WO2013077439A1

WO2013077439A1 - Monitoring device and method for monitoring

Info

Publication number: WO2013077439A1
Application number: PCT/JP2012/080404
Authority: WO
Inventors: 允裕大野
Original assignee: 日本電気株式会社
Priority date: 2011-11-21
Filing date: 2012-11-16
Publication date: 2013-05-30
Also published as: JPWO2013077439A1

Abstract

The objective of the present invention is to suppress the amount of time for reading operational data from a storage unit even if capacity of the operational data becomes enormous in order to calculate the value of a service level category. This monitoring device includes: an operational data history storage unit, which includes a plurality of storage units, which stores a value of operational data indicating information of a node to be monitored, in association with an identifier capable of identifying the operational data; a monitoring category definition storage unit which associates a monitoring category, for monitoring a system, with the identifier of the operational data used for calculation of an index of the monitoring category and stores the same; a counter unit for counting the number of monitoring categories that have been associated with the identifier; and a selection unit which, in accordance with the number that has been counted, selects a storage unit to store a value of the operational data which has been associated with the identifier.

Description

Monitoring device and monitoring method

The present invention relates to a monitoring apparatus and a monitoring method for monitoring an IT (information technology) system.

IT usage forms include XaaS (X as a Service) that provides processing performance of a data center as a service via a network. In such an IT usage mode, a service provider that operates a data center and a client that uses the service are service level items (SLO: Service) that are indexes in which target values of service quality and contents are set. Often, contracts that include (Level Objective) are signed.
The service provider needs to monitor the data center in order to comply with the concluded contract, that is, to check whether the service provided to the client satisfies the target value set in the service level item. “Patent Document 1” describes an example of a monitoring device for operating and managing a data center.
A monitoring device that monitors the data center stores values such as usage information of the IT system to be monitored as operation data values. Then, the monitoring device reads the value of the stored operation data for every fixed period, and calculates the value of the service level item using a predetermined calculation formula.
JP 2009-146001 A

When the monitoring device monitors a wide variety of service level items, there are a wide variety of types of operational data required to calculate the service level items. For this reason, the volume of operational data stored in the monitoring device is enormous. In this case, when calculating the service level item, the monitoring device described in Patent Literature 1 has a problem that it takes a long time to read the value of the operation data from the storage unit.
The objective of this invention is providing the monitoring apparatus and monitoring method which solve the said subject.

A first invention for solving the above-described problem is an operational data history including a plurality of storage means for storing operational data values indicating information of nodes to be monitored in association with identifiers capable of identifying the operational data. A storage item, a monitoring item for monitoring the system, and an identifier of the operational data used for calculating the index of the monitoring item are stored in association with each other, and a monitoring item definition storage unit is associated with the identifier. A monitoring device comprising: counting means for counting the number of monitoring items; and selection means for selecting the storage means for storing the value of operation data associated with the identifier according to the counted number. is there.
The second invention that solves the above-described problem includes a plurality of storage means for storing a value of operation data indicating information of a node to be monitored in association with an identifier that can identify the operation data, and monitors the system A monitoring item for monitoring and an identifier of the operational data used for calculation of an index of the monitoring item, the number of monitoring items associated with the identifier is counted, and the counted number The monitoring device control method of selecting a storage means for storing the value of the operation data associated with the identifier according to the method.
A third invention for solving the above-mentioned problem includes a plurality of storage means for storing the value of operation data indicating information of a node to be monitored in association with an identifier capable of identifying the operation data, and monitors the system A process for storing a monitoring item to be associated with an identifier of the operational data used to calculate an index of the monitoring item, a process for counting the number of monitoring items associated with the identifier, A program for causing a computer to execute a process of selecting the storage means for storing a value of operation data associated with the identifier according to a counted number.

According to the present invention, it is possible to reduce the time for reading the operation data from the storage unit even if the amount of operation data necessary for calculating the value of the service level item becomes enormous.

FIG. 1 is a diagram illustrating an example of the configuration of the monitoring system 1000. FIG. 2 is a diagram illustrating an example of the configuration of the IT system 100. FIG. 3 is a diagram illustrating an example of operational data. FIG. 4 is a diagram illustrating an example of the configuration of the monitoring device 200. FIG. 5 is a diagram illustrating an example of data stored in the operation data history storage unit 220. FIG. 6 is a diagram illustrating an example of the screen of the SLO input console 231. FIG. 7 is a diagram illustrating an example of data stored in the SLO type storage unit 232. FIG. 8A is a diagram illustrating an example of SLO definition data. FIG. 8B is a diagram illustrating an example of SLO definition data. FIG. 9 is a diagram illustrating an example of data stored in the SLO definition data storage unit 240. FIG. 10 is a diagram illustrating an example of data stored in the SLO calculation result storage unit 252. FIG. 11 is a diagram illustrating an example of a table in which reference priorities are associated with selected storage units. FIG. 12 is a flowchart illustrating an example of an operation for calculating the SLO value from the operation data value. FIG. 13 is a flowchart illustrating an example of an operation for determining the reference priority of operational data. FIG. 14 is a diagram illustrating an example of data in which operational data identifiers and reference priorities are associated with each other. FIG. 15 is a flowchart illustrating an example of an operation for selecting a storage location of operation data. FIG. 16 is a flowchart illustrating an example of an operation for determining the reference priority of operational data. FIG. 17A is a diagram illustrating an example of a calculation process state transition map. FIG. 17B is a diagram illustrating an example of a calculation process state transition map. FIG. 18 is a flowchart illustrating an example of a process for creating a calculation process state transition map. FIG. 19 is a diagram illustrating an example of an overlapping transition map. FIG. 20 is a flowchart illustrating an example of a process for overwriting a duplicate transition map. FIG. 21 is a block diagram illustrating an example of the configuration of the monitoring apparatus 200.

