JP2006344025A - Operation performance data acquisition method, performance monitor server, work server, computer, and computing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for acquiring operation performance related to execution of a program automatically. <P>SOLUTION: In the operation performance data acquisition method: a work manager 200 has a step for generating a collection instruction with respect to work application including a name of work and a name of a host; a performance monitor server 100 has a step for distributing to a grid agent host 400 having the name of the host a collection item corresponding to the name of the work included in the collection instruction; the grid agent host 400 has a step for executing the work application and collecting the operation performance data related to the distributed collection item; and the performance monitor server 100 has a step for acquiring from the grid agent host 400 the operation performance data related to the distributed collection item. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique used in a system in which computers that execute programs are dynamically allocated.

In recent years, as one of systems in which computers that execute business application programs are dynamically allocated, processing of business application programs using a grid computing system (hereinafter referred to as grid computing processing) is superposed on an inexpensive terminal. It is attracting attention because it can obtain processing performance comparable to computers. In this grid computing process, an optimal computer that executes a business application program is allocated when the business application program is executed, and each computer is used efficiently.
Under such circumstances, conventionally, an administrator or the like has identified a communication path (such as an IP address) to a computer that executes a business application program, thereby monitoring the performance of the computer and improving the performance of the computer. Is known (see, for example, Patent Document 1 and Patent Document 2).
Japanese Patent Laid-Open No. 5-242034 (paragraphs 0013 to 0027, FIG. 2) Japanese Patent Laying-Open No. 2004-302741 (paragraphs 0020 to 0042, FIG. 1)

  However, the method disclosed in Patent Document 1 does not collect operation performance data such as the number of error packets and a cache hit rate for a computer that executes a business application program, and therefore cannot sufficiently monitor the performance of the computer.

  Further, in the method disclosed in Patent Document 2, the computer that executes the business application program is operating in consideration of the CPU usage rate, the memory usage rate, and the hard disk usage rate in the grid computing process. Performs arithmetic processing (calculation for data analysis related to biotechnology, physics, design, finance, etc.) divided by the main processor. However, for example, since the characteristics of various server functions (Web server, application server, database server, etc.) in the Web shopping system are not taken into consideration, the performance of the server in such a system cannot be sufficiently monitored.

  Further, in the method disclosed in Patent Document 1, even if operating performance data is collected, after a business application program is assigned to a computer, a user uses a business application program monitoring server as a monitoring target. It is necessary to manually set the communication path to the computer. Therefore, setting is difficult when there are many computers that execute business application programs.

  Therefore, an object of the present invention is to provide a technique for solving the above-described problems and automatically acquiring operation performance related to program execution.

  The present invention was devised to solve the above-described problem, and in an operation performance data acquisition method, a plurality of computers that execute various programs, the program is allocated to the computer, and the program is assigned to the allocated computer. A computing system comprising: a business server that requests execution of the program; and a performance monitoring server that collects operational performance data related to the execution environment of the program from the computer and monitors the performance of the computer based on the operational performance data The operation performance data acquisition method for acquiring the operation performance data, wherein the business server includes business information indicating the assigned program and computer identification information for identifying a computer executing the program. Including the step of generating a collection instruction for the program, the performance monitoring A server reads out a collection item corresponding to the business information included in the collection command from a memory and distributes the collected item to a computer having the computer identification information, and the computer has an operation performance related to the distributed collection item. A step of collecting data, and the performance monitoring server has a step of acquiring operation performance data relating to the collection item from a computer that has collected the operation performance data relating to the collection item by distributing the collection item. It was.

  According to the present invention, it is possible to automatically acquire operation performance related to program execution.

Next, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
(Computing system configuration)
FIG. 1 is a diagram illustrating a configuration example of a computing system according to an embodiment of the present invention. As shown in FIG. 1, the computing system 1 includes a performance monitoring server 100, a business manager (business server) 200, a grid manager (grid server) 300, and a plurality (three in the figure) of grid agent hosts 400. Are connected via the network 500. This network 500 is, for example, the Internet. The computing system 1 is a grid computing system that collects operational performance data related to execution of business applications from a plurality of grid agent hosts 400 that execute predetermined business application programs (hereinafter simply referred to as business applications).
In this specification, various programs including business applications may be realized by objects, threads, and processes.

