WO2013105174A1

WO2013105174A1 - Service level management device, service level management method and program, and service level monitoring system

Info

Publication number: WO2013105174A1
Application number: PCT/JP2012/007860
Authority: WO
Inventors: 遼介外川
Original assignee: 日本電気株式会社
Priority date: 2012-01-12
Filing date: 2012-12-07
Publication date: 2013-07-18
Also published as: JPWO2013105174A1

Abstract

A service level management device (100) is provided with an analysis unit (106), an aggregation unit (102) and a service level computation unit (104). Said analysis unit (106) analyzes a service level definition table (110), and on the basis of a unit conversion definition table (108) that defines a relationship between aggregation units with different attributes, defines multiple different service levels having the same attribute and different aggregation units, and defines a grouping aggregation unit that groups aggregation units having the same attribute. Said aggregation unit (102) finds an aggregate value for management data. On the basis of multiple service level definition tables (110), said service level computation unit (104) computes a service level from the aggregate value. The aggregation unit (102) aggregates the management data for each grouping aggregation unit to find the aggregate value. On the basis of the aggregate value for each grouping aggregation unit, the service level computation unit (104) computes the service level defined by the multiple service level definition tables (110).

Description

Service level management apparatus, service level management method, program, and service level monitoring system

The present invention relates to a service level management device, a service level management method, a program, and a service level monitoring system, and more particularly, to a service level management device, a service level management method, a program, and a service level for managing service quality. It relates to a monitoring system.

Recently, in the IT (Information Technology) field, usage forms called cloud computing such as SaaS (Software as a Service) have increased. In the cloud computing environment, services are the center of contracts, not software or hardware.

In such a contract, a service level agreement (SLA) is used as a means for assuring service quality. SLA is a service level (SL) (for example, service content, scope, and quality) related to each of one or more service level items (Service Level Objective: SLO) connected between a service provider and a user. ) A quantitative and explicit agreement. The service level of the service level item in the SLA is, for example, “the service operation rate is 99.9% or more”, “the average recovery time when a failure occurs is within 12 hours”, and the like.

An example of a service evaluation system is described in Patent Document 1. In the service evaluation system disclosed in Patent Document 1, first, a user of the service evaluation system transmits a data processing command to the service evaluation system via a network. Next, the service evaluation system collects operational data (operation log, number of processes, etc.) of the server group for each service (business unit) according to the data processing command, and calculates service levels such as operation rate and average processing time.

Furthermore, in the example of the service evaluation system described in Patent Document 2, first, the system converts SLA definition information (that is, service level items and service level definition information) given in advance into monitoring items (setting values, Monitoring method). Next, the service evaluation system monitors the individual items of the software service (for example, the operating status, the response status, and the number of connected users), and records the history. Next, the service evaluation system analyzes the recorded history, and records the result of analysis based on the history of a predetermined period (for example, from 0:00 to 24:00 on the previous day) as a usage record.

Further, in the example of the data processing system described in Patent Document 3, the data processing system includes a pre-existing total table, and a control table having information on the total table items and the number of data items. This data processing system searches for a summary table that minimizes the number of data out of the summary tables that can be aggregated based on the control table when calculating a summary value that meets the requested conditions. The total value is calculated based on the total table.

JP 2009-146001 A JP 2010-257272 A JP 2001-256242 A

In the system described above, even if the monitored service is the same, it is necessary to monitor a plurality of service levels according to the user and the service status, and there is an overlap in the calculation process of these service levels. As the system scale increases, the calculation processing load increases.

An object of the present invention is to provide a service level management apparatus, a service level management method, a program, and a service level monitoring system that can solve the above-described high processing load.

The service level management apparatus of the present invention
Analyze multiple service level definitions that define how to calculate the service level, which is the assurance level that the information processing system must satisfy, and use the same attribute based on a unit conversion definition table that defines the relationship between different aggregation units for each attribute. Analyzing means for defining a plurality of different service levels having the aggregation unit, an aggregation aggregation unit for collecting the aggregation units between the same attributes,
Aggregation means for obtaining the aggregated value by aggregating the management data for monitoring the operating state of the information processing system in the aggregation unit;
Service level calculation means for calculating a service level from the aggregated values aggregated by the aggregation means based on a plurality of the service level definitions;
With
The aggregation means calculates the management data for each of the aggregation aggregation units defined by the analysis means to obtain an aggregation value,
The service level calculation means calculates the service level defined by a plurality of the service level definitions based on the aggregate value for each of the aggregate aggregation units aggregated by the aggregation means.

The service level management method of the present invention comprises:
A service level management method in a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
The management device is
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes,
Management data for monitoring the operating state of the information processing system is aggregated for each aggregation unit using the defined aggregation aggregation unit to obtain an aggregate value,
Based on the plurality of service level definitions defining the service level, the service levels defined in the plurality of service level definitions are calculated based on the aggregated values for each of the aggregated units.

The computer program of the present invention is:
In a computer that realizes a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
A procedure for calculating a management value obtained by monitoring the operation state of the information processing system, and calculating a total value for each total unit using the defined total unit
Based on a plurality of service level definitions defining the service level, and executing a procedure for calculating a service level from the aggregated values defined in the plurality of service level definitions based on the aggregated values for each aggregated unit It is for making it happen.

The service level monitoring system of the present invention comprises:
Comprising the service level management device of the present invention,
The operation state of the information processing system is monitored using the calculated service level.

The computer program of the present invention is:
A computer that realizes a service level management device that calculates a service level that is a guarantee level to be satisfied by the information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
This is to execute a procedure for outputting the defined aggregated unit.

It should be noted that an arbitrary combination of the above-described components and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

The various components of the present invention do not necessarily have to be independent of each other. A plurality of components are formed as a single member, and a single component is formed of a plurality of members. It may be that a certain component is a part of another component, a part of a certain component overlaps with a part of another component, or the like.

In addition, although a plurality of procedures are described in order in the method and computer program of the present invention, the order of description does not limit the order in which the plurality of procedures are executed. For this reason, when the method and computer program of the present invention are implemented, the order of the plurality of procedures can be changed within a range that does not hinder the contents.

Furthermore, the plurality of procedures of the method and computer program of the present invention are not limited to being executed at different timings. For this reason, another procedure may occur during the execution of a certain procedure, or some or all of the execution timing of a certain procedure and the execution timing of another procedure may overlap.

According to the present invention, it is possible to provide a service level management apparatus, a service level management method, a program, and a service level monitoring system that reduce processing load and are efficient.

The above-described object and other objects, features, and advantages will be further clarified by a preferred embodiment described below and the following drawings attached thereto.

It is a functional block diagram which shows the structure of the service level monitoring system which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the management data storage part of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the time definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the space definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the individual definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the unit variable definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level calculation definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the unit variable data storage part of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level memory | storage part of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the partial structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the management data collection definition table of the service level monitoring system which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level monitoring system which concerns on embodiment of this invention. It is a functional block diagram which shows the structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the method definition table of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the unit variable definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level calculation definition table of the service level management apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level variable definition table of the service level management apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the service level calculation definition table of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the unit variable table of the service level management apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the principal part structure of the service level monitoring system which concerns on embodiment of this invention. It is a figure which shows an example of the structure of the structure information table of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention. It is a flowchart which shows an example of operation | movement of the service level management apparatus which concerns on embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same reference numerals are given to the same components, and the description will be omitted as appropriate.

(First embodiment)
FIG. 1 is a functional block diagram showing a configuration of a service level monitoring system 1 according to the embodiment of the present invention.
As shown in FIG. 1, the service level monitoring system 1 of this embodiment is connected to an information processing system such as a data center 5 via a network 3, for example. The service level monitoring system 1 includes a service level management apparatus 100 that monitors the operating state of the data center 5 and manages a service level that is a guarantee level to be satisfied by the information processing system.
The data center 5 is a facility where a large number of devices such as computers and communication devices are installed and managed.

As described above, a service level agreement (SLA) that guarantees the quality of the service is made between the provider who provides the service to the user and the user. SLA is a contract of quantitative and explicit agreement for service level (SL) (eg, service content, scope and quality) for each of one or more service level items (Service Level Objective: SLO). is there. The service level of the service level item in the SLA is, for example, “the service operation rate is 99.9% or more”, “the average recovery time when a failure occurs is within 12 hours”, and the like.

In general, when calculating the service level value, the time span for aggregation (quarter, January, 1st, etc.), the scale of the system to be monitored (the entire data center, the entire service, the web (WEB) layer) , Application (APplication: AP) layer, database (DataBase: DB) layer (server), etc.) or calculation method (total, average, maximum, minimum, percentile, etc.). In addition to these, the service level used for the SLA can be defined only after the threshold is set.

Furthermore, even if the target service is the same, the service level requires a plurality of different definitions depending on the user and the service status. For example, regarding the range of time required for aggregation, the customer management needs a quarterly service level, but the user responsible for operation, such as the information system department, needs a daily unit. It is assumed that the service level is

The present invention is intended to improve the processing efficiency of the service level monitoring system 1 that calculates and monitors the service level for the service level item in the SLA. In particular, the present invention can be suitably applied to the operation of a system for which a service level contract is concluded. The service level management apparatus 100 of the present invention calculates and manages a service level for monitoring the operating state of an information processing system such as the data center 5.

The service level management apparatus 100 according to the embodiment of the present invention defines a method for calculating a service level that is a guarantee level to be satisfied by an information processing system (computer (not shown) or device (not shown) of the data center 5). Based on the unit conversion definition table 108 that analyzes a plurality of service level definitions (stored in the service level definition table 110) and defines the relationship between different aggregation units of each attribute, different services having different aggregation units with the same attribute The analysis unit 106 that defines the aggregation unit (unit variable) for grouping the aggregation units between the same attributes and the operating state of the information processing system (computer (not shown) or device (not shown) of the data center 5) Aggregation unit 10 that aggregates management data to be monitored (stored in management data storage unit 160) in an aggregation unit to obtain an aggregate value. When provided on the basis of a plurality of service level definition (stored in the service level definition table 110), and the service level calculation unit 104 for calculating a service level from a total value of aggregating unit 102 aggregates, the.
The aggregation unit 102 aggregates management data for each aggregation aggregation unit (unit variable) defined by the analysis unit 106 to obtain an aggregation value (unit variable data).
The service level calculation unit 104 is defined by a plurality of service level definitions (stored in the service level definition table 110) based on the aggregate value (unit variable data) for each aggregate aggregation unit (unit variable) aggregated by the aggregation unit 102. Service level is calculated.

Specifically, the service level management apparatus 100 according to the present embodiment is connected to a management data storage unit (indicated as “management data” in the figure) 160 that stores management data collected from the data center 5 via the network 3. Is done. The management data is service level original data monitored by the service level monitoring system 1 of the present embodiment. Management data is appropriately collected from the data center 5 via the network 3 and stored in the management data storage unit 160. The management data will be described in detail in an embodiment described later. The management data storage unit 160 may be included in the service level management device 100 or may be included in a storage device connected to the service level management device 100. Alternatively, the management data may be recorded on a recording medium that can be read by the service level management apparatus 100.

The service level management apparatus 100 according to the present embodiment includes a totaling unit 102, a service level calculation unit 104, an analysis unit 106, a unit conversion definition table storage unit (shown as “unit conversion definition table” in the figure) 108, A service level definition table 110, a unit variable definition table 150, and a service level storage unit (shown as "service level" in the figure) 170 are provided.

The service level management apparatus of the present invention includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and an I / O (Input / Output) (not shown), a keyboard, Implemented by an input device (not shown) such as a mouse, an output device (not shown) such as a display or printer, a computer connected to another device (not shown) via a network, or a device equivalent thereto Can do. These elements of the service level management apparatus are connected to each other via a bus (not shown), and the entire service level management apparatus is controlled together with the elements by the CPU. Then, the CPU reads out the program stored in the ROM to the RAM and executes it, so that each function of each unit of each service level management device of the present invention to be described later can be performed by any combination of computer hardware and software. Can be realized. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus. Each figure used for description shows not a configuration in hardware units but a block in functional units.
Moreover, in each figure, about the structure of the part which is not related to the essence of this invention, it has abbreviate | omitted and is not illustrated.

In the present embodiment, only one service level management apparatus 100 is shown, but the present invention is not limited to this. When monitoring a large-scale data center 5, a plurality of service level management devices 100 may be provided. Further, the service level management apparatus 100 may be configured by a virtual server or the like. The same applies to the service level management apparatus of other embodiments.

The computer program of the present embodiment analyzes a plurality of service level definitions that define a service level calculation method in a computer for realizing the service level management apparatus of the present invention, and has a plurality of different attributes having different aggregation units with the same attribute. The service level is defined to execute a procedure for defining an aggregation unit (unit variable) for grouping aggregation units between the same attributes, and a procedure for outputting the defined aggregation unit (unit variable).
The computer program may be in any form such as an object code or a program executed by an interpreter. The above-described program may be provided to a computer in a form incorporated in application software that can be operated on a general personal computer or a portable information terminal.

Further, the computer program of the present embodiment analyzes a plurality of service level definitions (stored in the service level definition table 110) that define a service level calculation method in a computer for realizing the service level management apparatus of the present invention. , A procedure for defining an aggregation unit that aggregates aggregation units between the same attributes for different service levels with the same attribute and different aggregation units, an aggregation unit that defines management data that monitors the operating status of the information processing system Aggregated for each aggregated unit using (unit variable), aggregated based on the procedure for obtaining the aggregated value (unit variable data), multiple service level definitions (stored in service level definition table 110) defining service levels Based on the aggregated value (unit variable data) for each unit, multiple service levels are defined. It is described defined aggregate value (stored in the service level definition table 110) from (in variable data) so as to perform the steps of calculating a service level.

The computer program of the present invention may be recorded on a computer-readable recording medium (not shown). The recording medium is not particularly limited, and various forms can be considered. The program may be loaded from a recording medium into a computer ROM, or downloaded to a computer through a network and loaded into the ROM.

The information stored in the unit conversion definition table storage unit 108, the service level definition table 110, the unit variable definition table 150, or the service level storage unit 170 in FIG. Can be stored in ROM or RAM. Alternatively, these pieces of information may be stored in a storage device (not shown) connected to the service level management apparatus 100 and accessible by the service level management apparatus 100 or another computer (not shown).

The service level definition table 110 includes definitions of a plurality of service levels that the service level monitoring system 1 should monitor. Various service levels to be monitored are included. For example, the service level may include availability such as availability, or information regarding performance such as response time (TAT) and CPU usage, or power consumption.