= First embodiment =
A monitoring system 1000 according to a first embodiment of the present invention will be described in detail with reference to the drawings.
<Overall configuration>
FIG. 1 is a diagram illustrating an example of a configuration of a monitoring system 1000 according to the first embodiment.
A monitoring system 1000 according to the first embodiment includes an IT system 100, a monitoring device 200, a network 300, a network 400, and a client terminal 500.
The IT system 100 and the monitoring apparatus 200 are connected via a network 300 so that they can communicate with each other. The IT system 100 is connected to the client terminal 500 through the network 400 so as to be communicable.
The network 300 and the network 400 may be the same network.
The client terminal 500 transmits a request for a desired process to the IT system 100 via the network 400.
The IT system 100 performs processing according to the request received from the client terminal 500. The IT system 100 transmits the processing result of the IT system 100 to the client terminal 500 via the network 400.
The IT system 100 transmits data representing usage information, performance information, or operation information of the IT system 100 to the monitoring device 200. Hereinafter, the usage information, performance information, or operation information is collectively referred to as “use information”.
The monitoring device 200 receives and stores data representing usage information of the IT system 100 from the IT system 100. The monitoring device 200 reads data values representing usage information and the like of the IT system 100 received and stored from the IT system 100 at regular intervals. And the monitoring apparatus 200 calculates the value of the service level item of the IT system 100 using a predetermined calculation formula. Hereinafter, the service level item is described as “SLO (Service Level Objective)”.
<Configuration of IT system 100>
FIG. 2 is a diagram illustrating an example of the configuration of the IT system 100.
The IT system 100 includes an operation data transmission unit 110 and various IT devices (120 to 170).
The various IT devices are, for example, a Web server 120, a WebAP (Application) server 130, a DB (Database) server 140, a storage server 150, a router 160, or a switching hub 170.
The operation data transmission unit 110 transmits operation data of various IT devices to the monitoring apparatus 200.
FIG. 3 is a diagram illustrating an example of operational data that the operational data transmission unit 110 transmits to the monitoring device 200.
The operational data includes operational data identifiers, operational data acquisition times, and operational data values including values such as usage information of various IT devices.
The identifier of the operational data includes at least information indicating the operational data acquisition range and operational data type such as usage information. For example, information on the combination of the operational data acquisition range and the operational data type becomes the operational data identifier. The identifier of the operation data may include the acquisition time width of the operation data.
The operation data shown in FIG. 3 includes the operation data acquisition range of node A, the type of operation data called stop time, the operation data acquisition time width of every day, and November 30, 2011 00:00:00. This is operational data including operational data acquisition time of seconds and operational data value of 10 minutes.
The operational data acquisition range is information for identifying which IT device usage information is the operational data. The operation data acquisition range is not limited to a single IT device, but may be a combination of a plurality of IT devices. Further, the operation data acquisition range may be a software program that operates on an IT device or system. Hereinafter, IT devices, software, and the like that are the operation data acquisition range will be referred to as “nodes”.
The type of operational data is information including information for specifying the type of usage information, performance information, or operation information of a node to be monitored, for example.
Node resource usage information includes, for example, operation data indicating usage and usage of node components (CPU (Central Processing Unit) memory, disk and network), event logs such as login and start / stop, and access Operation data related to operation logs of IT devices and software, such as event logs. The node performance information includes, for example, the processing elapsed time from the arrival of a request to the node until the response is output, the response time from the start of request transmission from the user node to the response display, and the request arrival rate per unit time This is operational data related to the processing capacity of the node, such as throughput. The node operation information includes, for example, an operation time indicating an actual measurement time after operation from the start of operation of the node to an operation stop, a stop time indicating an actual measurement time after operation from the operation stop of the node to the start of operation, Operation data related to operation, such as an assumed operation time indicating an assumed time before operation from the start to the operation stop.
The operation data acquisition time width is information representing a cycle in which the value of the operation data is output.
The operation data acquisition time is the time when the value of the operation data is output.
The value of operational data is a value indicated by operational data usage information or the like. The value of the operational data may be a measured value at the moment when the operational data is output, or may be an integrated value or an average value from when the previous operational data value is output until the current operational data value is output.
<Configuration of Monitoring Device 200>
FIG. 4 is a diagram illustrating an example of the configuration of the monitoring device 200.
The monitoring device 200 includes an operation data receiving unit 210, an operation data history storage unit 220, an SLO input unit 230, an SLO definition data storage unit 240, an SLO calculation unit 250, a reference priority determination unit 260, a storage location And a control unit 270.
The operation data history storage unit 220 further includes a high-speed storage unit 221 and a low-speed storage unit 222.
The SLO input unit 230 further includes an SLO input console 231 and an SLO type storage unit 232.
The SLO calculation unit 250 further includes an SLO calculation execution unit 251 and an SLO calculation result storage unit 252.
<Hardware configuration>
The monitoring device 200 may be configured based on a computer including a CPU (Central Processing Unit), a memory, a NIC (Network Interface Card), a storage unit, an input device, and an output device (all not shown).
In this case, the SLO calculation execution unit 251, the reference priority determination unit 260, and the storage location control unit 270 are SLO calculation execution programs configured with codes for the CPU to execute various operations according to the present embodiment. , A reference priority determination program, and a storage location control program (both not shown). However, the monitoring device 200 is naturally not limited to this mode, and may be mounted in advance as a dedicated circuit in hardware.