Each of these devices 100, 200, 300, and 400 is a general computer (computer), for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and an HDD. (Hard Disk Drive), KB / CRT (Key Board / Cathode Ray Tube), and a communication interface. In FIG. 1, the performance monitoring server 100, the business manager 200, and the grid manager 300 are shown as independent servers, but these servers may be configured by a single computer.
Below, the function of each apparatus 100-400 is demonstrated in detail.

[Performance monitoring server functions]
The performance monitoring server 100 collects operation performance data related to the execution environment of the business application from a plurality of grid agent hosts 400 that execute the business application, and monitors the performance of the grid agent host 400 based on the operation performance data. It mainly includes a control means (CPU or the like) 110 and a storage means (RAM or the like) 120.

  The control unit 110 includes a business execution notification management unit 111, a collection item distribution unit 112, and an operation performance data acquisition unit 113 that function according to program execution. The storage unit 120 includes a collection item list 121, a collection host, and the like. A list 122 and operating performance data 123 are stored.

The business execution notification management unit (collection command acquisition means) 111 receives a collection start request (collection command), a collection completion request, and the like from the business manager 200 as commands related to the acquisition timing of operation performance data.
The collection start request means the start of execution of a business application, and includes a business name to be executed (business information) and a name (host name: computer identification information) for identifying the grid agent host 400 to be used.
The collection completion request includes the business name for completing collection and the name (host name) of the grid agent host 400 used.

  The collection item distribution unit (collection item distribution means) 112 is a grid agent host 400 (business application is identified by the host name included in the collection start request) that collects performance performance data collection items predetermined for the business name. It is distributed to the assigned grid agent host 400).

  The operation performance data acquisition unit (operation performance data acquisition means) 113 specifies the collection items distributed from the grid agent host 400 to which the collection items of operation performance data have been distributed (the grid agent host 400 to which the business application is assigned). Operation performance data is acquired and stored in the storage means 120.

The collection item list 121 is information related to collection items of business applications and operation performance data to be collected.
The collection host list 122 is information indicating from which grid agent host 400 the collection items of the operation performance data are collected.
The operation performance data 123 is data obtained by collecting hardware, OS, and program operation information of the grid agent host 400. The operation performance data 123 may be associated with the business application together with the collection item list 121 and the like.

[Functions of business manager]
The business manager 200 assigns various business applications to the plurality of grid agent hosts 400, requests the assigned grid agent host 400 to execute the business applications, and collects the assigned business applications to the performance monitoring server 100. It notifies a start request (collection command), and includes a business execution control unit 210 and a business program DB 220.

  The business execution control unit (collection command generation means) 210 includes a business application business name (business information) and a collection start request (host name: computer identification information) for identifying the grid agent host 400 to be used (host name: computer identification information). Collection instruction). The business execution control unit 210 transmits and receives the following necessary commands and data to and from the performance monitoring server 100, the grid manager 300, and the grid agent host 400 in order to execute business applications. .

The business execution control unit 210 notifies the grid manager 300 of an instruction (host acquisition request) for requesting determination of the grid agent host 400 that executes the business application.
The business execution control unit 210 distributes a business application execution request (business execution) and a business program (file) for executing the business application to the determined grid agent host 400.
The business execution control unit 210 notifies the performance monitoring server 100 of a collection start request (collection command), a collection completion request, and the like as commands related to operation performance data acquisition timing.
The business program DB 220 stores business program data and the like.

[Grid Manager functions]
The grid manager 300 determines a grid agent host 400 that executes a business application (executes a business program), and includes an agent management unit 310 and a host information list DB 320.

When the agent management unit 310 receives a host acquisition request from the business manager 200, the agent management unit 310 selects (assigns) an optimal grid agent host 400 from a host information list that is data indicating a list of grid agent hosts 400.
The host information list DB 320 stores a host information list.

[Grid agent host functions]
The grid agent host (computer) 400 executes a business application (executes a business program), and includes a business program execution unit 410, a collection item management unit 420, an operation performance data collection unit 430, and a collection item list. DB440 is comprised.