The service level definition table 110 defines the types of service levels, the time and space ranges to be monitored, and the like. The service level monitoring system 1 can calculate the service level defined in the service level definition table 110 and monitor the service level with a set threshold. Note that the monitoring process itself based on the service level threshold is not related to the essence of the present invention, and thus detailed description thereof is omitted.

The unit conversion definition table storage unit 108 stores a unit conversion definition table that defines unit conversion used when the analysis unit 106 collects different aggregation units with the same attribute. As described above, when calculating the value of the service level, the time range for aggregation (quarter, January, 1st, etc.), the scale of the system to be monitored (the entire data center, the entire service, the web ( (WEB) layer, application (AP) layer, database (DataBase: DB) layer (server), etc.) or calculation method (total, average, maximum, minimum, percentile, etc.) is required.

In this embodiment, the relationship between the counting units is defined for each counting time, counting space, or counting method, and the unit conversion definition table is provided. For the total time, the total unit may be, for example, one year, half year, quarter, two months, one month, two weeks, one week, one day, one hour, or ten minutes. Reciprocal relationships between the total units, for example, 365 days for one year, 24 hours for one day, and the like are stored in advance in the unit conversion definition table. Similarly, regarding the total space or the total method, the interrelationships between the respective total units are held as unit conversion definition tables. Various unit conversion definition tables stored in the unit conversion definition table storage unit 108 will be described in detail in an embodiment described later.

The unit conversion definition table storage unit 108 stores a plurality of unit conversion definition tables. Each unit conversion definition table defines a relationship between different aggregation units of each attribute.
Based on each unit conversion definition table, the analysis unit 106 defines and outputs a plurality of different service levels having the same attribute and different aggregation units as unit variables for grouping the aggregation units between the same attributes.
The unit variable definition table 150 holds unit variables defined by the analysis unit 106.

For example, the TAT maximum value (MAX) in which the aggregation time is one day and the aggregation space is a node is defined as the service level, and the TAT maximum value (MAX) in which the aggregation time is one month and the aggregation space is a node. ) Is defined as a service level. The analysis unit 106 refers to the unit conversion definition table because the two service levels are the same in the attribute TAT maximum value (MAX) and the aggregation time is different. To be consolidated. Then, the analysis unit 106 defines the maximum value (MAX) of TAT whose aggregation time is one day and whose aggregation space is a node as a unit variable (aggregation aggregation unit).

In this way, the analysis unit 106 analyzes a plurality of service level definitions (hereinafter also referred to as “service level definition 110”) in the service level definition table 110. And the analysis part 106 defines what has a smaller total unit as a unit variable. Or the analysis part 106 defines the aggregation which should be performed previously as a unit variable, when the aggregation method has a subordinate relationship. Then, the analysis unit 106 aggregates the aggregation units.

The aggregation unit 102 uses the management data stored in the management data storage unit 160 to aggregate using the unit variables defined by the analysis unit 106, and outputs the aggregation results as unit variable data. In the present embodiment, the unit variable data output from the counting unit 102 is input to the service level calculation unit 104.

The service level calculation unit 104 calculates and outputs each service level defined in the service level definition table 110 using the unit variable data tabulated by the tabulation unit 102. At this time, the service level calculation unit 104 avoids duplication processing by redefining other service level calculation methods using the aggregated unit (unit variable). For example, the maximum value (MAX) of TAT in which the aggregation unit is one month and the aggregation space is a node is one month based on unit variable data in which the aggregation unit aggregated by the aggregation unit 102 is one day and the aggregation space is a node. You only have to count minutes.
The service level calculation method may be redefined by the analysis unit 106.

In the present embodiment, the calculated service level is stored in the service level storage unit 170. The service level is displayed on a display device or the like. Alternatively, the service level is processed by various application programs and presented to the user.

In the configuration as described above, a service level management method by the service level management apparatus 100 according to the embodiment of the present invention will be described below.
FIG. 2 is a flowchart showing an example of the operation of the service level management apparatus 100 according to the embodiment of the present invention.
The service level management method of the present invention is a management apparatus (service level management apparatus 100) for managing a service level that is a guarantee level to be satisfied by an information processing system (computer (not shown) or equipment (not shown) of the data center 5). The service level management apparatus 100 analyzes a plurality of service level definitions (stored in the service level definition table 110) that define a service level calculation method, and has different aggregation units with the same attribute. Aggregate aggregation unit (unit variable) for grouping aggregation units between the same attributes for a plurality of different service levels is defined (step S101), and management data for monitoring the operating state of the information processing system (stored in the management data storage unit 160) Using the defined aggregation unit (unit variable) Aggregated for each variable) to obtain an aggregated value (unit variable data) (step S103), and aggregated aggregated units (units) based on a plurality of service level definitions (stored in the service level definition table 110) defining service levels Based on the total value (unit variable data) for each variable), service levels defined by a plurality of service level definitions (stored in the service level definition table 110) are calculated (step S105).

As described above, according to the service level monitoring system 1 according to the embodiment of the present invention, when a plurality of service levels are calculated, duplication of calculation procedures is automatically detected, and processing for avoiding duplication is performed. Therefore, it is possible to reduce the processing load and increase the efficiency.

(Second Embodiment)
FIG. 3 is a functional block diagram showing the configuration of the service level monitoring system 1 according to the embodiment of the present invention.
The service level monitoring system 1 of this embodiment includes a service level management device 200. The service level management apparatus 200 of this embodiment shows an example of a specific configuration for realizing the service level management apparatus of the present invention. The service level management apparatus 200 of this embodiment has the same configuration as that of the service level management apparatus 100 of the above embodiment, and further includes a service level display unit 208 and a unit variable data storage unit (in the figure, “unit variable data”). ) 210 and a service level calculation definition table 220.

In the embodiment of the present invention, each data is described as being stored in a table format. However, the present invention is not limited to this and can be stored in various forms. For example, each data may be in the form of a file described in an arbitrary programming language or a text file described in CSV (Comma Separated Values) format. Alternatively, each data may be in a database format. It is desirable that each data can be viewed and edited by a user using an arbitrary editor or the like.

The management data storage unit 160 stores management data collected from the data center 5.
In the present invention, management data relates to information such as unscheduled downtime availability information, response time (TAT), CPU usage rate, memory usage, disk usage, or batch processing completion rate, etc. Includes information or power consumption.

As shown in FIG. 4, the management data storage unit 160 shows at least one record 165 (FIG. 4 shows five records 165a to 165e. Hereinafter, when there is no need to distinguish each record, “ 165 "). Each record 165 records a log of each data collected from the monitored system. Each record 165 includes a recording date 161, a management data name 162, a target 163, or a value 164. The configuration of the management data stored in the management data storage unit 160 in FIG. 4 is an example, and the present invention is not limited to this.

The recording date 161 indicates the date when the management data was recorded in the management data storage unit 160. Further, the recording date 161 may be the date when management data is collected. The management data name 162 indicates the name of the recorded management data. The object 163 indicates a space where management data is collected, for example, a node, a service, or a data center. A value 164 indicates the value of management data actually collected. Here, values and units are recorded, but only numerical values may be used.
In addition, although the data of each item of each table described in the embodiment of the present invention is shown by type, name (TAT or node A), etc., it is not limited to this, for example, by identifiable information such as an identifier. I just need it.

FIG. 4 shows that “TAT” of “service A” recorded in “2011/08/01, 15:00” is “10 msec”. Similarly, “unscheduled stop time” of “service A” recorded at the same time is “2 minutes”. This example means, for example, that the unscheduled stop time in the past hour was “2 minutes”. Further, the value of “operation status” of “node A” at the same time is “response received”. This example means that there was a response to the ping for checking the operating status at the recorded time.

As shown in FIG. 5, the service level definition table 110 stores a plurality of definitions such as the type of service level to be monitored by the service level monitoring system 1 and the range of time and space to be monitored. The service level monitoring system 1 can calculate the service level defined in the service level definition table 110 and monitor the service level with a set threshold value. Note that the monitoring process itself based on the service level threshold is not related to the essence of the present invention, and thus detailed description thereof is omitted.

The service level of the present invention includes various things. For example, service level includes availability such as availability and service time, reliability such as average recovery time, or performance information such as response time (TAT), CPU usage, memory usage, or disk usage, consumption The amount of electric power, the amount of carbon dioxide conversion, or Power Usage Effectiveness (PUE) can be included.

As shown in FIG. 5, the service level definition table 110 shows at least one record 117 (FIG. 5 shows three

records

117a, 117b, and 117c. Hereinafter, when there is no need to distinguish each record in particular. , “117”). Each record 117 includes a service level (hereinafter also referred to as “SL (Service Level)”) name 111, an SL type 112, an aggregation variable 113, an aggregation time 114, an aggregation space 115, or an aggregation method 116.

SL name 111 indicates a name for identifying a service level. The SL type 112 indicates the type of service level to be calculated. The total variable 113 indicates the type of management data used for calculating the service level value. The total time 114 indicates a time range in which the total variable 113 is totaled and the service level value is calculated. The total space 115 indicates the range of the space where the total variable 113 is totaled and the service level value is calculated. The aggregation method 116 indicates a method for aggregating the aggregation variable 113. The configuration of the service level definition table 110 in FIG. 5 is an example, and the present invention is not limited to this.

The SL type 112 includes, for example, availability such as operation rate or service time, reliability such as average recovery time, and response time (Turn Around Time: TAT), CPU usage rate, memory usage, or disk usage. Information on performance, power consumption, carbon dioxide equivalent, or power usage efficiency (PUE).
The aggregate variable 113 is, for example, service level values listed in the SL type 112 such as unscheduled stop time, response time (TAT), information on performance such as CPU usage rate, memory usage, disk usage, and power consumption. Is the type of management data required to calculate
The total time 114 is, for example, one year, quarter, two months, one month, two weeks, one week, one day, one hour, ten minutes, or one minute.

The total space 115 includes, for example, a plurality of data centers, one data center, a management unit defined based on an operation system, a customer, a customer department, a specific system, an entire service, an AP layer, a WEB layer, and a DB layer. Layers, nodes, etc.
The aggregation method 116 is, for example, sum, average, maximum, minimum, percentile, and the like. Alternatively, the counting method 116 may specify a calculation formula. In the example of the service level definition table 110 illustrated in FIG. 5, the aggregation method of “operation rate” is defined as “expression”, but this is defined separately.

The unit conversion definition table storage unit 108 in FIG. 3 stores a plurality of unit conversion definition tables that define unit conversions used when the analysis unit 106 in FIG. 3 collects different aggregation units with the same attribute.
In the present embodiment, the unit conversion definition table storage unit 108 includes at least one unit conversion definition table that defines the relationship between different aggregation units of each attribute shown in FIGS. In this embodiment, the unit conversion definition table storage unit 108 includes the time definition table 120 in FIG. 6, the space definition table 130 in FIG. 7, and the individual definition table 140 in FIG.

The time definition table 120 defines the relationship between time aggregation units. As shown in FIG. 6, the time definition table 120 shows at least one record 127 (FIG. 6 shows six records 127a to 127f. In the following, if there is no need to distinguish each record, “127 ").
Each record 127 can include a unit name 121 and unit converted values 122 to 126. As can be seen from the figure, the time aggregation units are inclusive of each other, and the time definition table 120 defines the inclusion relationship of the time aggregation units.
Note that the configuration of the time definition table 120 in FIG. 6 is an example, and the present invention is not limited to this.

The unit name 121 indicates the name of the time unit. The time unit means a time width of a variable used for calculating a service level value. For example, if the time unit is “1 day”, the time width of the variable used to calculate the service level value is “a value measured at intervals of 1 day” or “statistics such as average and sum of management data for 1 day” It means “processed value”. The unit conversion values 122 to 126 indicate the number of time aggregation units when the time in parentheses is based on the aggregation unit.
For example, in FIG. 6, “90” is recorded in “Unit conversion value (1 day)” of “Quarter”. This indicates that “quarter” is “90 days” when “1 day” is used as a unit. On the other hand, “-” is recorded in “Unit conversion value (1 year)” of “Quarter”. This indicates that there is no corresponding data because “1 year” is larger than “quarter”.

The space definition table 130 defines the relationship of the aggregation units of the space to be managed. As shown in FIG. 7, the space definition table 130 shows at least one record 135 (FIG. 7 shows five records 135a to 135e. Hereinafter, when it is not necessary to distinguish each record, “135 ").

Each record 135 can include a unit name 131 and unit conversion values 132 to 134. As can be seen from the figure, in this embodiment, the space aggregation units are in an inclusive relationship, and the space definition table 130 defines the inclusive relationship of the space aggregation units. The inclusion relationship is, for example, a relationship such as unit size, long / short, wide / narrow. Furthermore, the inclusion relationship can also include the relationship of the common parts between the aggregation units. In other embodiments, there may be cases where the aggregate units of space are not in an inclusive relationship. These relationships may be defined and stored in the space definition table 130.
The configuration of the space definition table 130 in FIG. 7 is an example, and the present invention is not limited to this.

The unit name 131 indicates the name of the space unit. The space unit means the width of a variable space used for calculating a service level value. For example, if the spatial unit is “AP layer”, the width of the space of the variable used for calculating the service level value is “value measured in the entire AP layer” or “management data of nodes included in the AP layer, It means “statistically processed value such as sum”.

Unit conversion values 132 to 134 indicate the number of space units when the space units in parentheses are used as a reference. For example, in FIG. 7, “10000” is recorded in “unit conversion value (node)” of “DC”. This indicates that “DC” is composed of “10000 nodes” when “node” is used as a unit. On the other hand, “−” is recorded in “Unit conversion value (plural DC)” of “DC”. This indicates that there is no corresponding data because “plural DC” is larger than “DC”.

The individual definition table 140 is a definition relating to a specific monitoring item, and stores information defining a counting method in which service levels are arbitrarily defined by mathematical formulas or the like. As shown in FIG. 8, the individual definition table 140 shows at least one record 146 (FIG. 8 shows one record of 146a. Hereinafter, if it is not necessary to distinguish each record, “146” is shown. Included).

Each record 146 includes an SL type 141, a tabulation method 142, a variable 143, a time unit 144, or a space unit 145.
The configuration of the individual definition table 140 in FIG. 8 is an example, and the present invention is not limited to this.