The operation data history storage unit 220, the SLO type storage unit 232, the SLO definition data storage unit 240, and the SLO calculation result storage unit 252 can be realized based on storing each data in the storage unit. The storage unit is, for example, a hard disk device or a semiconductor storage device.
The operation data receiving unit 210 can be realized, for example, when the CPU controls a NIC (Network Interface Card) based on a predetermined program.
The SLO input console 231 is a screen for inputting data to the monitoring apparatus 200, and functions as a user interface. The SLO input console 231 can be realized based on, for example, the CPU displaying a predetermined screen on the display according to a predetermined program. The system monitor inputs data based on the operation of the screen displayed on the display using a keyboard, a mouse, or the like.
<Operation data receiving unit 210>
The operation data receiving unit 210 receives operation data including “operation data value” from the IT system 100. The operation data receiving unit 210 causes the operation data history storage unit 220 to store the received operation data.
<Operation Data History Storage Unit 220>
The operation data history storage unit 220 stores the operation data received from the IT system 100 by the operation data receiving unit 210.
The operational data history storage unit 220 includes a high-speed storage unit 221 having a high read speed and a low-speed storage unit 222 having a low read speed.
The high-speed storage unit 221 includes a high-speed readable storage unit such as an SSD (Solid State Drive) memory disk drive. The low-speed storage unit 222 includes a low-speed readable storage unit such as a hard disk drive.
The high-speed storage unit 221 may be configured by, for example, a high-speed readable SAS (Serial Attached SCSI) at a speed of 15,000 rpm. The low-speed storage unit 222 may be configured with a low-speed read SATA (Serial Advanced Technology Attachment) at 7200 revolutions per minute.
The operation data history storage unit 220 may be configured by combining three or more storage units having different reading speeds.
FIG. 5 is a diagram illustrating an example of data stored in the operation data history storage unit 220.
The operational data history storage unit 220 stores operational data, for example, for each operational data having a common identifier, in chronological order of operational data acquisition time.
The operation data stored in the operation data history storage unit 220 may be associated with a reference priority using a reference priority determination unit 260 described later. In this case, the operation data stored in the operation data history storage unit 220 is divided and stored in the high-speed storage unit 221 and the low-speed storage unit 222 according to the reference priority.
In the example of operation data stored in the operation data history storage unit 220 illustrated in FIG. 5, operation data having a reference priority “1” is stored in the high-speed storage unit 221. Further, the operation data having the reference priority “2” is stored in the low speed storage unit 222.
<SLO input unit 230>
The SLO input unit 230 includes an SLO input console 231 and an SLO type storage unit 232. The SLO input unit 230 inputs SLO definition data that defines the SLO to the monitoring apparatus 200. The SLO definition data in this embodiment includes, for example, the name of the type of SLO, the SLO calculation range, the SLO calculation period width, and the SLO target value.
The name of the type of SLO is information that identifies the quality and content of the service to be monitored. As the name of the type of SLO, for example, the operating rate of the monitoring target and the achievement rate of TAT (Turn Around Time) in a certain period can be mentioned.
The SLO calculation range is information representing the range of nodes to be monitored. The SLO calculation range may be one node or a combination of a plurality of nodes. The SLO calculation range may be a system configured by a combination of a plurality of nodes. Alternatively, the SLO calculation range may be a layer such as a Web layer, a DB layer, or an AP layer. Alternatively, the SLO calculation range may be a system such as a web three-layer system. When a combination of a plurality of nodes is used as the SLO calculation range, the SLO calculation range and the names of the individual nodes constituting the SLO calculation range need to be stored separately in association with each other.
The SLO calculation period width is information that represents a cycle of counting SLOs. The SLO calculation period width is, for example, monthly, quarterly, or yearly.
The SLO target value is a target value set to SLO. For example, when the name of the type of SLO is an operation rate, the SLO target value is set to a number such as 99%, 99.9%, or 99.99%.
FIG. 6 is a diagram illustrating an example of the screen of the SLO input console 231. The SLO input console 231 is a user interface that is used when a system monitor inputs SLO definition data to the monitoring apparatus 200. From the screen shown in FIG. 6, the system monitor selects the name of the SLO type, the SLO calculation range, the SLO calculation period width, and the SLO target value from the pull-down menu, and inputs the SLO definition data. The SLO definition data input from the SLO input unit 230 is stored in the SLO definition data storage unit 240.
FIG. 7 is a diagram illustrating an example of data stored in the SLO type storage unit 232. For example, the SLO type storage unit 232 calculates the name of the SLO type, the type of operational data necessary to calculate the value of the SLO belonging to the type, and the value of SLO belonging to the type from the operational data. Are stored in association with each other. In the data shown in FIG. 7, for example, the value of the SLO of the type “actual operating time” is calculated as a calculation formula 1 using the values of the types of operation data of “assumed operating time” and “stop time”. Based on this, it is calculated. Further, in the data shown in FIG. 7, for example, the value of the SLO of the type “operating rate” is calculated as a calculation formula 2 using the operation data values of the types “assumed operating time” and “stop time”. Based on this, it is calculated.
When the system monitor selects the name of the SLO type from the SLO input console 231, the type of operation data necessary for calculating the SLO value is stored in the SLO definition data storage unit 240 together with the name of the selected SLO type. And the calculation formula are input in association with each other. For example, when “actual operation time” is selected as the name of the type of SLO, “expected operation time” and “stop time”, which are types of operation data used to calculate the value of “actual operation time”, and “actual operation time” The calculation formula (calculation formula 1) used for calculating the value “” is input to the SLO definition data storage unit 240 and stored therein.