The business program execution unit 410 executes a business program distributed from the business manager 200 and returns a predetermined processing result to the business manager 200.
The collection item management unit (collection item management means) 420 acquires the collection items distributed from the performance monitoring server 100 and notifies the performance monitoring server 100 of the performance data related to the acquired collection items. The collection item management unit 420 writes the collected collection items to the collection item list DB 440 and reads the collection items of the operation performance data to be collected from the collection item list DB 440.

The operation performance data collection unit (operation performance collection means) 430 periodically collects operation performance data related to the execution environment of the business application based on the collection items written in the collection item list DB 440.
The collection item list DB 440 stores the collection items distributed from the performance monitoring server 100.

[Specific example of collection item list]
FIG. 2 is a diagram illustrating an example of the collection item list. As illustrated in FIG. 2, the collection item list 121 includes a business name 2001, a collection item 2002, a status 2003, and a data ID 2004.
All the business applications specified by the business name 2001 indicate those executed by the business manager 200. For example, in FIG. 2, the business manager 200 executes three businesses specified by the business names of “Web server A”, “application server B”, and “database server C”. In this case, the three grid agent hosts 400 each function as “Web server A”, “application server B”, and “database server C” by executing the business programs specified in the three businesses. It means to do. The business name 2001 may be anything that can uniquely identify a business application, and may be a value such as a business ID composed of a numerical value or a character string, for example.

  The collection item 2002 indicates what kind of contents of operation performance data is collected for the business application specified by the business name 2001, and is predetermined. In FIG. 2, the number of error packets, the transaction time, and the cache hit rate are illustrated. However, for example, contents such as a CPU usage rate, a memory usage rate, and a hard disk usage rate may be collected items.

A status 2003 indicates a collection status for the corresponding business application. For example, the initial state is set to “uncollected”, “collecting” when collection is started, and “collection completed” when collection is completed.
The data ID 2004 uniquely assigns data other than the collection item 2002 and the collection item list 121 to each collection item.
In the collection item list 121, an event notification condition based on conditions such as a collection interval, a collection period, and a collection item exceeding a threshold (upper limit value) may be set.

[Specific example of collection host list]
FIG. 3 is a diagram illustrating an example of the collection host list. The collection host list 122 has a data ID 3001 and a host name 3002 as shown in FIG.
The data ID 3001 is the same ID as the data ID 2004 of the collection item list 121 (see FIG. 2). Using this, it is possible to specify which collection item (see FIG. 2) the data is related to.
The host name 3002 identifies the grid agent host 400 that has collected the performance data. As the host name 3002, as shown in FIG. 3, in addition to values such as “Host A” and “Host C”, values such as an IP address and a MAC address may be designated.

[Specific examples of operational performance data]
FIG. 4 is a diagram illustrating an example of operation performance data. As shown in FIG. 4, the operational performance data 123 includes a data ID 4001, a host name 4002, a time 4003, and a value 4004.
The data ID 4001 is the same ID as the data ID 2004 of the collection item list 121 (see FIG. 2). Using this, it is possible to specify which collection item (see FIG. 2) the data is related to.
The host name 4002 has the same contents as the host name 3002 in the collection host list 122 (see FIG. 3), and identifies the grid agent host 400.
The time 4003 is a time (for example, date and time) when data corresponding to the data ID 4001 is collected.
A value 4004 is a specific value (such as the number of error packets) that represents operation performance data for the corresponding data.

  In the example shown in FIG. 4, for example, the grid agent host 400 specified by “Host A” has 10 minutes as the value of the data collection item (transaction time: see FIG. 2) specified by the data ID “0002”. It is shown that “223”, “232”, and “231” are collected as operation performance data at intervals (see the item of time 4003). “Host B” and “Host C” are values of data collection items (cache hit rate: see FIG. 2) specified by the data ID “0003”, and “96” and “82” are operating performance data. , Each collected in the same time.