The SL type 141 indicates the type of service level calculated based on the individual definition defined in the individual definition table 140 (hereinafter also referred to as “individual definition 140”). The aggregation method 142 indicates a calculation formula or a function used for calculating the service level value. Note that m represents a scheduled operation time, and n represents an unscheduled stop time. The scheduled operating time shall be a constant value. Next, a variable 143 indicates a variable used for calculating a service level value. The time unit 144 indicates the range of the total time for calculating the service level value. The space unit 145 indicates the width of the total space for calculating the service level value.

For example, in FIG. 8, “unscheduled stop time” is a variable. The time unit 144 is “1 day” and the space unit 145 is “service”. Note that the collection method defined as “expression” by the aggregation method 116 of the record 117c in the service level definition table 110 in FIG. 5 corresponds to the aggregation method 142 defined in the individual definition table 140. In FIG. 5, the tabulation method 116 is shown as “expression”, but the individual definition table 140 can define information that can identify the formula of the tabulation method 142, for example, a function name.

Returning to FIG. 3, the analysis unit 106 is based on the unit conversion definition composed of the time definition table 120, the space definition table 130, or the individual definition table 140 recorded in the service level definition table 110 and the unit conversion definition table storage unit 108. Detect duplicate service level calculation processes. The analysis unit 106 then defines a service level calculation definition that is a definition for calculating a service level value while avoiding duplication, and a definition for calculating a value (total value) used for calculating the service level value. The unit variable definition is output.

In the present embodiment, the output unit variable definition and service level calculation definition are stored in the unit variable definition table 150 and the service level calculation definition table 220, respectively. Hereinafter, the unit variable definition stored in the unit variable definition table 150 is referred to as “unit variable definition 150”. Further, the service level calculation definition stored in the service level calculation definition table 220 is referred to as “service level calculation definition 220”.

The analysis unit 106 first refers to the service level definition table 110, reads the service level definition, and based on the time definition table 120, the space definition table 130, or the individual definition table 140 recorded in the unit conversion definition table storage unit 108. Analyze the overlaps in calculating service level values. The analysis unit 106 summarizes the aggregation units for the service levels of the different aggregation units with the same attribute, defines unit variables, and generates a service level calculation definition for calculating each service level from the unit variables. By using the unit variable definition and the service level calculation definition, the aggregation unit 102 and the service level calculation unit 104, which will be described later, can calculate the service level value while avoiding duplication.

In the present embodiment, the analysis unit 106 defines, as a unit variable, the smallest unit that is the innermost among the service levels defined in the service level definition 110 between service levels in which the aggregation units are in an inclusive relationship. .

The analysis unit 106 outputs the unit variable definition 150 to the counting unit 102. The unit variable is a variable used by the service level calculation unit 104 to calculate a service level value. Furthermore, the unit variable definition 150 is a definition used by the counting unit 102 to calculate unit variables.

The unit variable definition table 150 stores definitions of unit variables that summarize the aggregation units defined by the analysis unit 106. As shown in FIG. 9, the unit variable definition table 150 shows at least one record 158 (FIG. 9 shows one record 158a. In the following, if there is no need to distinguish each record, “158” is shown. Can be included. Each record 158 includes a unit variable name 151, a totaling method 152, a variable 153, a totaling time 154, a totaling space 155, a time unit 156, or a space unit 157.
The configuration of the unit variable definition table 150 in FIG. 9 is an example, and the present invention is not limited to this.

The unit variable name 151 indicates the name of the unit variable. The totaling method 152 corresponds to the totaling method 116 of the service level definition table 110 in FIG. 5, and indicates a variable totaling method used when calculating a service level value. A variable 153 corresponds to the aggregation variable 113 of the service level definition table 110 in FIG. 5 and indicates the type of management data used for calculating the service level value.

The total time 154 corresponds to the total time 114 of the service level definition table 110 in FIG. 5, and indicates a time range in which the variable 153 is totaled and the service level value is calculated. The total space 155 corresponds to the total space 115 of the service level definition table 110 in FIG. 5 and indicates the range of the space where the variables 153 are totaled and the service level value is calculated.

The time unit 156 indicates the timing (time interval) for reading the original data (management data) of the variable 153 used when calculating the service level value. As the time unit 156, a unit equal to or smaller than the total time 154 is selected. For example, the management data collection or recording time unit (5 minutes) may be used as it is as the time unit 156. Alternatively, as the time unit 156, a time unit such as a 10-minute interval in which the management data collection interval is an integer multiple of 5 minutes may be set. These may be set appropriately by providing a setting table separately.

The space unit 157 indicates the range of the target space from which the original data (management data) of the variable 153 used when calculating the service level value is read. As the space unit 157, a unit equal to or smaller than the size of the total space 155 is selected.

Each record 158 includes a specific total time 154 of the variable 153 used for calculating the unit variable, for example, a range in which what time is designated, or a specific range of the total space 155, for example, the node A , B, D and the like. These may be set appropriately by providing a setting table separately. When the analysis unit 106 defines the unit variable definition 150, these setting tables may be referred to.

For example, the unit variable “TAT (1 day, node)” in FIG. 9 is the variable “TAT” recorded in the management data storage unit 160 in units of “10 minutes” and “node”, This means reading in the range of each “node” and calculating the maximum value (MAX) within that range.

The service level calculation definition table 220 is a definition used when the service level calculation unit 104 calculates a service level value. As shown in FIG. 10, the service level calculation definition table 220 shows at least one record 229 (shown in FIG. 10 is three records 229a to 229c. Hereinafter, when there is no need to distinguish each record in particular, "229"). Each record 229 includes an SL name 221, an aggregation method 222, a variable 223, an aggregation time 224, an aggregation space 225, a time unit 226, a space unit 227, or an aggregation determination 228.
Note that the configuration of the service level calculation definition table 220 in FIG. 10 is an example, and the present invention is not limited to this.

The SL name 221 indicates the name of the service level calculated by the service level calculation unit 104 based on the service level calculation definition 220. The aggregation method 222 indicates a method for calculating a service level value. The variable 223 indicates the name of the variable used when calculating the service level value.

The totaling time 224 indicates the width of time during which the variable 223 is totaled and the service level value is calculated. The total space 225 indicates the width of the space where the variables 223 are totaled and the service level value is calculated. The time unit 226 indicates the time width of the variable 223 used when calculating the service level value. The space unit 227 indicates the space width of the variable 223 used when calculating the service level value. Here, as the time unit 226 or the space unit 227, a unit equal to or smaller than the size of the total time 224 or the total space 225 is selected. What unit is selected may be set appropriately by providing a separate setting table.
The aggregation determination 228 indicates information (such as a flag) for identifying whether or not the target service level is a calculation aggregation target.

For example, in FIG. 10, the aggregation determination 228 is defined as “1” by the analysis unit 106 as “1”, and “0” otherwise. For example, “TAT_B” in FIG. 10 includes a variable 223 “TAT (1 day, node)” recorded in units of “1 day” and “node”, and the service level calculation unit 104 sets “January” and It means reading in the range of “service” and calculating “MAX”, that is, the maximum value. A variable 223 “TAT (1 day, node)” corresponds to the unit variable defined in the unit variable definition 150 of FIG.

Also, the aggregation determination of “TAT_B” is “1”, which means that the service level calculation unit 104 reads the variable 223 from the unit variable data storage unit 210. In the case of FIG. 10, it is assumed that the service level calculation unit 104 reads the variable 223 from the management data storage unit 160 for the service level whose aggregation determination 228 is “0”. Note that the record 229 of the service level calculation definition 220 may include the total time 224 of the variable 223 and the specific range of the total space 225 used for calculating the service level value. The specific range may be set as appropriate by providing a setting table separately. When the analysis unit 106 defines the service level calculation definition 220, these setting tables may be referred to.

The aggregation unit 102 in FIG. 3 refers to the unit variable definition table 150, aggregates the management data read from the management data storage unit 160 based on the unit variable definition, and outputs the aggregation value. In the present embodiment, the output total value is stored as unit variable data in the unit variable data storage unit 210.

The unit variable data storage unit 210 records the unit variable data (hereinafter also referred to as “unit variable data 210”) output by the totaling unit 102. The unit variable data 210 is data used by the service level calculation unit 104 to calculate a service level value. As shown in FIG. 11, the unit variable data storage unit 210 has at least one record 217 (FIG. 11 shows three

records

217a, 217b, and 217c. Hereinafter, it is not necessary to distinguish each record in particular. May be labeled "217").

Each record 217 includes a recording date and time 211, a unit variable name 212, a totaling time 213, a totaling space 214, a totaling method 215 or a value 216.
Note that the structure of the unit variable data storage unit 210 in FIG. 11 is an example, and the present invention is not limited to this.

The recording date 211 indicates the date when the unit variable data 210 record was recorded. The unit variable name 212 corresponds to the unit variable name 151 of the unit variable definition 150 in FIG. 9, and indicates a name for identifying the unit variable data 210. The total time 213 indicates a time range in which the values of the unit variable data 210 are totaled. The total space 214 indicates a spatial range in which the values of the unit variable data 210 are totaled. The totaling method 215 corresponds to the totaling method 152 of the unit variable definition 150 in FIG. 9 and shows a method of calculating the value of the unit variable data 210. A value 216 is a value of the calculated unit variable data 210.

For example, FIG. 11 shows “TAT (1 day, 1 day, when calculating the maximum value for“ Node A ”in the time range“ 2011/07/31, 15:01 to 2011/08/08, 15:00 ””. Node) "is" 5.1 sec ". In FIG. 11, only the record 217 of the value of TAT (1 day, node) of the node A is shown as the data of August 1, but the record 217 of the value of another node is also a unit variable data storage unit. Suppose that it is recorded in 210.

Returning to FIG. 3, the service level calculation unit 104 refers to the service level definition table 110 and, based on each service level definition, is recorded in the management data and unit variable data storage unit 210 recorded in the management data storage unit 160. The service level value is calculated using the unit variable data as input. The service level calculation unit 104 can store the calculated service level value in the service level storage unit 170.

The service level storage unit 170 stores the service level value calculated by the service level calculation unit 104. As shown in FIG. 12, the service level storage unit 170 shows at least one service level record 177 (FIG. 12 shows two

records

177a and 177b. In the following, there is no need to distinguish each record in particular. Can be labeled "177"). Each record 177 includes a recording date 171, an SL name 172, an aggregation time 173, an aggregation space 174, an aggregation method 175 or a value 176.
The structure of the service level storage unit 170 in FIG. 12 is an example, and the present invention is not limited to this. For example, the aggregation method 175 is not necessarily required because it can be known by referring to the aggregation method 116 of the SL name 111 in the corresponding service level definition table 110 of FIG.

The recording date 171 indicates the date when each service level record 177 was recorded in the service level storage unit 170. The SL name 172 corresponds to the SL name 111 in FIG. 5 and indicates a name for identifying the service level. The total time 173 indicates a time range in which the service level values are totaled. The total space 174 indicates a space range in which the service level value is calculated. The totaling method 175 corresponds to the totaling method 116 in FIG. 5 and shows a method for calculating a service level value. The value 176 is the calculated service level value.

For example, in FIG. 12, the value of “TAT_A” obtained by calculating the maximum value for “node A” in the time range “2011/07/31 15:01 to 2011/08/08 15:00” is “5”. .1 sec ". Similarly, in the time range of “2011/07/01 15:01 to 2011/08/01 15:00”, the maximum value calculated for “Service A” is “TAT_B” and the value is “6.4 sec”. is there.

11 and 12 indicate the time ranges by the start time and the end time, but are not limited thereto. Since the total time width can be acquired based on the unit variable name 212 or the SL name 172, the

total time

213 and 173 may be only the start time or the end time, or the start time or the end time and the total time width. You may make it have the information of. If the timing of the total time is predetermined in operation, the time is not necessarily required, and only the date may be used.

The service level display unit 208 refers to the service level storage unit 170 and presents the service level value calculated by the service level calculation unit 104 to the user. The service level display unit 208 may be configured by a display device such as a display (not shown) or an application that presents a service level value or an operation state to the user.
The service level display unit 208 may directly receive and present the service level calculated by the service level calculation unit 104 without referring to the service level storage unit 170.

The operation of the service level monitoring system 1 of the present embodiment configured as described above will be described below.
13 to 17 are flowcharts showing an example of the operation of the service level monitoring system 1 according to the embodiment of the present invention. FIG. 13 shows the operation of the analysis unit 106 in this embodiment. 14 and 15 show the details of each processing procedure in FIG. FIG. 16 shows the operation of the counting unit 102 in the present embodiment. FIG. 17 shows the operation of the service level calculation unit 104 in the present embodiment.

First, the operation of the analysis unit 106 will be described with reference to FIG.
First, the analysis unit 106 reads the service level definition with reference to the service level definition table 110 (step S201).
Next, the analysis unit 106 determines, for each service level definition, whether there is a service level that summarizes the aggregation units with the same attribute, that is, whether the calculated service level is an aggregation target (step S203). The analysis unit 106

extracts records

117a and 117b having the same aggregation variable 113 and aggregation method 116 from the service level definition table 110 in FIG. 5 and sets them as aggregation targets (YES in step S203).

Here, only one set of

records

117a and 117b is an aggregation target, but actually, a plurality of sets of records 117 are respectively extracted. Further, only the two

records

117a and 117b are extracted, but two or more records 117 may be extracted as aggregation targets. On the other hand, the analysis unit 106 determines that the record 117c in which the record 117 having the same aggregation variable 113 and the aggregation method 116 does not exist is excluded from the aggregation target (NO in Step S203). Note that the aggregation target means a service level in which a process that overlaps with another service level exists in the calculation process.

For example, in “TAT_A” and “TAT_B” in the service level definition 110 of FIG. 5, the aggregation variable 113 is “TAT” and the aggregation method 116 is “MAX” (YES in step S203). Determines that “TAT_A” and “TAT_B” are aggregation targets, and proceeds to step S205. On the other hand, since a service level having the same aggregation variable 113 and aggregation method 116 as “operation rate_A” does not exist in the service level definition 110 (NO in step S203), the analysis unit 106 selects “operation rate_A” as an aggregation target. It determines with there not being, and progresses to step S207.

Next, the analysis unit 106 outputs a unit variable definition based on the service level definition 110 and the time definition 120, the space definition 130 or the individual definition 140 recorded in the unit conversion definition table storage unit 108, and the unit variable definition table 150 (step S205).
Next, the analysis unit 106 outputs a service level calculation definition based on the service level definition 110 and the unit variable definition 150, and stores the service level calculation definition in the service level calculation definition table 220 (step S207).
In this manner, the service level management apparatus 200 according to the present embodiment analyzes a plurality of service level definitions 110, and unit variable definition 150 and service level calculation definition 220 for service levels that can be aggregated with the same attribute. Can be output.