8A and 8B are diagrams illustrating examples of SLO definition data stored in the SLO definition data storage unit 240 input from the SLO input unit 230, respectively.
The SLO definition data shown in FIG. 8A includes an SLO type “operation rate”, an SLO calculation range “node A and node B”, an SLO calculation period width “3 months”, and “99.9%”. SLO target value.
The SLO definition data shown in FIG. 8B includes the SLO type “actual operation time”, the SLO calculation range “node A”, the SLO calculation period width “3 months”, and the SLO target value “1880 hours”. Including.
Instead of providing the SLO type storage unit 232, the SLO input console 231 may provide an input field for an SLO calculation formula. In this case, the system monitor inputs the name of the SLO type and the SLO calculation formula directly into the SLO input console 231. The SLO input unit 230 acquires the type of operational data used for calculating the SLO value by using a syntax analysis that divides the SLO calculation formula into an operation symbol and a character string, and stores it in the SLO definition data storage unit 240. To do.
<SLO definition data storage unit 240>
FIG. 9 is a diagram illustrating an example of data stored in the SLO definition data storage unit 240. For example, the SLO definition data storage unit 240 stores the SLO identifier in association with the SLO definition data input from the SLO input unit 230.
In addition, for example, the SLO definition data storage unit 240 stores an SLO identifier and an identifier of operation data necessary for calculating an identified SLO value based on the SLO identifier in association with each other.
The data identified using the SLO identifier 001 shown in FIG. 9 corresponds to the SLO definition data shown in FIG. 8A. Similarly, the data identified using the SLO identifier 002 shown in FIG. 9 corresponds to the SLO definition data shown in FIG. 8B.
The data corresponding to the SLO definition data shown in FIG. 8A identified by the SLO identifier 001 is the type of operation data necessary for calculating the SLO value is “estimated operation time” and “stop time”, and the SLO calculation The range is “Node A” and “Node B”. Therefore, the identifier of the operation data necessary for calculating the SLO definition data corresponding to FIG. 8A, that is, the combination of the type of operation data and the acquisition range of the operation data is “the assumed operation time of node A”, “node A ”Stop time”, “expected operation time of node B”, and “stop time of node B”. Therefore, the SLO definition data shown in FIG. 8A is stored in the SLO definition data storage unit 240 in association with the identifiers of the four operational data.
The same applies to the data corresponding to the SLO definition data shown in FIG. 8B identified by the SLO identifier 002.
<SLO calculation unit 250>
The SLO calculation unit 250 calculates the SLO value from the operation data based on the SLO definition data stored in the SLO definition data storage unit 240. The SLO calculation unit 250 includes an SLO calculation execution unit 251 and an SLO calculation result storage unit 252.
The SLO calculation execution unit 251 calculates the value of SLO using the value of operation data and the SLO calculation formula. The SLO calculation execution unit 251 stores the calculated SLO value in the SLO calculation result storage unit 252. Details of the SLO calculation execution unit 251 will be clarified in the description of the operation described later.
The SLO calculation result storage unit 252 stores the SLO value calculated by the SLO calculation execution unit 251 as SLO calculation result data.
FIG. 10 is a diagram illustrating an example of data stored in the SLO calculation result storage unit 252. The SLO calculation result data includes the SLO identifier, the name of the SLO type, the SLO calculation range, the SLO calculation target period, the calculated SLO value, and the SLO calculation time.
The SLO calculation target period is information representing a period that is backed by the SLO calculation period width from the period in which the SLO value is calculated. The SLO calculation target period goes back in months if the SLO calculation period width is in units of months, and goes back in days if it is in days. For example, if the SLO value is calculated at 03:10:15 on April 1, and the SLO calculation period is 3 months, the SLO calculation target period is January 1, 0: 0: 0 From March 31 to 23:59. If the SLO value is calculated at 03:10:15 on April 1, and the SLO calculation period is 1 day, then the SLO calculation target period is March 31, 0: 0: 0 From March 31 to 23:59:59.
The SLO calculation result data shown in FIG. 10 includes an SLO identifier “SLO — 001”, an SLO type name “operation rate”, an SLO calculation range “node A and node B”, and “three months”. SLO calculation target period (specifically, from January 1, 00:00:00 to March 31, 23:59:59), an SLO value of “99.97%”, and “4 The data includes the SLO calculation time of “Monday 1 03:10:15”.
<Reference priority determination unit 260>
The reference priority determination unit 260 gives a reference priority to the operation data. The reference priority is a parameter for selecting a storage destination of operation data from a plurality of storage units (high-speed storage unit 221 and low-speed storage unit 222 in the present embodiment) included in the operation data history storage unit 220. is there.
The reference priority determination unit 260 counts how many SLO identifiers each identifier of the operational data is associated with. In other words, the reference priority determination unit 260 counts how many SLOs a certain operational data is used for. The reference priority determination unit 260 associates the reference priority with the identifier of the operation data according to the counted number. Details of the reference priority determination unit 260 will be made clear in the description of the operation described later.
<Storage location control unit 270>
The storage location control unit 270 includes a plurality of storage units (a high-speed storage unit 221 and a low-speed storage unit 222) in which the operation data history storage unit 220 includes the storage destination of the operation data according to the reference priority associated with the operation data. ) To select. Preferably, the storage location control unit 270 selects the high-speed storage unit 221 for operation data with a high reference priority, and selects the low-speed storage unit 222 for operation data with a low reference priority.
The storage location control unit 270 stores, for example, a table shown in FIG. 11 in which the reference priority is associated with the selected storage unit. The storage location control unit 270 rearranges the value of the operation data stored in the operation data history storage unit 220 to an appropriate storage unit according to the reference priority associated with the operation data. Details of the storage location control unit 270 will be clarified in the description of the operation described later.
<Description of operation>
The operation of this embodiment will be described in detail.