[Specific example of business program DB]
FIG. 5 is a diagram illustrating an example of the business program DB. 5 includes a business name 5001, a host name 5002, and business program data 5003.
The business name 5001 has the same contents as the business name 2001 in the collection item list 121 (see FIG. 2).
The host name 5002 has the same contents as the host name 3002 in the collection host list 122 (see FIG. 3), and identifies the grid agent host 400 that executes the business program.
The business program data 5003 represents information (execution file) that defines the relevance and execution order of programs such as processes and threads constituting the corresponding business application.

  In FIG. 5, for example, it is indicated that the business name of “application server B” is assigned by the grid manager 300 to the grid agent host 400 with the host name “HostA”. When the business application identified by the business name 5001 is not assigned to the grid agent host 400 (business name “Web server”), a blank (NULL) is set in the host name 5002.

[Specific example of host information list DB]
FIG. 6 is a diagram illustrating an example of the host information list DB. In FIG. 6, it has a host name 6001 and a status 6002.
The host name 6001 has the same contents as the host name 3002 of the collection host list 122 (see FIG. 3), and identifies the grid agent host 400 managed by the grid manager 300.
A state 6002 indicates whether the corresponding grid agent host 400 is used for a business application. This state 6002 is set to, for example, “available”, “in use”, “unusable”, etc. by the agent management unit 310 (see FIG. 1).

[Specific example of collection item list DB]
FIG. 7 is a diagram illustrating an example of the collection item list DB. In FIG. 7, a business name 7001, a collection item 7002, and a data ID 7003 are included.
The business name 7001, the collection item 7002, and the data ID 7003 have the same contents as the business name 2001, the collection item 2002, and the data ID 2004 in the collection item list 121 (see FIG. 2), and the collection item list 121 (see FIG. 2). The collection items of the operation performance data distributed to the grid agent host 400 are shown.

[Operational data collection item setting operation]
Next, an operation related to setting of collection items of operation performance data in the computing system 1 will be described with reference to FIG. FIG. 8 is a sequence diagram showing a flow of setting operation performance data collection items linked to execution of a business application in the business manager.

  In step 8001, the business manager 200 requests the grid manager 300 (agent management unit 310) to determine the grid agent host 400 that executes the business application by the business execution control unit 210 as a host acquisition request via the network 500. To do. This request is made, for example, when a plurality of business applications are executed by an operation of the business manager 200 by an operator.

  In Step 8002, the grid manager 300 receives the host acquisition request, refers to the host information list DB 320 by the agent management unit 310, and assigns the optimal grid agent host 400 for executing the business application. Specifically, the agent management unit 310 refers to the host information list DB 320 (see FIG. 6), sets the status 6002 of the grid agent host 400 to “in use”, and sets the host name (the value specified by 6001). ) To the business manager 200 (business execution control unit 210) via the network 500.

  In step 8003, the business manager 200 uses the business execution control unit 210 to set the notified host name to the host name 5002 (see FIG. 5) of the business program DB 220. Then, the business execution control unit 210 generates a collection start request (collection command), and notifies the performance monitoring server 100 (business execution notification management unit 111) of the generated collection start request via the network 500.

  In step 8004, in the performance monitoring server 100, the business execution notification management unit 111 instructs to create the business name and host name values included in the received collection start request and conditions for the collection items of the operational performance data ( Is also output to the collection item distribution unit 112.

  In step 8005, the collection item distribution unit 112 (performance monitoring server 100), from the collection item list 121 (see FIG. 2) of the storage unit 120, collects items, business names, and data based on the business names acquired by the output. The ID values are extracted, and these values are distributed via the network 500 to the grid agent host 400 (collection item management unit 420) having the host name acquired in Step 8004. Further, the grid agent host 400 is requested to set collection items (collection item settings).

  When the collection item distribution is normally executed, in step 8006, the grid agent host 400 (collection item management unit 420) notifies the performance monitoring server 100 (collection item distribution unit 112) of the success of the process. Similarly, in step 8007, the collection item distribution unit 112 notifies the business execution notification management unit 111 of the success of the process, and in step 8008, the business execution notification management unit 111 (performance monitoring server 100) displays the business manager 200 ( The business execution control unit 210) is notified of the success of the processing.