Next, the procedure of the unit variable definition process S210 of the analysis unit 106 corresponding to step S205 of FIG. 13 will be described in detail with reference to FIG.
First, the analysis unit 106 defines an aggregation unit (unit variable) for the service level determined as the aggregation target in step S203 of FIG. 13 (step S211). The aggregation unit (unit variable) is the aggregation time 114 or the aggregation space 115 of the aggregation variable 113 of the service level definition 110 that can be used in common when calculating values of a plurality of service levels that have overlapping calculation processes. It is a pair. The analysis unit 106 selects the minimum aggregation time 114 and the aggregation space 115 from a plurality of service levels having the same aggregation method 116 among the service levels of the service level definition 110 determined to be an aggregation target, and defines them as unit variables. . The analysis unit 106 may refer to the time definition 120 or the individual definition 140 recorded in the unit conversion definition table storage unit 108 when defining the unit variable.

For example, among the service levels recorded in the service level definition 110 of FIG. 5, the aggregation targets are “TAT_A” and “TAT_B”, and their aggregation times 114 are “1 month” and “1 day”, respectively. Here, referring to the time definition 120, “1 day” is an inner aggregation unit included in “1 month”. Therefore, the analysis unit 106 selects “1 day” as the unit variable based on the time definition 120. Similarly, since the aggregation space 115 to be aggregated is “service” and “node”, the analysis unit 106 selects an inner “node” included in “service” as a unit variable based on the space definition 130. Therefore, the analysis unit 106 defines the unit variable as “1 day” and “node”.

Next, the analysis unit 106 defines the unit variable name 151 of the unit variable definition 150 in FIG. 9 based on the service level definition 110 and the unit variable defined in step S211 (step S213). The unit variable is a value used when the service level calculation unit 104 calculates a service level value.

For example, the aggregation targets in the service level definition 110 are “TAT_A” and “TAT_B”, and the variable is “TAT”. As described above, in this example, based on the result of step S211, the unit variables are “1 day” and “node”. Accordingly, the analysis unit 106 defines the unit variable name 151 as “TAT (1 day, node)”. That is, the unit variable is defined by the aggregation time, the aggregation space, the aggregation method, or a combination thereof, together with the attribute of the service level variable to be aggregated.

Next, the analysis unit 106 refers to the service level definition table 110, compares the SL type 112 and the aggregation method 116 of the service level to be aggregated with the individual definition 140, and finds a service level that matches the individual definition 140. It is determined whether or not it exists (step S215). If there is a matching service level (YES in step S215), step S217 and step S219 are bypassed for that service level, and the process proceeds to step S221. The analysis unit 106 defines the aggregation method 142, variable 143, time unit 144, and space unit 145 recorded in the individual definition 140 as the aggregation method 152, variable 153, time unit 156, and space unit 157 of the unit variable definition table 150 as they are. (Step S221).

For example, the aggregation targets in the service level definition 110 in FIG. 5 are “TAT_A” and “TAT_B”, and there is no item that matches the individual definition 140 for the SL type 112 and the aggregation method 116 (NO in step S215). . Therefore, the information of the individual definition 140 is not used as the unit variable definition 150.

Next, after the determination in step S215, apart from the service level processing that matches the individual definition 140, the service level that does not match (NO in step S215) proceeds to step S217.
Based on the service level definition 110, the analysis unit 106 defines a totaling method 152, a totaling time 154, or a totaling space 155 for calculating unit variables (step S217).
For example, based on the service level definition 110, the analysis unit 106 defines the aggregation method 152 of the unit variable “TAT (1 day, node)” in the unit variable definition 150 as “MAX”. Similarly, the analysis unit 106 defines the total time 154 in the unit variable definition 150 as “1 day” and the total space 155 as “node” based on the service level definition 110 and the unit variables defined in step S211.

Next, the analysis unit 106 defines a variable 153 used when calculating a unit variable based on the service level definition 110. Further, the analysis unit 106 defines a time unit 156 and a space unit 157 for calculating the unit variable based on the unit variable, the time definition 120, and the space definition 130 (step S219). At this time, the analysis unit 106 selects a unit of the total time 154 and the total space 155 or less as the time unit 156 and the space unit 157.

For example, the analysis unit 106 first defines the variable 153 of the unit variable “TAT (1 day, node)” as “TAT” based on the service level definition 110. Next, since the total time 154 of the unit variable “TAT (1 day, node)” is “1 day”, the analysis unit 106 uses the time definition 120 as the time unit 156 for calculating the unit variable. A smaller unit, and “10 minutes”, which is an integer multiple (2 times) of the management data collection timing, is defined. Similarly, since the total space 155 of the unit variable “TAT (1 day, node)” is “node”, the analysis unit 106 uses the space definition 130 as a space unit 157 for calculating the unit variable based on the space definition 130. Define a “node” which is the same unit.

Next, the analysis unit 106 outputs a unit variable definition based on the results of steps S211 to S219 and stores the unit variable definition in the unit variable definition table 150 (step S221).
In this way, the service level management apparatus 200 of the present embodiment can generate the unit variable definition table 150.

Next, the procedure of the service level calculation definition process S230 of the analysis unit 106 corresponding to step S207 of FIG. 13 will be described in detail with reference to FIG.
First, the analysis unit 106 classifies the service level of the service level definition 110 according to whether or not the service level is a calculation aggregation target based on the result of step S203 of FIG. 13 (step S231). The analysis unit 106 generates

records

229a, 229b, and 229c corresponding to the

records

117a, 117b, and 117c in the service level definition table 110 in FIG. 5 in the service level calculation definition table 220 in FIG. That is, the SL name 111 of each record 117 of the service level definition 110 is defined as the SL name 221 of each record 229 of the service level calculation definition 220.

Then, the analysis unit 106 sets the aggregation determination 228 of the service level calculation definition 220 of the service level classified as the aggregation target in step S231 to “1”, and sets the aggregation determination 228 of the service level classified as the aggregation target to “0”. Are defined respectively. The aggregation determination 228 is an index used to determine which of the unit variable data storage unit 210 and the management data storage unit 160 is to read variables when the service level calculation unit 104 calculates a service level value. . That is, the aggregation determination 228 indicates whether to use the aggregate value of the management data aggregated by the aggregation unit 102 or the management data itself.

For example, from the result of step S201 in FIG. 13 based on the service level definition 110, the analysis unit 106 sets the aggregation determination 228 of “TAT_A” of the record 229a and “TAT_B” of the record 229b to “1”, respectively. It is defined as Similarly, the analysis unit 106 defines “0” as the aggregation determination 228 of “operation rate_A” of the record 229c of the service level calculation definition 220.

Next, the analysis unit 106 compares the SL type 112 and the aggregation method 116 of the service level definition 110 with the SL type 141 and the aggregation method 142 of the individual definition 140, and compares the SL type 112 and the aggregation method 116 of the service level definition 110 with each other. Then, it is determined whether or not the service level record 117 in which the SL type 141 and the tabulation method 142 of the individual definition 140 coincide with each other exists in the service level definition 110 (step S233). If there is a matching service level (YES in step S233), step S235 and step S237 are bypassed, and the process proceeds to step S239. The analysis unit 106 defines the aggregation method 142, variable 143, time unit 144, and space unit 145 of the individual definition 140 as the aggregation method 222, variable 223, time unit 226, and space unit 227 of the service level calculation definition 220, respectively ( Step S239).

For example, the SL type 112 “operating rate” defined in “operating rate_A” of the service level definition 110 and the SL type 141 “operating rate” recorded in the individual definition 140 match (YES in step S233). The unit 106 stores the aggregation method 142 “(m−Σni) / m”, the variable 143 “unscheduled stop time”, the time unit 144 “one day”, and the space unit 145 “service” recorded in the individual definition 140. The “operation rate_A” totaling method 222, the variable 223, the time unit 226, and the space unit 227 of the record 229c of the service level calculation definition 220 are defined (step S239).

Next, after the determination in step S233, apart from the service level processing that matches the individual definition 140, for the service level that does not match the individual definition 140 (NO in step S233), the process proceeds to step S235, and the analysis unit 106: Perform the process.
The analysis unit 106 defines a variable 223 used for calculation for a service level that does not have an item that matches the individual definition 140 (step S235). First, based on the result of step S231, the analysis unit 106 defines the unit variable defined in step S205 as the variable 223 used for calculation of the service level value of which the aggregation determination 228 is “1”. Next, based on the service level definition 110, the analysis unit 106 defines a variable 223 that is used to calculate a service level value for which the aggregation determination 228 is “0”. Further, the analysis unit 106 defines a total time 224 and a total space 225 based on the service level definition 110 for a service level that does not have an item that matches the individual definition 140.

For example, based on the unit variable definition 150 defined in step S205, the analysis unit 106 defines the unit variable “TAT (1 day, node)” as the variable 223 for calculating “TAT_A” and “TAT_B”. Further, the analysis unit 106 uses the total time 114 and the total space 115 of the record 117a based on the service level definition 110 as the total time 224 “1 day” and the total space 225 “of the TAT_A” of the record 229a of the service level calculation definition 220. Define as node. Similarly, the analysis unit 106 defines “TAT_B” of the record 229 b of the service level calculation definition 220 based on the record 117 b of the service level definition 110.

Next, based on the service level definition 110, the analysis unit 106 defines a tabulation method 222 used when calculating a service level value (step S237). Further, the analysis unit 106 defines a time unit 226 and a space unit 227 for calculating a service level value based on the service level definition 110, the unit variable definition 150, the time definition 120, or the space definition 130. At this time, the analysis unit 106 selects, as the time unit 226 and the space unit 227, units less than the size of the total time 114 and the total space 115 recorded in the service level definition 110.

For example, the variable 223 for calculating “TAT_A” in step S235 is defined as a unit variable “TAT (1 day, node)”. According to the service level definition 110, the service level “TAT_A” of the record 117a of the service level definition 110 and the unit variable “TAT (1 day, node)” of the variable 223 of the record 229a of the service level calculation definition 220 (FIG. 9). (Corresponding to the record 158a of the unit variable definition 150) is the same. Therefore, when calculating the service level “TAT_A”, the analysis unit 106 uses the unit variable data of “TAT (1 day, node)” that has already been calculated by the counting unit 102 based on the unit variable definition 150 as a unit. It is only necessary to read from the variable data storage unit 210. As described above, the analysis unit 106 compares the service level definition 110 and the variable 223 and, when determining that they are the same, defines “Read” as the aggregation method 222 of the service level “TAT_A” of the record 229a. The aggregation method 222 “Read” means that the analysis unit 106 reads data from the unit variable data storage unit 210.

Next, the analysis unit 106 defines a counting method 222 of “TAT_B”. In step S235, the variable 223 for calculating “TAT_B” is defined as “TAT (1 day, node)”. According to the service level definition 110, the service level “TAT_B” of the service level definition 110 and the unit variable “TAT (1 day, node)” of the variable 223 are not the same. Therefore, based on the service level definition 110, the analysis unit 106 defines “MAX” of the aggregation method 116 of the record 117b as the aggregation method 222 of “TAT_B” of the record 229b of the service level calculation definition 220.

Next, the total time 224 of “TAT_A” and the total space 225 of the record 229a of the service level calculation definition 220 are the total time 154 of the record 158a of the unit variable definition 150 corresponding to the variable 223 “TAT (1 day, node)”. Since it is the same as the total space 155, based on the unit variable “1 day” and “node” defined in step S211, the analysis unit 106 sets the time unit 226 to “1 day” and the space unit 227 to “ Each node is defined. Also, “TAT_B” is defined similarly.

Next, the analysis unit 106 outputs a service level calculation definition that is a definition for calculating a service level value based on the results of steps S231 to S237, and stores the service level calculation definition in the service level calculation definition table 220 (step S239). ).
In the present embodiment, as a result of the above operation, the service level calculation definition 220 shown in FIG. 10 is obtained.

As described above, the analysis unit 106 can analyze the duplication of the calculation process of the service level definition 110 and output the service level calculation definition 220 and the unit variable definition 150.

Next, the operation of the counting unit 102 will be described in detail. FIG. 16 is a flowchart showing the operation of the counting unit 102.
First, the totaling unit 102 reads the unit variable definition 150 (step S241).
Next, the totaling unit 102 determines the range of the variable 153 used for calculating the value of the unit variable based on the unit variable definition 150 (step S243).

For example, according to the unit variable definition 150, the variable 153 for calculating “TAT (1 day, node)” is “TAT”. Similarly, the time unit 156 is “10 minutes”, and the space unit 157 is “node”. Therefore, the totaling unit 102 determines that “TAT” recorded in units of “10 minutes” and “node” is a variable for calculating the value of the unit variable. Since the total time 154 is “1 day”, the total unit 102 defines the management data “1 day” recorded in units of “10 minutes” as the range of the variable 153.

Note that the counting unit 102 may refer to the time definition 120 and the space definition 130 in order to determine the range and number of variables used to calculate the value of the unit variable.

Next, the totaling unit 102 reads the variable 153 used for calculating the value of the unit variable from the management data storage unit 160 based on the unit variable definition 150 (step S245).
For example, in order to calculate the value of “TAT (1 day, node)” of the record 158a of the unit variable definition 150, the counting unit 102 calculates “10 minutes” and “node” units from the calculation time point to one day before. Is read from the management data storage unit 160.

Next, the totaling unit 102 calculates the value of the unit variable based on the read variable 153 and the totaling method 152 (step S247).
In this example, since the aggregation method 152 of “TAT (1 day, node)” is “MAX” based on the unit variable definition 150, the aggregation unit 102 selects the TAT value from the range read in step S 245. Select the maximum value.

Next, the totaling unit 102 outputs the unit variable value 216 calculated in step S247 to the unit variable data storage unit 210 (step S249). At this time, the totaling unit 102 adds the recording date and time 211, the unit variable name 212, the totaling time 213, the totaling space 214, and the totaling method 215 to the unit variable data record 217 of FIG. Record.