<Description of operation for calculating SLO value>
FIG. 12 is a flowchart illustrating an example of an operation in which the SLO calculation execution unit 251 calculates the SLO value from the operation data value.
The SLO calculation execution unit 251 starts the calculation of the SLO value with the SLO value calculation period as a trigger (S101). A specific method of detecting the SLO value calculation period by the SLO calculation execution unit 251 may be, for example, a method of detecting the progress of the calculation period width from the time when the SLO value was calculated last time. Further, the method for detecting the calculation period of the SLO value may be a method in which the timing for calculating the SLO is specifically determined for each SLO calculation period width. For example, when the SLO calculation period width is in units of months, the method of detecting the SLO value calculation period may be a method of calculating the SLO value at 00:00:00 on the first day of every month. Alternatively, the method for detecting the SLO value calculation period may be a method in which the system supervisor clearly indicates the timing for calculating the SLO value.
The SLO calculation execution unit 251 acquires “an identifier of operation data necessary for calculating the SLO” corresponding to the SLO whose calculation period has come (S102).
The SLO calculation execution unit 251 uses the current time and the SLO calculation time width to calculate the range to which the operation data acquisition time necessary to calculate the SLO value belongs (S103). For example, assuming that the current time is 00: 00: 00: 00 on April 1 and the SLO calculation period is 3 months, the range of operation data acquisition time required to calculate SLO is January 1 0:00 From 0 minutes 0 seconds to March 31 23:59:59.
The SLO calculation execution unit 251 determines the value of the operation data necessary for calculating the value of the SLO based on the “type of operation data necessary for calculating the SLO” and the range to which the operation data acquisition time calculated in S103 belongs. And the value of operational data is acquired (S104).
The SLO calculation execution unit 251 calculates the SLO value based on the acquired operational data value and the SLO calculation formula (S105).
The SLO calculation execution unit 251 stores the calculated SLO value in the SLO calculation result storage unit 252 as SLO calculation result data (S106).
<Description of operation for determining operational data reference priority>
FIG. 13 is a flowchart illustrating an example of an operation in which the reference priority determination unit 260 determines the reference priority of operational data.
The reference priority determination unit 260 starts the operation of determining the reference priority of the operation data using the addition or update of the SLO definition data storage unit 240 as a trigger (S201). The method in which the reference priority determination unit 260 detects the update of the SLO definition data storage unit 240 is stored, for example, by the reference priority determination unit 260 periodically polling the SLO definition data storage unit 240. A method of detecting whether or not there is addition / update in the SLO definition data may be used. Or the start of the operation | movement which determines the reference priority of the operation data of the reference priority determination part 260 may specify the timing when a system supervisor determines a reference priority.
Here, as an example, the update trigger is such that the total number of SLO identifiers stored in the SLO definition data storage unit 240 is “n> 0”.
The reference priority determination unit 260 acquires one SLO identifier from the SLO definition data storage unit 240 (S202).
For the sake of explanation, it is assumed that the number of identifiers of operational data necessary for calculating the SLO value corresponding to the SLO identifier acquired in S202 is “m> 0”.
The reference priority determination unit 260 acquires one operational data identifier corresponding to the SLO identifier acquired in S202 (S203).
The reference priority determination unit 260 determines whether or not the operational data identifier acquired in S203 is the acquired operational data identifier (S204).
If the reference priority determination unit 260 determines that the operation data identifier is not already acquired (NO in S204), that is, if the reference priority determination unit 260 determines that the operation data identifier is a newly acquired operation data identifier, the reference priority determination unit 260 sets the operation data identifier to the count number “1”. Is stored (S205).
When the reference priority determination unit 260 determines that the operation data identifier has already been acquired (YES in S204), the reference priority determination unit 260 increases the count number of the operation data identifier by one (S206).
The reference priority determination unit 260 repeats the operations of S204 to S206 for all operation data identifiers corresponding to the SLO identifier acquired in S202 (S207).
The reference priority determination unit 260 acquires the next SLO identifier when the operations of S204 to S206 are repeated for all the operational data identifiers corresponding to the SLO identifier.
The reference priority determination unit 260 repeats the operations of S204 to S207 for all the SLO identifiers stored in the SLO definition data storage unit 240 (S208).
When the reference priority determination unit 260 repeats the operations of S204 to S207 for all the SLO identifiers stored in the SLO definition data storage unit 240, the process proceeds to the processing of S209.
The reference priority determination unit 260 associates the reference priority with the operation data identifier based on the count number of each operation data identifier (S209).
The reference priority determination unit 260 may use the count value as the reference priority. Further, the reference priority determination unit 260 may use a value obtained by weighting the count value as the reference priority.
The reference priority determination unit 260 transmits data in which the operation data identifier and the reference priority are associated to the storage location control unit 270 (S210).
FIG. 14 is a diagram illustrating an example of data in which operational data identifiers and reference priorities are associated with each other. In the data shown in FIG. 14, for example, the reference priority “2” is associated with the operation data identifier “node A assumed operation time”, and the reference priority “1” is the operation data identifier “node B stop time”. Associated with.
<Description of operation for controlling storage location of operational data>
FIG. 15 is a flowchart illustrating an example of an operation in which the storage location control unit 270 selects a storage location for operation data.
The storage location control unit 270 optimizes the storage location of the operation data in the operation data history storage unit 220 at a predetermined timing.
The storage location control unit 270 refers to the data associated with the operation data identifier and the reference priority received from the reference priority determination unit 260 (S301).
