maintenance center management server obtains configuration data, soundness evaluation results, and occurrence risk from each configured virtualization platform.

Images