For example, in the example of the unit variable data 210 shown in FIG. 11, as shown in the record 217a, it is recorded as “Node A” between “2011/07/31 15:01 to 2011/08/08 15:00”. This means that the maximum value of the management data “TAT” was “5.1 sec”. The totaling unit 102 outputs the totaling time 213 in the unit variable data based on the totaling time 154 recorded in the unit variable definition 150 in FIG. 9 and the recording date and time 211 when the unit variable data in FIG. 11 is calculated. In the example of FIG. 11, the totaling time 154 of “TAT (1 day, node)” is “1 day” based on the unit variable definition 150 of FIG. Let the day be the total time 213 in the unit variable data 210. The aggregation method 215 is “MAX” of the aggregation method 152 of the unit variable definition 150.

As described above, the counting unit 102 can calculate the unit variables based on the unit variable definition 150 and the management data.

Next, the operation of the service level calculation unit 104 will be described. FIG. 17 is a flowchart showing the operation of the service level calculation unit 104.

First, the service level calculation unit 104 reads each record 229 from the service level calculation definition 220 (step S251).
Next, the service level calculation unit 104 determines a variable 223 to be used for calculating a service level value based on the record 229. Next, the service level calculation unit 104 sets the range of the variable 223 used for calculating the service level value based on the total time 224, the total space 225, the time unit 226, and the space unit 227 recorded in the service level calculation definition 220. Determination is made (step S253).

For example, according to the service level calculation definition 220 of FIG. 10, the variable 223 for calculating “TAT_B” of the record 229b is “TAT (1 day, node)”. Similarly, the time unit 226 of the record 229b is “1 day”, and the space unit 227 is “node”. Therefore, the service level calculation unit 104 determines “TAT” recorded in units of “1 day” and “node” as a variable for calculating the value of “TAT_B”. Further, since the total time 224 is “January” and the total space 225 is “service”, the service level calculation unit 104 forms “service” for “January” recorded in units of “one day”. The value of “node” is set as the range of the variable 223.

Note that the service level calculation unit 104 may refer to the time definition 120 and the space definition 130 in order to determine the range and the number of variables used for calculating the service level value.

Next, the service level calculation unit 104 reads the variable 223 corresponding to the range defined in step S243 from either the unit variable data storage unit 210 or the management data storage unit 160. Here, the service level calculation unit 104 determines from which storage unit the variable 223 is read based on the aggregation determination 228 of the service level calculation definition 220. If the aggregation determination 228 is “1”, the service level calculation unit 104 reads the variable 223 from the unit variable data storage unit 210. On the other hand, if the aggregation determination 228 is “0”, the service level calculation unit 104 reads the variable 223 from the management data storage unit 160 (step S255).

For example, since the aggregation determination 228 of “TAT_A” in the record 229a and “TAT_B” in the record 229b in FIG. 10 are both “1”, the service level calculation unit 104 stores the variable 223 defined in step S253 as unit variable data storage. Read from the unit 210. On the other hand, since the aggregation determination 228 of “operation rate_A” of the record 229c is “0”, the service level calculation unit 104 reads the variable 223 defined in step S253 from the management data storage unit 160.

Next, the service level calculation unit 104 calculates a service level value based on the aggregation method 222 of the service level calculation definition 220 using the variables read in step S245. The service level calculation unit 104 outputs the calculated service level value to the service level display unit 208 (step S257).

For example, since the aggregation method of “TAT_B” is “MAX” based on the service level calculation definition 220, the service level calculation unit 104 selects the maximum value from the values read in step S255. Note that the service level value output by the service level calculation unit 104 may be stored in the service level storage unit 170 of FIG. 12 as an example.
As described above, the service level calculation unit 104 can calculate a service level value.

In the present embodiment, the service level monitoring system 1 may further include a management data collection unit 240 and a management data collection definition table 230, as shown in FIG.
The management data collection unit 240 first collects management data related to a monitoring target system (not shown) based on the management data collection definition table 230. Next, the management data collection unit 240 outputs the management data of each item to the management data storage unit 160 as a management data log based on the management data collection definition table 230. The monitoring target system can be, for example, a server alone, a WEB three-tier system composed of a plurality of servers, a server group stored in one rack, the entire data center, or all servers constituting a service, Or all servers used by a certain user, a server group used by a part of a certain user group, or a server of a plurality of data centers as a whole.

The management data collected by the management data collection unit 240 includes information on availability of the monitored system such as unscheduled stop time, response time (TAT), CPU usage rate, memory usage rate, disk usage rate, batch processing completion rate, Such as performance information and power consumption.

The management data collection definition table 230 describes definitions related to management data collected by the management data collection unit 240. As shown in FIG. 19, the management data collection definition table 230 shows at least one record 234 (FIG. 19 shows three

records

234a, 234b, and 234c. Hereinafter, there is no need to distinguish each record in particular. Is configured to include “234”. Each record 234 includes a management data name 231, a collection method 232, or a recording interval 233.
The configuration of the management data collection definition table 230 in FIG. 19 is an example, and the present invention is not limited to this.

Management data name 231 indicates a name for identifying management data. The collection method 232 indicates a method for collecting management data. The recording interval 233 indicates an interval at which collected management data is recorded in the management data storage unit 160.

For example, the record in FIG. 19 and “TAT” in 234a mean that management data “TAT” “measured by middleware” every “5 minutes” is output to the management data storage unit 160 every 5 minutes. Similarly, the “operation status” of the record 234 c means that “response confirmation by Ping” is performed “every 5 minutes” and the result is output to the management data storage unit 160.

The management data collected by the management data collection unit 240 is recorded as a log in the management data storage unit 160.
Note that the service level calculation unit 104 described above may directly receive management data as an input from the management data collection unit 240 instead of the management data stored in the management data storage unit 160, and calculate a service level value. .

The operation of the management data collection unit 240 will be described below. FIG. 20 is a flowchart showing the operation of the management data collection unit 240 in the present embodiment.
First, the management data collection unit 240 reads the management data collection definition table 230 (step S261).
Next, the management data collection unit 240 collects management data of the target system (data center 5) according to the collection method 232 and the recording interval 233 defined for each record 234 of the management data collection definition table 230 (step S263). .

For example, the management data collection unit 240 collects “TAT” measured by the middleware of the target system every 5 minutes based on the management data collection definition table 230. Alternatively, the management data collection unit 240 transmits a Ping to the target system based on the management data collection definition table 230, checks whether there is a response every 5 minutes, and collects “operation status”.

Next, the management data collection unit 240 sequentially outputs the values of each management data as records 165 to the management data storage unit 160 (FIG. 4) and stores them (step S265).

For example, in the example of the management data storage unit 160 shown in FIG. 4, in the record 165a, “TAT” measured by “Service A” at “2011/08/08 15:00” is “10 msec”. Show. Similarly, the record 165c indicates that the “operation status” of “node A” of “2011/08/01 15:00” is “response received”. This means that there was a response to Ping transmitted by the management data collection unit 240. Further, the record 165b indicates that the “unscheduled stop time” of “Service A” of “2011/08/01 15:00” was “2 minutes / 1 hour”. This means that there was an unscheduled stop time of 2 minutes in the past hour from the recording date.

As described above, the management data collection unit 240 can collect management data of the monitoring target system and record it in the management data storage unit 160 as a management data log.

As described above, according to the service level monitoring system 1 according to the embodiment of the present invention, it is possible to automatically detect overlapping processes when calculating a plurality of service levels.
The reason is that the analysis unit 106 detects time and spatial overlap portions of the calculation process existing in the service level definition 110 based on the definition related to overlap (time definition 120, space definition 130, or individual definition 140). Because.

Further, as described in the above document, there are the following three problems when calculating a plurality of service levels for the same monitoring target.
The first problem is that there is an overlap in the service level calculation process, such as reading out the recorded history information in duplicate.

The service evaluation system of Patent Document 1 calculates a service level value according to a command input by a user. For example, when calculating a service level value for one month, the system reads all history information for one month. Further, when calculating the value of the daily service level, the system reads the history information for one day. Therefore, when calculating the two types of service levels, the system reads the history information for one month redundantly.

Therefore, the service evaluation system disclosed in Patent Document 1 has a problem that history information is read redundantly when calculating a plurality of service level values exemplified above for a specific period.

This also applies to the service evaluation system disclosed in Patent Document 2. The system of Patent Document 2 calculates a service level value according to SLA definition information given in advance, instead of inputting a command. When calculating values of a plurality of service levels, a plurality of definitions are given, and the duplication is not considered.

The second problem is that pre-aggregated data is required to avoid duplication.
In Patent Document 3, in order to solve the first problem, the system includes a plurality of tabulation tables in which database data is tabulated in advance for a predetermined item, and a control table having information on the tabulation table items and the number of data items. Prepare. When the system calculates the aggregate value that meets the requested conditions, the aggregate that minimizes the number of aggregate data among the aggregate data that can be aggregated that meets the aggregation conditions based on the control table. Configured to search the table.

However, in recent years, since the services to be monitored have become large-scale, the size of history information has increased accordingly, and it is difficult to aggregate and prepare individual aggregation tables in advance.
Therefore, the data processing system of Patent Document 3 has a problem that it requires aggregated data that has been previously aggregated from the original data in order to improve data processing efficiency.

The third problem is that a person judges and avoids duplication of service level calculation processes.
Since the service evaluation system of Patent Document 1 calculates a service level in accordance with a user's instruction, when the calculation process overlaps is obvious, it is possible to avoid the overlap by setting a certain rule. It is. However, the user has to judge where each point overlaps and set a rule. That is, the system of Patent Document 1 has a problem that a person has to determine a condition and set a rule for calculation in order to avoid duplication during service level calculation.

For example, in the service evaluation systems of Patent Document 1 and Patent Document 2, consider the case where the service availability for each month of a certain service is calculated for half a year. In this case, both systems first read the history information every day and calculate the service operating rate for one day. Next, the system reads out history information for one month every month, and calculates a service operation rate for one month. This is continued for half a year. At this time, in the service evaluation systems of Patent Document 1 and Patent Document 2, all the half-year history information must be read at least twice. Furthermore, in order to avoid the above-described duplication in the service evaluation system of Patent Document 1, it is necessary to specify the duplication process and set a rule for calculation.

According to the service level monitoring system 1 of the present invention, the above problem can be solved.
That is, according to the service level monitoring system 1 according to the embodiment of the present invention, it is possible to efficiently calculate a plurality of service levels by reducing the number of times and the amount of data read from the storage device.
The reason is that it is configured to include the following elements. First, the analysis unit 106 outputs a service level calculation definition 220 and a unit variable definition 150 in order to calculate a service level value while avoiding an overlapping portion included in the service level definition 110. The totaling unit 102 calculates unit variables based on the unit variable definition 150 and outputs them to the unit variable data storage unit 210. Next, the service level calculation unit 104 calculates a service level value based on the service level calculation definition 220. At that time, the service level calculation unit 104 reads necessary variables from the management data storage unit 160 and the unit variable data storage unit 210. Note that the storage unit from which the service level calculation unit 104 reads the variable 223 is defined as the aggregation determination 228 in the service level calculation definition 220. With the above operation, the service level monitoring system 1 can reduce the number and the number of times the same data is repeatedly read from the management data storage unit 160.

Furthermore, according to the service level monitoring system 1 according to the embodiment of the present invention, it is possible to increase the efficiency of data processing without preparing aggregate data in advance.
The reason is that it is configured to include the following elements. When the analysis unit 106 calculates values of a plurality of service levels based on the service level definition 110, the overlapping part is automatically detected, and a definition for calculating the overlapping part is output as a unit variable definition 150. . Next, the totaling unit 102 calculates a unit variable based on the management data and the unit variable definition 150 and outputs the unit variable to the unit variable data storage unit 210. The service level calculation unit 104 reads the variable 223 from the unit variable data storage unit 210 and the management data storage unit 160 according to the service level calculation definition 220, and calculates a service level value.

(Third embodiment)
FIG. 21 is a functional block diagram showing the configuration of the service level monitoring system 1 according to the embodiment of the present invention.
The service level monitoring system 1 of this embodiment is different from the above-described embodiment in that the service level calculation method is defined as the highest aggregation unit (unit variable) when the service level calculation methods are in a vertical relationship with each other. Here, the hierarchical relationship is defined by the order of a plurality of service level calculation processing procedures, the priority order, or the dependency relationship. The procedure to be processed earlier is set as the upper level, and the procedure for performing the calculation process after the upper level procedure or using the value calculated in the higher level procedure is set as the lower level.

As shown in FIG. 21, the service level management apparatus 300 according to the present embodiment is similar to the service level management apparatus 200 according to the above-described embodiment of FIG. 3 and includes a totaling unit 102, a service level calculation unit 104, and a service level display unit 208. And an analysis unit 306 and a unit conversion definition table 308.
3 is the same as the service level definition table 110, the unit variable definition table 150, the service level storage unit 170, the unit variable data storage unit 210, and the service level calculation definition table 220 of the service level management apparatus 200 of the above embodiment of FIG. A service level definition table 310, a unit variable definition table 330, a service level storage unit 370, a unit variable data storage unit 350, and a service level calculation definition table 340 each having a structure are provided.

As shown in FIG. 22, the service level definition table 310 of this embodiment has the same configuration as the service level definition table 110 of FIG. 5 of the above embodiment. FIG. 22 shows an example in which records 317 a to 317 d different from the service level definition table 110 of FIG. 5 are stored in the service level definition table 310.

In this embodiment, the unit conversion definition table 308 of FIG. 21 includes the time definition table 120 of FIG. 6 of the above embodiment, the space definition table 130 of FIG. 7, and the individual definition table 140 of FIG. It has the method definition table 320 of FIG.

The method definition table 320 (hereinafter also referred to as “method definition 320”) defines a relationship such as an inclusion relationship, dependency relationship, master-slave relationship, or priority order of the service level aggregation method 116 recorded in the service level definition table 310. Is. The method definition table 320 can also include definitions of common parts of the aggregation method. As shown in FIG. 23, the method definition table 320 shows at least one record 325 (FIG. 23 shows five records 325a to 325e. Hereinafter, if it is not necessary to distinguish each record, “325” is shown. Can be included). Each record 325 includes an aggregation method name 321, an aggregation type 322, an upper technique 323, or a lower technique 324.
The configuration of the method definition table 320 in FIG. 23 is an example, and the present invention is not limited to this.

The aggregation method name 321 indicates the name of the method for calculating the service level value. The aggregation method name 321 is an item corresponding to the aggregation method 116 recorded in the service level definition 310.

“Aggregation type 322” indicates a classification of an operation performed by the corresponding aggregation method. FIG. 23 shows two types of “four arithmetic operations” and “comparison” as examples. The aggregation type 322 is used by the analysis unit 306 in order to consider aggregation of aggregation methods regarding aggregation methods belonging to the same aggregation type.