The storage location control unit 270 rearranges the operation data stored in the operation data history storage unit 220 in the storage unit according to the reference priority associated with the operation data identifier of the operation data (S302).
The timing at which the value of the operation data stored in the operation data history storage unit 220 is rearranged may be designated each time by the system monitor. Alternatively, the timing may be every time the data in the SLO definition data storage unit 240 is updated.
Further, when the operational data receiving unit 210 receives a new operational data value, the storage location control unit 270 has already stored operational data having the same identifier as the received operational data in the operational data history storage unit 220. Check whether or not Then, the storage location control unit 270 may apply the reference priority assigned to the operation data to the received operation data, select an appropriate storage unit, and select a storage destination.
Further, when the operation data receiving unit 210 receives a new value of operation data, the storage location control unit 270 may select the storage destination by uniformly considering the reference priority “1”.
The monitoring apparatus 200 according to the present embodiment takes time to read out the operation data value when calculating the SLO value even if the storage capacity of the operation data necessary for calculating the SLO value becomes enormous. The effect which suppresses can be acquired.
The reason is that the monitoring apparatus 200 according to the present embodiment controls the operation data used for calculation of a large number of SLO values so that higher reference priorities are associated and stored in a storage medium that can be read at high speed. It is.
Note that the monitoring apparatus 200 of the present embodiment is not limited to the configuration shown in FIG.
FIG. 20 is a block diagram illustrating an example of the minimum configuration of the monitoring apparatus 200.
The same number is attached | subjected to the structure similar to FIG. Detailed description is omitted.
Monitoring device 200 includes an operation data history storage unit 220, an SLO definition data storage unit 240, a reference priority determination unit 260, and a storage location control unit 270.
The operation data history storage unit 220 stores operation data including operation data values indicating information of nodes to be monitored in association with operation data identifiers. The operation data history storage unit 220 includes a plurality of storage units.
The SLO definition data storage unit 240 stores SLO definition data. The SLO definition data is data in which monitoring items for monitoring the system such as the type of SLO are associated with operation data used for calculating the index of the monitoring item such as a calculation formula. Therefore, it can be said that the SLO definition data storage unit 240 is a monitoring item definition storage unit.
The reference priority determination unit 260 counts monitoring items associated with the identifier. Therefore, the reference priority determination unit 260 can also be called a counting unit.
The storage location control unit 270 selects a storage unit of the operation data history storage unit 220 that stores the value of the operation data associated with the identifier according to the counted number. Therefore, the storage location control unit 270 can also be called a selection unit.
The monitoring apparatus 200 with the minimum configuration shown in FIG. 21 configured as described above can obtain the same effects as the monitoring apparatus 200 shown in FIG.
The reason is that the storage location control unit 270 can select the storage unit of the operational data history storage unit 220 based on the number of SLO definition data measured by the reference priority determination unit 260 in the SLO definition data storage unit 240. Because.
= Second embodiment =
A monitoring system 1000 according to a second embodiment of the present invention will be described in detail with reference to the drawings.
In the monitoring system 1000 according to the second embodiment, the operation in which the reference priority determination unit 260 determines the reference priority of the operation data using the calculation process state transition map and the overlap transition map is the first embodiment. And different.
The calculation process state transition map is a map representing a state transition until one SLO is calculated from a plurality of operation data. In the calculation process state transition map, the operational data is in the initial state, and the SLO is in the final state. The overlap transition map is a state transition map showing a process until two or more SLOs are calculated from a plurality of operation data.
<Description of operation for determining operational data reference priority>
FIG. 16 is a flowchart for explaining an example of the operation in which the reference priority determination unit 260 determines the reference priority of operational data.
The reference priority determination unit 260 starts an operation of determining the reference priority of the operational data using the addition or update of data in the SLO definition data storage unit 240 as a trigger (S401).
For example, the reference priority determination unit 260 periodically polls the SLO definition data storage unit 240 to detect the update of the SLO definition data storage unit 240, and whether there is an addition / update to the stored SLO definition data. Whether or not may be detected. Alternatively, the system monitor may clearly indicate the timing for determining the reference priority.
The reference priority determination unit 260 acquires one SLO definition data corresponding to the SLO identifier stored in the SLO definition data storage unit 240 (S402).
Next, the reference priority determination unit 260 creates a calculation process state transition map for the SLO definition data acquired in S402 (S403).
The reference priority determination unit 260 determines whether or not a duplicate transition map is already stored (S404).
When the overlapping transition map is not stored (NO in S404), the reference priority determining unit 260 stores the calculation process state transition map created in S403 as the overlapping transition map (S405), and returns to the step of S402.
When the overlapping transition map is already stored (YES in S404), the reference priority determination unit 260 detects the overlap between the calculation process state transition map created in step S403 and the stored overlapping transition map. The overlapping transition map is overwritten (S406).
The reference priority determination unit 260 repeats the operations of S402 to S406 for all the SLO definition data stored in the SLO definition data storage unit 240, and sequentially overwrites the overlapping transition map (S407).
Next, the reference priority determination unit 260 refers to the duplicate transition map and determines the reference order of the operational data. Specifically, the reference priority determination unit 260 determines the reference priority of each operation data based on the addition of the number of each state that passes from a certain operation data to the SLO calculation value (S408).