At this time, the grid agent host 400 (collection item management unit 420) writes the value of the collected collection item received in the collection item list DB 440.
In addition, the performance monitoring server (business execution notification management unit 111) changes the status 2003 of the collected collection item to “collecting” in the collection item list 121 (see FIG. 2), and also collects the collection host list 122 (FIG. 3). In the reference), the host name 3002 is set to a value indicating the grid agent host 400 to which a value such as a collection item is distributed.

In step 8009, the business manager 200 (business execution control unit 210) makes a business application execution request (business execution) to the grid agent host 400 (business program execution unit 410) having the host name set in the business program DB 220. And a business program for executing the business application are distributed via the network 500.
In step 8010, the grid agent host 400 receives the execution request and executes the business program by the business program execution unit 410 to notify the business manager 200 (business execution control unit 210) of a predetermined processing result.

  In step 8011, the business manager 200 (business execution control unit 210) requests the performance monitoring server 100 (business execution notification management unit 111) to complete the collection of operation performance data of the business application via the network 500 (step 8011). Collection completion request). This collection completion request is notified to the performance monitoring server 100 when predetermined, for example, when an end command is input in the business manager 200 or a business application is completed in the grid agent host 400.

  As a result, the performance monitoring server 100 (business execution notification management unit 111) receives the collection completion request, and the value of the state 2003 in the collection item list 121 (see FIG. 2) based on the business name included in the collection completion request. Is changed to “collection complete”, and “NULL” is set as the host name value in the collection host list 122 (see FIG. 3).

  In step 8012, the performance monitoring server 100 (the job execution notification management unit 111) sets the grid agent host 400 (having a value before changing the host name value of the collection host list 122 (see FIG. 3) to “NULL”. The collection item management unit 420) is requested to delete data such as a collection item to be collected via the network 500 (collection item deletion).

  In step 8013, the grid agent host 400 receives the collection item deletion request, and the collection item management unit 420 deletes data such as the collection item from the collection item list DB 440. The monitoring server 100 (business execution notification management unit 111) is notified of the success of the process. In step 8014, the performance monitoring server 100 (business execution notification management unit 111) notifies the business manager 200 (business execution control unit 210) of the success of the process.

  In step 8015, the business manager 200 (the business execution control unit 210) receives the notification of the success of the processing and releases the grid agent host 400 used for the business application execution (collection of operation performance data) via the network 500. To the grid manager 300 (host release).

  In step 8016, the grid manager 300 receives the host release notification, and changes the state 6002 of the host information list DB 320 (see FIG. 6) to “available” when the agent management unit 310 finishes performance monitoring normally. Then, the success of the process is notified to the business manager 200 (business execution control unit 210).

  In the above-described flow, the business server 200 manages a pair of the business application and the grid agent host 400 that executes the business application, and notifies the performance monitoring server 100 of the collection timing of the operation performance data. Therefore, the performance monitoring server 100 can automatically set and release the necessary performance data collection items for the grid agent host 400 that executes the business application. This completes the operation related to the setting of the collection item of the operation performance data.

[Operation performance data acquisition operation]
Next, operations related to acquisition of operation performance data in the computing system 1 will be described with reference to FIG. FIG. 9 is a sequence diagram illustrating a flow in which the performance monitoring server acquires operation performance data.

  In step 9001, when the status 2003 of the collection item list 121 (see FIG. 2) indicates “being collected”, the performance monitoring server 100 periodically performs the collection host list 122 ( The corresponding data ID 3001 is notified to the grid agent host 400 specified by the value of the host name 3002 in FIG. 3), and the acquisition of the operation performance data is requested (data acquisition).

  In step 9002, the grid agent host 400 receives the data ID notification from the collection item management unit 420 and requests the operation performance data collection unit 430 to collect the operation performance data (data acquisition). Specifically, the collection item management unit 420 reads the collection item 7002 (see FIG. 7) corresponding to the received data ID from the collection item list DB 440 and notifies the operation performance data collection unit 430.

In Step 9003, the operation performance data collection unit 430 collects the operation performance data at that time and notifies the collection item management unit 420 of the collection result (operation performance data).
In step 9004, the grid agent host 400 (collection item management unit 420) notifies the performance monitoring server 100 of the collection result (operation performance data) via the network 500.