The high-order method 323 indicates a totaling method that enables different totals using the totaling result. FIG. 23 illustrates “total” as the upper method 323 of “average”, and “reorder” as the upper method 323 of “percentile” and “MAX”.
The subordinate method 324 indicates a totaling method that can be totaled using the results totaled by the corresponding totaling method. In the upper method 323 and the lower method 324, the related aggregation method names are recorded corresponding to each other's vertical relationship. FIG. 23 shows “average” as the sub-method 324 of “total”, and “MAX”, “MIN”, and “percentile” as the sub-method 324 of “sort”. In the upper method 323 or the lower method 324 of the method definition table 320, “−” means that there is no corresponding method.

In the present embodiment, only “four arithmetic operations” and “comparison” are given as examples of aggregation types of the method definition 320, but other calculation formulas recorded in the individual definition 140 are also included. be able to. Those that can form a tree structure based on dependency relationships based on input / output values, such as the upper method 323 or the lower method 324, may be newly defined as “aggregation type 322”. Further, even in the case of individual calculation formulas recorded in the individual definition 140, the upper method 323 or the lower method 324 may be extracted by parsing the calculation formula so as to be divided into operation symbols and character strings. Such upper method 323 or lower method 324 may be recorded as a method definition 320 in association with a calculation formula.

The analysis unit 306 refers to the service level definition table 310 and calculates a plurality of service levels based on the time definition 120, the space definition 130, the individual definition 140, or the method definition 320 recorded in the unit conversion definition table 308. The unit variable definition and the service level calculation definition, which are rules for calculating the service level value while avoiding the duplication, are output.

In the present embodiment, the analysis unit 306 defines, as a unit variable, the highest unit among the service levels defined in the service level definition 110 between service levels in which the aggregation units are vertically related to each other. The analysis unit 306 uses the method definition 320 to determine the hierarchical relationship of the service level aggregation units.

As shown in FIGS. 25 and 26, the unit variable definition table 330 and the service level calculation definition table 340 have the same configurations as the unit variable definition table 150 in FIG. 9 and the service level calculation definition table 220 in FIG. Have FIG. 25 shows an example in which a record 338 a different from the unit variable definition table 150 of FIG. 9 is recorded in the unit variable definition table 330. FIG. 26 shows an example in which records 349a to 349d different from the service level calculation definition table 220 of FIG. 10 are recorded in the service level calculation definition table 340. Hereinafter, the service level definition table 310 is also referred to as “service level definition 310”, the unit variable definition table 330 is also referred to as “unit variable definition 330”, and the service level calculation definition table 340 is also referred to as “service level calculation definition 340”.

The operation of the service level monitoring system 1 of the present embodiment configured as described above will be described below.
FIG. 24 is a flowchart showing an example of the operation of the service level management apparatus 300 of the service level monitoring system 1 of the present embodiment. In this embodiment, the operation of the management data collection unit 240 (FIG. 18) described in the above embodiment is the same. Hereinafter, the operation of the analysis unit 306 will be described.

First, the analysis unit 306 reads the service level definition 310 of FIG. 22 (step S301). The following processing is repeated for each service level definition defined in the service level definition table 310.
Next, the analysis unit 306 refers to the method definition 320 stored in the unit conversion definition table 308 for the aggregation method 116 recorded in the read record 317, and the aggregation type 322 of the aggregation method 116 of the target service level. Is “comparison” (step S303).

For example, in the service level definition 310 of FIG. 22, the aggregation method 116 of “TAT_C” of the record 317d is “average”. In accordance with the method definition 320 of FIG. 23, the analysis unit 306 determines that the aggregation type 322 of the aggregation method name 321 “average” is “four arithmetic operations” (NO in step S303). On the other hand, the analysis unit 306 calculates the totaling method 116 of “TAT_A” in the record 317a of the service level definition 310 and “TAT_B” in the record 317b by “MAX” and “98th percentile”. It is determined that the type 322 is “comparison” (YES in step S303).

When the analysis unit 306 determines in step S303 that the aggregation type 322 of the aggregation method 116 of the record 317 is other than “comparison” (NO in step S303), the analysis unit 306 has an aggregation type other than “comparison”. The service level determined to be is processed in the same manner as the unit variable definition process (step S210) of FIG. And the analysis part 306 progresses to step S230 after a process.

On the other hand, when the analysis unit 306 determines in step S303 that the aggregation type 322 of the aggregation method 116 of the record 317 is “comparison” (YES in step S303), the analysis unit 306 first determines the service level definition 310. , It is determined whether there is a service level in which the total time 114 and the total space 115 are the same and the total method 116 is different among the target service levels (step S305).
If there is no service level in which the target service level is the same as the total service time 114 and the total space 115 and the total method 116 is different (NO in step S305), the analysis unit 306 does not set the target service level as an aggregation target. Then, the analysis unit 306 proceeds to step S230.
On the other hand, when there is a service level in which the target service level is the same as the total time 114 and the total space 115 and the total method 116 is different (YES in step S305), the analysis unit 306 sets the target service level as an aggregation target, The process proceeds to step S307.

For example, in the service level definition 310 of FIG. 22, the total time 114 and the total space 115 of “TAT_A” of the record 317a and “TAT_B” of the record 317b are the same, but the total method 116 is different. Therefore, the analysis unit 306 determines the two types of service levels “TAT_A” and “TAT_B” as aggregation targets. If at least one of the total time 114 and the total space 115 is different, the analysis unit 306 determines that none of the service levels is an aggregation target based on the service level definition 310.

Then, based on the service level definition 310, the analysis unit 306 sets the total time 114 and the total space 115 as a total unit for the service level having the same total time 114 and total space 115 (step S307).
For example, in the example of the service level definition 310 in FIG. 22, the analysis unit 306 defines the total time 114 “January” and the total space 115 “service” as a total unit for calculating “TAT_A” and “TAT_B”. To do.

Next, the analysis unit 306 defines the name of the unit variable used for calculating the service level value determined as the aggregation target (step S309).
For example, as shown in FIG. 25, the analysis unit 306 defines the unit variable name 151 used for calculating the values of “TAT_A” and “TAT_B” in the unit variable definition 330 as “reference SL_1”.

Next, the analysis unit 306 refers to the method definition 320 in FIG. 23 based on the service level aggregation method 116 recorded in the service level definition 310 in FIG. The corresponding method is defined as the aggregation method 152 of the unit variable definition 330 (step S311).
For example, the aggregation methods 116 of “TAT_A” and “TAT_B” determined to be aggregation targets in step S305 are “MAX” and “98 percentile”, respectively. Based on the method definition 320, the high-order method 323 of these aggregation methods 116 is “reordering”. Therefore, the analysis unit 306 defines “reordering” as the aggregation method 152 of the unit variable “reference SL_1” defined in step S309.

Next, the analysis unit 306 defines a variable 153 used for aggregation of unit variables based on the service level definition 310 and the method definition 320. Further, the analysis unit 306 defines the total time 154, the total space 155, the time unit 156, and the space unit 157 based on the time definition 120, the space definition 130, the method definition 320, or the service level definition 310 (step S313).

For example, the analysis unit 306 defines a unit variable “TAT” as the variable 153 for calculating “reference SL_1”. The analysis unit 306 also defines “January” and “Service” as the total time 154 and the total space 155 for calculating “reference SL_1” based on the service level definition 310. Furthermore, the analysis unit 306 defines “10 minutes” and “service” as the time unit 156 and the space unit 157 based on the time definition 120 and the space definition 130, respectively.

Then, the analysis unit 306 outputs a unit variable definition based on the results of Steps S307 to S313, and creates the unit variable definition table 330 shown in FIG. 25 (Step S315).
In this way, the analysis unit 306 generates the unit variable definition 330.

The operation after step S315 or step S210 is performed by the analysis unit 306 based on the service level definition 310 and the unit variable definition 330 as in the service level calculation definition process (step S230) of FIG. The service level calculation definition is output and stored in the service level calculation definition table 340 shown in FIG.

24, since “TAT_A” and “TAT_B” are aggregation targets in the service level definition 310, these variables 223 are “reference SL_1” defined in the above processing. Similarly, the aggregation determination 228 for “TAT_A” and “TAT_B” is “1”. On the other hand, the other items are the same as those in the service level definition 310 regardless of whether or not they are aggregation targets.

As described above, the analysis unit 306 can analyze duplication of calculation processes in the service level definition 310 and output the unit variable definition 330 and the service level calculation definition 340.

Next, the operation of the counting unit 102 will be described.
The tabulation unit 102 of this embodiment operates in the same manner as the procedure described with reference to FIG. 16 in the service level management apparatus 200 of the above embodiment.
The aggregation unit 102 calculates unit variables based on the unit variable definition 330 and outputs the unit variables to the unit variable data storage unit 350 as in the above embodiment.

For example, the case where the counting unit 102 calculates “reference SL_1” based on the unit variable definition 330 of FIG. 25 will be described. According to the record 338a in FIG. 25, the counting unit 102 first calculates the value of “TAT” recorded in units of “10 minutes” and “service” from the management data storage unit 160 one month before the measurement time point. Read to value. Next, the totaling unit 102 rearranges the read management data in ascending order according to the totalization method “rearrangement” recorded in the unit variable definition 330. Finally, the totaling unit 102 outputs the rearranged management data to the unit variable data storage unit 350 as “reference SL_1”.

The above is the operation of the counting unit 102 that calculates unit variables based on the unit variable definition 330 and the management data storage unit 160 and outputs the unit variables to the unit variable data storage unit 350.

Next, the operation of the service level calculation unit 104 of this embodiment will be described.
The service level calculation unit 104 of this embodiment operates in the same manner as the procedure described with reference to FIG. 17 in the service level management apparatus 200 of the above embodiment.

For example, a case where the service level calculation unit 104 calculates a service level value according to the service level calculation definition table 340 shown in FIG. 26 will be described.
When calculating “TAT_A” of the record 349a, since the aggregation determination 228 is “1”, the service level calculation unit 104 reads “reference SL_1” from the unit variable data storage unit 350. Note that the “reference SL_1” to be read is a value that is totaled in units of “January, service” based on the total time 224, the total space 225, the time unit 226, and the space unit 227. The service level calculation unit 104 calculates the maximum value (MAX) based on the aggregation method 222 from the set of management data values read as “reference SL_1”.
In the case of “TAT_B” in the record 349b, the service level calculation unit 104 calculates the 98th percentile value from the set of values.

On the other hand, since the aggregation determination 228 of “operation rate_A” and “TAT_C” is “0”, the service level calculation unit 104 reads the variable used for calculating the service level value from the management data storage unit 160. Note that the range of the variable 223 read by the service level calculation unit 104 based on the service level calculation definition 340 is “unscheduled stop” collected in “one day” and “service” for “operation rate_A”. “Time” is “January”. Similarly, for “TAT_C”, “10 minutes” and “TAT” collected in “node” units are “1 day” and “service”. Next, the service level calculation unit 104 applies the tabulation method 222 recorded in the service level calculation definition 340 to the read variable 223.

As described above, the service level calculation unit 104 can calculate a service level value based on the service level calculation definition 340, unit variable data, and management data.

As described above, according to the service level monitoring system 1 according to the embodiment of the present invention, the effects of the above embodiment can be obtained, and the aggregation type can be changed to “sort” when calculating a plurality of service levels. It is possible to summarize the calculation process of the classified aggregation method.
The reason is that it is configured to include the following elements. First, in addition to the above embodiment, the method definition 320 is added as a duplicate definition. With reference to the method definition 320, the analysis unit 306 defines the calculation method for calculating the service level so that the service level is calculated while avoiding the overlapping portion as in the calculation time and the calculation space in the above embodiment. The unit variable definition 330 and the service level calculation definition 340 are output.

Furthermore, according to the service level monitoring system 1 according to the embodiment of the present invention, when a plurality of service levels using a totaling method in which the totaling type is classified as “sorting” is totalized, data reading is made efficient. it can.
The reason is that it operates as follows. First, based on the unit variable definition 330, the totaling unit 102 calculates a variable necessary for calculating the service level as a unit variable and records it in the unit variable data storage unit 350. Next, the service level calculation unit 104 reads a unit variable recorded in the unit variable data storage unit 350 based on the service level calculation definition 340 and calculates a service level value.

(Fourth embodiment)
FIG. 27 is a functional block diagram showing the configuration of the service level monitoring system 1 according to the embodiment of the present invention.
The service level management apparatus 400 of the service level monitoring system 1 according to the present embodiment generates a definition that combines the aggregation units with different service level definitions, with the unit variable defined in the above embodiment as a unit. The difference is that the duplication of the service level calculation process is further reduced.

As shown in FIG. 27, the service level management device 400 in the service level monitoring system 1 of this embodiment includes a service level calculation unit 104 and a service level display unit similar to the service level management device 300 of the above embodiment of FIG. 208, a unit conversion definition table 308, and a unit variable definition table 330, and further, an analysis unit 406, a service level variable calculation unit 402, a service level variable definition table 430, and a service level variable data storage unit 450. And comprising.
In addition, the service level definition table 410, the service level storage having the same structure as the service level definition table 110, the service level storage unit 170, and the service level calculation definition table 220 of the service level management apparatus 200 of the above embodiment of FIG. A unit 470 and a service level calculation definition table 440.

As shown in FIG. 28, the service level definition table 410 of this embodiment has the same configuration as the service level definition table 110 of FIG. 5 or the service level definition table 310 of FIG. FIG. 28 shows an example in which records 417 a to 417 c different from the service level definition table 110 or the service level definition table 310 are stored in the service level definition table 410.

In this embodiment, the service level calculation definition table 440 in FIG. 30 has the same configuration as the service level calculation definition table 220 in FIG. 10 or the service level calculation definition table 340 in FIG. FIG. 30 shows an example in which records 449 a to 449 c different from the service level calculation definition table 220 or the service level calculation definition table 340 are stored in the service level calculation definition table 440.

Note that the configuration of the service level management apparatus 400 according to the present embodiment is not limited to the service level management apparatus 300 according to the above-described embodiment illustrated in FIG. 21, and the present embodiment is not limited to the service level management apparatus 100 illustrated in FIG. 3 may be combined with the configuration of the service level management apparatus 200 in FIG.

In the present embodiment, the analysis unit 406 defines a rule (service level variable definition 430) for converting a plurality of service level definitions 410 into units using the aggregation unit defined by the unit variables. The service level calculation unit 104 calculates a service level by converting the plurality of service level definitions 410 as a unit according to (service level variable definition 430) defined by the analysis unit 406.