The reference priority determination unit 260 transmits data in which the operation data identifier and the reference priority are associated to the storage location control unit 270 (S409).
Here, details of the operation step (S403) for creating the calculation process state transition map will be described with reference to FIG. 8A, FIG. 17A, and FIG.
As already described, FIG. 8A is a diagram illustrating an example of SLO definition data.
FIG. 17A is a diagram illustrating an example of a calculation process state transition map corresponding to the SLO definition data illustrated in FIG. 8A.
FIG. 18 is a flowchart showing an example of processing for creating a calculation process state transition map from SLO definition data.
The reference priority determination unit 260 creates a final state from the SLO definition data (S4031). The final state is a state having, as attributes, the name of the type of SLO, the identifier of operation data necessary for calculating the value of SLO, and the calculation formula in the SLO definition data. In the calculation process state transition map shown in FIG. 17A, N105 corresponds to the final state.
The reference priority determination unit 260 creates an initial state for each operational data identifier necessary for calculating the SLO value (S4032). The initial state is a state having, as attributes, the operational data acquisition range and the operational data type. In the calculation process state transition map shown in FIG. 17A, N101-N104 corresponds to the initial state.
The reference priority determination unit 260 connects the initial state and the final state, and creates a calculation process state transition map (S4033).
Next, details of the step (S406) of overwriting the overlapping transition map will be described with reference to FIGS. 17A, 17B, 19, and 20. FIG.
As already described, FIG. 17A is a diagram showing a calculation process state transition map corresponding to the SLO definition data shown in FIG. 8A.
FIG. 17B is a diagram illustrating an example of a calculation process state transition map corresponding to the SLO definition data illustrated in FIG. 8B.
FIG. 19 is a diagram illustrating an example of an overlapping transition map created by superimposing calculation process state transition maps shown in FIGS. 17A and 17B.
FIG. 20 is a flowchart illustrating an example of a process for overwriting a duplicate transition map.
The reference priority determination unit 260 overwrites the duplicate transition map with the stored duplicate transition map and the calculation process state transition map created from the SLO identifier as inputs.
The reference priority determination unit 260 extracts one state from the calculation process state transition map (S4061).
The reference priority determination unit 260 examines whether or not the state extracted in step S4061 and the state that matches the state are in the overlapping transition map (S4062).
If there is no matching state (NO in S4062), the reference priority determination unit 260 proceeds to (S4066).
If there is a matching state (YES in S4062), the reference priority determination unit 260 proceeds to step (S4063).
For example, when the reference priority determination unit 260 extracts the state of N106 from the calculation process state transition map shown in FIG. 17B, the state of N106 matches the state of N101 in the overlapping transition map shown in FIG. 17A. The reference priority determination unit 260 examines whether or not the states connected to the output side links in the matched state (for example, N101 and N106 here) match each other (S4063).
In the specific examples shown in FIGS. 17A and 17B, the reference priority determination unit 260 matches the state N105 connected to the output side link of N101 and the state N108 connected to the output side link of N106. Consider whether or not to do so.
If they match (YES in S4063), the reference priority determination unit 260 proceeds to step S4067 without updating the overlapping transition map.
If they do not match (NO in S4063), the reference priority determination unit 260 compares the inclusion relations of arbitrary attributes in the respective states (here, for example, N105 and N108) (S4064). Here, the reference priority determination unit 260 compares the inclusion relations of the calculation formulas, for example. In the calculation formula of N105 and the calculation formula of N108, the calculation formula of N108 is included in the calculation formula of N105. Therefore, the reference priority determination unit 260 sets the included state (N108) to the output side link of the original state (N101 = N106) and the input side link of the included state (N105). (S4065).
When the state extracted from the calculation process state transition map is not in the overlapping transition map (NO in S4062), the reference priority determination unit 260 adds the state to the overlapping transition map (S4066).
The reference priority determination unit 260 extracts the next state from the calculation process state transition map, and repeats the processing of steps S4062 to S4066 (S4067).
The reference priority determination unit 260 creates an overlapping transition map as shown in FIG. 19 from the calculation process state transition map shown in FIGS. 17A and 17B.
In the above example, the reference priority determination unit 260 compares the inclusion relations of the calculation formulas. However, the reference priority determination unit 260 may use, for example, a method of comparing the inclusion relationship of SLO calculation period widths. Moreover, the reference priority determination part 260 may use the method of comparing the inclusion relation of the SLO calculation range.
In the above example, when the reference priority determination unit 260 determines that the states match in (S4063), the reference priority determination unit 260 did not update the overlapping transition map. However, when the parameter of the count number is provided for each state of the overlapping transition map and the reference priority determination unit 260 determines that the states match in (S4063), the reference priority determination unit 260 determines the count number of the matching states. May be increased. The count number can be used as a weight when determining the reference order of the operational data.
When the SLO definition data storage unit 240 is frequently updated based on such a configuration, the monitoring system 1000 according to the second embodiment has the effect of reducing the cost when assigning the reference priority to the operation data. Can be obtained.
The reason is that the reference priority determination unit 260 of the monitoring system 1000 determines the reference priority of the operational data using the calculation process state transition map and the overlap transition map.
While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. 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.
This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2011-253430 for which it applied on November 21, 2011, and takes in those the indications of all here.