The service level variable definition table 430 is a definition used by the service level calculation unit 104 to calculate the value of the SL variable. The SL variable indicates a value used by the service level calculation unit 104 to calculate a service level value.

FIG. 29 shows an exemplary data structure of the service level variable definition table 430. The service level variable definition table 430 shows at least one record 434 (FIG. 29 shows two

records

434a and 434b. In the following, when there is no need to distinguish each record, a reference numeral “434” is attached. ) Can be included. Each record 434 includes a totaling method 152, a variable 153, a totaling time 154, a totaling space 155, and a time unit 156 similar to the records in the unit variable definition table 150 in FIG. 9 or the unit variable definition table 330 in FIG. Or a spatial unit 157, and further includes a calculation order 431, an SL variable name 432, or a reference source 433.
Note that the configuration of the service level variable definition table 430 in FIG. 29 is an example, and the present invention is not limited to this.

The calculation order 431 means the order in which SL variables are calculated. The SL variable name 432 indicates a name for identifying the SL variable. The aggregation method 152 indicates a method for calculating the value of the SL variable. A variable 153 indicates the name of a variable used for calculating the value of the SL variable. The total time 154 indicates a time range in which the variables are totaled and the value of the SL variable is calculated. The total space 155 indicates the range of the space where the variables are totaled and the value of the SL variable is calculated. The time unit 156 indicates the time width of the variable used for calculating the SL variable. The space unit 157 indicates the space width of the variable used for calculating the SL variable. A reference source 433 indicates a location where a variable is read. In the example of FIG. 25, when the reference source 433 is “0”, the service level variable calculation unit 402 reads a variable from the management data storage unit 160. On the other hand, when the reference source is “1”, the service level variable calculation unit 402 reads a variable from the service level variable data storage unit 450.

The operation of the service level management apparatus 400 of this embodiment configured as described above will be described below.
FIG. 31 is a flowchart showing the operation of the service level management apparatus 400 of the service level monitoring system 1 according to the embodiment of the present invention. The operation of the management data collection unit 240 (FIG. 18) described in the above embodiment is the same as that in the above embodiment. Hereinafter, the operation of the analysis unit 406 will be described.

First, the analysis unit 406 reads the service level definition 410 (step S201).
The operations in step S201 and step S203 are the same as the operation of the analysis unit 106 shown in the flowchart of FIG. 13 in the above embodiment. Further, the operations in steps S211 to S219 are the same as the operations of the analysis unit 106 shown in the flowchart of FIG.

Next, the analysis unit 406 searches for an SL variable based on the service level definition 410 and the processing result (step S401). The SL variable indicates a variable used for calculating the service level. Here, it is determined whether or not the service level value itself calculated using the unit variable can be used as a variable for calculating another service level.

For example, in the case of the service level definition 410 shown in FIG. 28, the unit variable is defined as “TAT (1 day, node)”. However, based on the time definition 120 of FIG. 6, when comparing the total unit with the total time 224 and total space 225 of “TAT_B” and “TAT_C”, “TAT_B” may fall between the unit variable and “TAT_C”. Recognize. In this case, the analysis unit 406 determines that “TAT_B” is an SL variable.

Next, the analysis unit 406 defines SL variables based on the service level definition 110 and the processing result (step S403).
For example, in the case of the service level definition 410 shown in FIG. 28, “TAT (1 day, node)” is defined as a unit variable based on the processing result. In this case, the analysis unit 406 defines the unit variable as an SL variable as it is. In addition to the definition as a unit variable, the analysis unit 406 may define the calculation order 431 and the reference source 433. The calculation order 431 is an order in which SL variables are calculated. When there are a plurality of SL variables, priority is given to the calculation order of SL variables having a small time unit and space unit.

If the spatial units are equal, the calculation order of SL variables with small time units is given priority. The reference source 433 indicates a storage device that reads the variable 153 used for calculating the SL variable. Here, “0” means the management data storage unit 160. Similarly, “1” means the service level variable data storage unit 450. Next, the analysis unit 406 defines “TAT_B” as an SL variable. At this time, the analysis unit 406 defines the SL variable name 432 of “TAT_B” as “TAT (1 day, service)” based on the service level definition 410. The variable 153 is “TAT (1 day, node)” based on the processing result. Further, since the SL variable “TAT (1 day, service)” is calculated using the value of the variable “TAT (1 day, node)”, the time unit 156 is “1 day” and the space unit 157 is “node”. Is. Also, since the SL variable “TAT (1 day, service)” is recorded in the service level variable data storage unit 450, the reference source 433 is “1”.

Next, the analysis unit 406 outputs a service level variable definition 430 based on the processing result (step S405).
Next, the analysis unit 406 outputs the service level calculation definition 440 (step S230). The operation of the analysis unit 406 that outputs the service level calculation definition 220 is the same as the service level calculation definition process S230 of FIG. 15 of the above embodiment.
In this way, the analysis unit 406 of the present embodiment can output the service level variable definition 430 and the service level calculation definition 440.

Next, an operation in which the service level variable calculation unit 402 calculates the SL variable based on the service level variable definition 430 will be described.
FIG. 32 is a flowchart showing the operation of the service level variable calculation unit 402 of this embodiment.

First, the service level variable calculation unit 402 reads the service level variable definition 430 (step S411).
Next, the service level variable calculation unit 402 calculates the calculation order 431 recorded in the service level variable definition 430, the SL variable name 432, the aggregation method 152, the variable 153, the aggregation time 154, the aggregation space 155, the time unit 156, and the space unit 157. Alternatively, based on the reference source 433, the management data is read (step S413), the SL variable is calculated (step S415), and is output to the service level variable data storage unit 450 (step S417).

For example, when the SL variable is calculated according to the service level variable definition 430 of FIG. 29, the service level variable calculation unit 402 first determines from the management data storage unit 160 “10 minutes” for the past day from the calculation time point, and , “TAT” in “node” unit is read. Next, the service level variable calculation unit 402 records the maximum value of the read values as “TAT (1 day, node)” in the service level variable data storage unit 450. Next, the service level variable calculation unit 402 sets the value of “TAT (1 day, node)” for the past day recorded in the service level variable data storage unit 450 to the number of nodes constituting “service”. Read based on. Next, the service level variable calculation unit 402 records the maximum value of the read values in the service level variable data storage unit 450 as “TAT (1 day, service)”.

The service level variable calculation unit 402 may refer to the time definition 120 and the space definition 130 in order to define the number of variables 153 used for calculation of the SL variable value, the dependency relationship, and the like.
As described above, the service level variable calculation unit 402 can calculate the SL variable based on the service level variable definition 430.

Further, the service level calculation unit 104 in the present embodiment operates in the same manner as the service level calculation unit 104 of the above-described embodiment described with reference to the flowchart of FIG. Then, the service level calculation unit 104 can output the service level value and store it in the service level storage unit 470.

As described above, according to the service level management apparatus 400 according to the embodiment of the present invention, the same effect as that of the above embodiment can be obtained, and another service level can be used to calculate the service level. It becomes possible to do.
The reason is that it is configured as follows. First, the analysis unit 406 outputs a service level variable definition 430 and a service level calculation definition 440 based on the service level definition 410. The service level variable definition 430 can define not only a common variable for calculating a service level but also a service level itself used as a variable for calculating other service levels.

Furthermore, according to the service level management apparatus 400 of the present embodiment, it is possible to reduce the amount and frequency of reading of the original management data by using another service level in order to calculate the service level. .
The reason is that it is configured as follows. First, the service level variable calculation unit 402 records the SL variable in the service level variable data storage unit 450 based on the service level variable definition 430. Next, the service level calculation unit 104 reads a variable from the management data storage unit 160 or the service level variable data storage unit 450 based on the service level calculation definition 440. Next, the service level calculation unit 104 calculates a service level using the read variable. At that time, it is not necessary to read the original management data again by using the SL variable calculated from the management data. In addition, since the SL variable calculated using the SL variable is also recorded in the service level variable data storage unit 450, the amount and the number of times of reading the variable are further reduced.

(Fifth embodiment)
FIG. 33 is a functional block diagram showing the configuration of the service level monitoring system 1 according to the embodiment of the present invention.
The service level management apparatus 500 of the service level monitoring system 1 according to the present embodiment is different from the above embodiment in that it can automatically determine without using variables to be used for service level calculation in advance.

As shown in FIG. 33, the service level management device 500 in the service level monitoring system 1 of this embodiment includes an analysis unit 106, a service level display unit 208, and the service level management device 200 of the above embodiment of FIG. , And a totaling unit 502 and a service level calculation unit 504.

Further, the service level management apparatus 500 of this embodiment includes a service level definition table 110, a unit conversion definition table storage unit 108, and a unit variable definition table 150 similar to the service level management apparatus 200 of the above embodiment of FIG. A unit variable data storage unit 210, a service level calculation definition table 220, and a service level storage unit 170, and further includes a unit variable table 510.

The service level management apparatus 500 according to the present embodiment is not limited to the configuration of the service level management apparatus 200 according to the above-described embodiment shown in FIG. 3, but the service level management apparatus 100 shown in FIG. 1, the service level management apparatus 300 shown in FIG. The present invention can also be applied to the configuration of 27 service level management apparatuses 400.

The unit variable table 510 is a table that includes information such as unit variable names, tabulation methods, variables, time units, and space units recorded in the unit variable data storage unit 210.

FIG. 34 shows an example of the unit variable table 510. Referring to FIG. 34, the unit variable table 510 shows at least one record 515 (FIG. 34 shows three records 515a to 515c. Hereinafter, when it is not necessary to distinguish each record, “515” is shown. Can be included. Each record 515 includes a unit variable name 511, a total time 512, a total space 513, or a total method 514.
The configuration of the unit variable table 510 in FIG. 34 is an example, and the present invention is not limited to this. For example, the record 515 may further include an aggregated variable name, value, time unit, space unit, or the like.

For example, referring to the unit variable table 510 of FIG. 34, the maximum value obtained by tabulating the period “2011/07/31 15:01 to 2011/08/01 15:00” for “node A” of the record 515a is “TAT ( It is recorded in the unit variable data storage unit 210 as a value of “1 day, node)”. The same applies to the other records 515.

In addition to the operation of the aggregation unit 102 of the above embodiment, the aggregation unit 502 calculates a unit variable according to the unit variable definition 150 and outputs the unit variable name 212 regarding the unit variable output to the unit variable data storage unit 210 when outputting the unit variable. The information such as the total time 213, the total space 214, and the total method 215 is output to the unit variable name 511, the total time 512, the total space 513, and the total method 514 of the unit variable table 510, respectively. Information such as the variable 153, the time unit 156, and the space unit 157 corresponding to the aggregated unit variable definition 150 may be further output to the unit variable table 510.

The unit variable table 510 of this embodiment does not include the value 216, unlike the data items of the unit variable data storage unit 210 of FIG. Since the value 216 includes numerical data and the like, the data size becomes relatively large, and the capacity of the unit variable data storage unit 210 becomes large when information management is continued for a long time in a large-scale system. For this reason, the process of referring to the data in the unit variable data storage unit 210 increases the load. On the other hand, since the information recorded in the unit variable table 510 can be configured to include only identification information such as the unit variable name 511, the capacity can be kept relatively small, and the reference processing load of the unit variable table 510 can be reduced. be able to.

Also, the service level calculation unit 504 calculates a service level value according to the service level calculation definition 220. At this time, the service level calculation unit 504 determines whether or not the variable recorded in the service level calculation definition 220 is recorded in the unit variable data storage unit 210 in addition to the operation of the service level calculation unit 104 of the above embodiment. This is determined with reference to the unit variable table 510. When the corresponding variable exists in the unit variable table 510, the service level calculation unit 504 reads the variable from the unit variable data storage unit 210. On the other hand, when the corresponding variable does not exist in the unit variable table 510, the service level calculation unit 504 reads the variable from the management data storage unit 160.

As described above, according to the service level management apparatus 500 of the present embodiment, the same effect as that of the above embodiment can be obtained, and the service level calculation unit 504 can be either the management data storage unit 160 or the unit variable data storage unit 210. It is possible to automatically determine whether the necessary variables are read from without being defined in advance.
The reason is that the totaling unit 502 is configured to output information about the unit variables to be output to the unit variable data storage unit 210 to the unit variable table 510. The service level calculation unit 504 can determine whether or not a variable necessary for the service level value is recorded in the unit variable data storage unit 210 by referring to the unit variable table 510.

Furthermore, according to the service level management apparatus 500 of the present embodiment, the service level calculation unit 504 can calculate the service level value independently of the totaling unit 502.
The reason is that since the totaling unit 502 outputs the unit variable table 510, the service level calculation unit 504 refers to the unit variable table 510 to calculate the service level value even at a time different from the unit variable calculation time. Because it is possible to do. For example, the service level calculation definition 220 can be determined based on the unit variable table 510 and the service level definition 110 without depending on the analysis unit 106.

(Sixth embodiment)
FIG. 35 is a functional block diagram showing the main configuration of the service level management apparatus 600 of the service level monitoring system 1 according to the embodiment of the present invention.
The service level management apparatus 600 of the service level monitoring system 1 of this embodiment is different from the above embodiment in that the unit conversion definition is generated based on the system configuration information.

As shown in FIG. 35, the service level management apparatus 600 of the present embodiment has a configuration similar to that of the service level management apparatus 200 of the above embodiment of FIG. The unit conversion definition production | generation part 602 and the structure information table 610 are provided.

Note that the configuration of the service level management apparatus 600 of the present embodiment is not limited to the configuration of the service level management apparatus 200 of FIG. 3, but the service level management apparatus 100 of FIG. 1 of the above embodiment and the service level management apparatus 300 of FIG. The service level management apparatus 400 in FIG. 27 and the service level management apparatus 500 in FIG. 33 can also be applied.

The unit conversion definition generation unit 602 generates a time definition 120 or a space definition 130 stored in the unit conversion definition table storage unit 108 based on the service level definition 110 and the configuration information of the information processing system (stored in the configuration information table 610). To do.

The configuration information table 610 records configuration information of each node of the information processing system to be monitored.
FIG. 36 shows an example of the configuration information table 610. Referring to FIG. 36, the configuration information table 610 shows at least one record 615 (FIG. 36 shows four records 615a to 615d. In the following, if it is not necessary to distinguish each record, “615” is shown. Can be included. Each record 615 includes a number 611, an IP address 612, a service ID 613, or a software name (shown as “software” in the figure) 614.
The configuration of the configuration information table 610 in FIG. 36 is an example, and the present invention is not limited to this.