1000 Monitoring System 100 IT System 110 Operation Data Transmission Unit 120 Web Server 130 WebAP Server 140 DB Server 150 Storage Server 160 Router 170 Switching Hub 200 Monitoring Device 210 Operation Data Receiving Unit 220 Operation Data History Storage Unit 221 High Speed Storage Unit 222 Low Speed Storage Unit 230 SLO input unit 231 Input console 232 SLO type storage unit 240 SLO definition data storage unit 250 SLO calculation unit 251 SLO calculation execution unit 252 SLO calculation result storage unit 260 Reference priority determination unit 270 Storage location control unit 300 Network 400 Network 500 Client Terminal

Claims

Operational data history storage means including a plurality of storage means for storing operation data values indicating information of nodes to be monitored in association with identifiers capable of identifying the operation data;
A monitoring item definition storage unit for storing a monitoring item for monitoring the system and an identifier of the operation data used for calculating an index of the monitoring item;
Counting means for counting the number of monitoring items associated with the identifier;
Selection means for selecting the storage means for storing a value of operational data associated with the identifier according to the counted number;
Including monitoring equipment.
The operational data history storage means includes at least storage means with different data reading speeds,
The monitoring device according to claim 1, wherein the selection unit stores a value of operational data associated with the large number of the counted identifiers in a storage unit having a high reading speed.
3. The monitoring apparatus according to claim 1, wherein the identifier that can identify the operation data includes a type of usage information and information that identifies a monitoring target node.
The counting means includes
For each monitoring item, create a calculation process state transition map from operation data associated with an identifier used for calculating the monitoring item index until the monitoring item index is calculated,
Created a duplicate transition map by detecting duplication between calculation process state transition maps created for each monitoring item,
The system monitoring apparatus according to any one of claims 1 to 3, wherein the number of monitoring items associated with the identifier is counted based on the created duplicate transition map.
The system monitoring apparatus according to any one of claims 1 to 4, wherein the node information includes at least one of performance information, operation information, and usage information of the node.
Including a plurality of storage means for storing values of operation data indicating information of nodes to be monitored in association with identifiers capable of identifying the operation data;
A monitoring item for monitoring the system and an identifier of the operational data used for calculating the index of the monitoring item are stored in association with each other,
Count the number of monitoring items associated with the identifier,
According to the counted number, the storage means for storing the value of the operation data associated with the identifier is selected.
Monitoring device control method.
A plurality of storage means for storing operation data values indicating information of nodes to be monitored in association with identifiers capable of identifying the operation data values;
A process for storing a monitoring item for monitoring the system and an identifier of the operation data used for calculating an index of the monitoring item in association with each other;
A process of counting the number of monitoring items associated with the identifier;
A process of selecting the storage means for storing a value of operational data associated with the identifier according to the counted number;
A program that causes a computer to execute.