Numeral 611 indicates the number of a node included in the monitored system. An IP address 612 indicates an IP address assigned to each node. The service ID 613 indicates the ID of a service that uses each node. The software name 614 indicates the name of software installed on each node.

For example, the node with the number 611 “1” of the record 615a in FIG. 36 has the IP address “128.6.1.1.101”, belongs to the service “service001_e_commerce”, and installs “Apache”. Indicates that In FIG. 36, the IP address is described in IPv4. However, it may be described using IPv6 or name information having similar properties and identifiable in units of nodes.

For example, when the unit conversion definition generation unit 602 classifies each node based on the configuration information 610, it first classifies each node based on the IP address. In the example of FIG. 36, since the IP addresses of the nodes having the numbers 611 of “1” to “3” are “1286.16.1. ***”, the unit conversion definition generating unit 602 has the number 611 of “1”. ”To“ 3 ”are determined to exist in the same“ data center ”or in a cluster belonging to the same network. Here, “***” means an arbitrary value (0 to 256).

Next, the unit conversion definition generation unit 602 classifies each node based on the service ID 613. Since all the service IDs 613 of the nodes with numbers “1” to “3” in FIG. 36 are “service001_e_commerce”, the unit conversion definition generation unit 602 has the same “services” with the nodes with numbers “1” to “3”. ".

Next, the unit conversion definition generation unit 602 classifies each node based on the software name. For example, in FIG. 36, since the software names having the numbers 611 “1” and “2” are “Apache”, the unit conversion definition generation unit 602 determines that the two nodes belong to the same “layer”. To do. Note that the unit conversion definition generation unit 602 may be configured to input the relationship between software and layers in advance and determine the layer to which the node belongs. For example, the unit conversion definition generation unit 602 may determine that “Apache” belongs to the “WEB layer”, and “MySQL” belongs to the “DB layer”.

The unit conversion definition generation unit 602 generates a time definition 120 and a space definition 130 based on the service level definition 110 and the configuration information 610 and outputs them to the unit conversion definition table storage unit 108.

Next, the operation of the present embodiment will be described with reference to FIG. 37 and FIG. 37 and 38 are flowcharts showing the operation of the unit conversion definition generation unit 602 of this embodiment.

First, the operation in which the unit conversion definition generation unit 602 outputs the time definition 120 will be described with reference to the flowchart of FIG.
First, the unit conversion definition generation unit 602 reads the service level definition 110 in FIG. 5 (step S601).
Next, the unit conversion definition generation unit 602 collects information related to the total time 114 from each record 117 of the service level definition 110 (step S603).
For example, in the case of the service level definition 110 in FIG. 5, the unit conversion definition generation unit 602 reads information “January” or “1 day”.

Next, the unit conversion definition generation unit 602 determines a unit from the read total time 114 (step S605).
For example, in the case of the service level definition 110 in FIG. 5, the unit conversion definition generation unit 602 determines “month” or “day” as a unit.

Next, the unit conversion definition generation unit 602 calculates a relationship between units based on information input in advance (step S607).
For example, in the case of the service level definition 110 in FIG. 5, the relationship can be calculated by defining in advance “January” = “30 days”. Also, in the time definition 120, for example, when the unit conversion definition generation unit 602 converts “half year” to “day” or “hour” unit, it is defined in advance as “half year” = “6 months”, and separately. The unit conversion definition generation unit 602 may be configured to calculate “six months” = “180 days” based on the defined information “January” = “30 days”.

Next, the unit conversion definition generation unit 602 outputs the unit of the total time 114 collected from the service level definition 110 and the relationship between the units as the time definition 120 (step S609). FIG. 6 is an example of the time definition 120 output by the unit conversion definition generation unit 602.
As described above, the unit conversion definition generation unit 602 can output the time definition 120.

Next, an operation in which the unit conversion definition generation unit 602 outputs the space definition 130 will be described with reference to the flowchart of FIG.
First, the unit conversion definition generation unit 602 reads the configuration information 610 (step S611).
Next, the unit conversion definition generation unit 602 classifies each node for each item recorded in the configuration information 610. First, the unit conversion definition generation unit 602 classifies each node based on the IP address. Next, the unit conversion definition generation unit 602 classifies each node based on the service ID 613. Next, the unit conversion definition generation unit 602 classifies each node based on the software name 614 (step S613).

Next, the unit conversion definition generation unit 602 obtains the number of elements of each classification item based on the information of each node classified in step S613 (step S615).
For example, referring to the example of the operation step, the unit conversion definition generation unit 602 has three nodes belonging to the space whose IP address is “128.6.1. ***” based on the configuration information 610. There are three nodes belonging to the service ID 613 “service001_e_commerce”, two nodes belonging to the “AP layer”, and one node belonging to the “DB layer”. Here, “***” means an arbitrary value (0 to 256).

Next, the unit conversion definition generation unit 602 outputs the information collected up to the previous step as the space definition 130 (step S617). FIG. 7 is an example of the space definition 130 output by the unit conversion definition generation unit 602.
As described above, the unit conversion definition generation unit 602 can output the space definition 130.

As described above, according to the service level management apparatus 600 according to the embodiment of the present invention, the same effect as that of the above embodiment can be obtained. Further, based on the service level definition 110 and the configuration information 610, time can be automatically set. Definition 120 and space definition 130 can be output.
This is because the unit conversion definition generation unit 602 outputs the time definition 120 and the space definition 130 based on the service level definition 110 and the configuration information.

The present invention has been described above with reference to each embodiment and example, but the present invention is not limited to the above embodiment and example. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

In addition, each component described in each of the above embodiments does not necessarily have to be individually independent. For example, in each component, a plurality of components may be realized as one module, or one component may be realized as a plurality of modules. Each component is configured such that a component is a part of another component, or a part of a component overlaps a part of another component. Also good.

Furthermore, each component in each embodiment described above and a module that realizes each component may be realized by hardware, or may be realized by a computer and a program if possible. However, it may be realized by mixing hardware modules, computers, and programs. The program is provided by being recorded on a non-volatile computer-readable recording medium such as a magnetic disk or a semiconductor memory, and read by the computer when the computer is started up. The read program causes the computer to function as a component in each of the above-described embodiments by controlling the operation of the computer.

In each of the embodiments described above, a plurality of operations are described in order in the form of a flowchart. However, the order of description does not limit the order in which the plurality of operations are executed. For this reason, when each embodiment is implemented, the order of the plurality of operations can be changed within a range that does not hinder the contents.

Furthermore, in each embodiment described above, a plurality of operations are not limited to being executed at different timings. For example, another operation may occur during execution of a certain operation, or the execution timing of a certain operation and another operation may partially or entirely overlap.

Furthermore, in each of the embodiments described above, it is described that a certain operation becomes a trigger for another operation, but the description does not limit all relationships between the certain operation and other operations. For this reason, when each embodiment is implemented, the relationship between the plurality of operations can be changed within a range that does not hinder the contents. The specific description of each operation of each component does not limit each operation of each component. For this reason, each specific operation | movement of each component may be changed in the range which does not cause trouble with respect to a functional, performance, and other characteristic in implementing each embodiment.

As described above, the embodiments of the present invention have been described with reference to the drawings. However, these are exemplifications of the present invention, and various configurations other than the above can be adopted.

A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.
(Appendix 1)
A service level management method in a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
The management device is
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes,
Management data for monitoring the operating state of the information processing system is aggregated for each aggregation unit using the defined aggregation aggregation unit to obtain an aggregate value,
A service level management method for calculating service levels defined in a plurality of the service level definitions based on the aggregated values for the aggregated units based on the service level definitions defining the service levels.
(Appendix 2)
In the service level management method described in Appendix 1,
A service level management method in which the management device defines the highest unit among the service levels defined in the service level definition as the aggregated unit between service levels in which the aggregated units are vertically related to each other .
(Appendix 3)
In the service level management method described in appendix 1 or appendix 2,
A service in which the management device defines, as the aggregated aggregate unit, the smallest unit that is the innermost among the service levels defined in the service level definition between service levels in which the aggregated units are inclusive of each other Level management method.
(Appendix 4)
In the service level management method according to any one of appendix 1 to appendix 3,
The management device is
Using the aggregation unit defined in the aggregation aggregation unit, defining a rule for unit conversion of a plurality of the service level definitions,
A service level management method for calculating the service level by converting a plurality of the service level definitions according to the defined rule.
(Appendix 5)
In the service level management method according to any one of appendix 1 to appendix 4,
The management device is
Store the defined aggregated unit in the unit variable definition table,
A service level management method for referring to the unit variable definition table and totaling the management data to obtain a total value.
(Appendix 6)
In the service level management method according to any one of appendix 1 to appendix 5,
A service level management method in which the management device further generates the unit conversion definition table based on the service level definition and configuration information of the information processing system.

(Appendix 7)
In a computer that realizes a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
A procedure for calculating a management value obtained by monitoring the operation state of the information processing system, and calculating a total value for each total unit using the defined total unit
Based on a plurality of service level definitions defining the service level, and executing a procedure for calculating a service level from the aggregated values defined in the plurality of service level definitions based on the aggregated values for each aggregated unit Program to let you.
(Appendix 8)
In the program described in Appendix 7,
For causing the computer to further execute a procedure of defining the highest unit among the service levels defined in the service level definition as the aggregated unit between service levels in which the aggregated units are vertically related to each other program.
(Appendix 9)
In the program described in Appendix 7 or Appendix 8,
In the service level in which the aggregation units are inclusive of each other, the computer further executes a procedure for defining, as the aggregation aggregation unit, the smallest unit that is the innermost among the service levels defined in the service level definition Program to let you.
(Appendix 10)
In the program according to any one of appendix 7 to appendix 9,
A procedure for defining a rule for converting a plurality of the service level definitions using the aggregation unit defined in the aggregation aggregation unit;
A program for causing a computer to further execute a procedure of calculating the service level by converting a plurality of service level definitions according to the defined rule.
(Appendix 11)
In the program according to any one of appendix 7 to appendix 10,
A procedure for storing the defined aggregated unit in the unit variable definition table;
A program for causing a computer to further execute a procedure of referring to the unit variable definition table and totaling the management data to obtain a total value.
(Appendix 12)
In the program according to any one of appendix 7 to appendix 11,
A program for causing a computer to further execute a procedure for generating the unit conversion definition table based on the service level definition and the configuration information of the information processing system.

(Appendix 13)
In a computer that realizes a service level management device that calculates a service level that is a guarantee level to be satisfied by the information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
A program for executing a procedure for outputting the defined aggregated unit.
(Appendix 14)
In the program described in Appendix 13,
For causing the computer to further execute a procedure for defining the highest unit among the service levels defined in the service level definition as the aggregated unit between service levels in which the aggregated units are vertically related to each other program.
(Appendix 15)
In the program described in appendix 13 or appendix 14,
In the service level in which the aggregation units are inclusive of each other, the computer further executes a procedure for defining, as the aggregation aggregation unit, the smallest unit that is the innermost among the service levels defined in the service level definition Program to let you.

This application claims priority based on Japanese Patent Application No. 2012-004398 filed on January 12, 2012, the entire disclosure of which is incorporated herein.

Claims

Analyze multiple service level definitions that define how to calculate the service level, which is the assurance level that the information processing system must satisfy, and use the same attribute based on a unit conversion definition table that defines the relationship between different aggregation units for each attribute. Analyzing means for defining a plurality of different service levels having the aggregation unit, an aggregation aggregation unit for collecting the aggregation units between the same attributes,
Aggregation means for obtaining the aggregated value by aggregating the management data for monitoring the operating state of the information processing system in the aggregation unit;
Service level calculation means for calculating a service level from the aggregated values aggregated by the aggregation means based on a plurality of the service level definitions;
With
The aggregation means calculates the management data for each of the aggregation aggregation units defined by the analysis means to obtain an aggregation value,
The service level management device, wherein the service level calculation means calculates the service level defined in the plurality of service level definitions based on the aggregate value for each aggregate aggregation unit aggregated by the aggregation means.
In the service level management apparatus according to claim 1,
The analysis means is a service level management device that defines the highest unit among the service levels defined in the service level definition as the aggregated unit between service levels in which the aggregated units are vertically related to each other .
In the service level management apparatus according to claim 1 or 2,
The analysis unit is a service that defines, as the aggregated aggregate unit, the smallest unit that is the innermost among the service levels defined in the service level definition between service levels in which the aggregated units are inclusive of each other. Level management device.
In the service level management apparatus in any one of Claims 1 thru | or 3,
The analysis means defines a rule for unit conversion of a plurality of the service level definitions using the aggregation unit defined in the aggregation aggregation unit,
The service level calculation unit is a service level management device that calculates the service level by converting a plurality of the service level definitions into units according to the rules defined by the analysis unit.
The service level management device according to any one of claims 1 to 4,
The analysis means stores the defined aggregated unit in a unit variable definition table,
The aggregation means refers to the unit variable definition table, and aggregates the management data to obtain an aggregate value.
In the service level management apparatus in any one of Claims 1 thru | or 5,
A service level management apparatus further comprising a generating unit that generates the unit conversion definition table based on the service level definition and the configuration information of the information processing system.
A service level management method in a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
The management device is
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes,
Management data for monitoring the operating state of the information processing system is aggregated for each aggregation unit using the defined aggregation aggregation unit to obtain an aggregate value,
A service level management method for calculating service levels defined in a plurality of the service level definitions based on the aggregated values for the aggregated units based on the service level definitions defining the service levels.
In a computer that realizes a management device that manages a service level that is a guarantee level to be satisfied by an information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
A procedure for calculating a management value obtained by monitoring the operation state of the information processing system, and calculating a total value for each total unit using the defined total unit
Based on a plurality of service level definitions defining the service level, and executing a procedure for calculating a service level from the aggregated values defined in the plurality of service level definitions based on the aggregated values for each aggregated unit Program to let you.
A service level management device according to any one of claims 1 to 6,
A service level monitoring system that monitors the operating state of an information processing system using a calculated service level.
A computer that realizes a service level management device that calculates a service level that is a guarantee level to be satisfied by the information processing system,
Analyzing a plurality of service level definitions that define the calculation method of the service level, defining a plurality of different service levels having the same attribute and different aggregation units, and defining an aggregation aggregation unit that aggregates the aggregation units between the same attributes;
A program for executing a procedure for outputting the defined aggregated unit.