  Thereafter, the performance monitoring server 100 acquires the operation performance data by the operation performance data acquisition unit 113 and accumulates the acquired operation performance data in the storage unit 120. The accumulated operation performance data 123 is displayed on a display unit such as a computer display (not shown). As a result, the administrator or the like can search the location of the bottleneck by checking the contents of the operation performance data 123 and monitoring the value of the collection item of “number of error packets”, for example. As a result, the performance of each grid agent host 400 can be improved.

(Operation of the performance monitoring server 100)
[Actions related to the collection item distribution part]
Next, the operation for distributing the collection items of the operation performance data in the performance monitoring server 100 will be described with reference to FIG. 10 (refer to FIG. 1 as appropriate). FIG. 10 is a flowchart showing processing of the collection item distribution unit of the performance monitoring server.
The collection item distribution unit 112 acquires a combination of a business name and a host name from the business execution notification management unit 111 (step S1), and acquires the acquired business name from the collection item list 121 (see FIG. 2) in the storage unit 120. Search (search from the top of the list in order) (step S2).

  Then, the collection item distribution unit 112 determines whether or not the acquired business name matches the value of the record in the collection item list 121 (step S3). If the acquired business name matches the value of the record (step S3: Yes), the collection item distribution unit 112 executes the business name with the matched value based on the host name acquired in step S1. On the other hand, the collection items of the records matched by the search are distributed via the network 500 (step S4).

  On the other hand, when the business name does not match the value of the record (step S3: No), the collection item distribution unit 112 determines whether or not all searches have been completed (step S5). When all the searches are finished (step S5: Yes), the collection item distribution unit 112 finishes the process. On the other hand, when there is an unsearched record (step S5: No), the process returns to step S2. repeat.

[Operations related to operation performance data acquisition unit]
Next, the operation performance data acquisition operation in the performance monitoring server 100 will be described with reference to FIG. 11 (see FIG. 1 as appropriate). FIG. 11 is a flowchart showing the processing of the operation performance data acquisition unit of the performance monitoring server.

  The operation performance data acquisition unit 113 determines whether or not a certain time has elapsed (step S11), and repeats the determination until the certain time has elapsed (step S11: No). When the certain time has elapsed (step S11: Yes) Then, the host name value indicated at the top of the collection host list 122 (see FIG. 3) is acquired (step S12). Here, the fixed time (fixed interval) means a collection interval of operation performance data, and is predetermined in, for example, a definition file stored in a memory. This collection interval may be given as one content included in the collection item.

  Then, the operation performance data acquisition unit 113 refers to the collection host list 122 (see FIG. 3), and the position of the value specifying the host name in the acquisition order has reached the end of the collection host list 122 (to be acquired). It is determined whether or not the host name has been lost (step S13). When the end of the collection host list 122 is reached (step S13: Yes), the operation performance data acquisition unit 113 returns to step S11.

  On the other hand, when the end of the collection host list 122 has not been reached (acquisition of the host name has succeeded) (step S13: No), the operation performance data acquisition unit 113 applies the acquired value of the host name to the grid agent host 400. Then, by notifying the data ID 3001 (see FIG. 3) associated with the acquired value, the acquisition of the operation performance data is requested (step S14). That is, the operation performance data acquisition unit 113 instructs the grid agent host 400 which collection items of the operation performance data are collected by the data ID 3001 (see FIG. 3).

  Subsequently, the operation performance data acquisition unit 113 adds the acquired operation performance data to the storage unit 120 (step S15). At this time, the values specified by the data ID 4001, the host name 4002, the time 4003, and the value 4004 are added to the operation performance data 123 of the storage unit 120 as shown in FIG.

  Then, the operation performance data acquisition unit 113 determines whether or not the next host name exists in the collection host list 122 (step S16). As a result, when there is a next host name (step S16: Yes), the operation performance data acquisition unit 113 acquires the next host name from the collection host list 122 (step S17), and returns to step S13. On the other hand, when there is no next host name (step S16: No), the operation performance data acquisition unit 113 ends the process.

  As described above, the computing system 1 according to the present embodiment is configured to include the performance monitoring server 100 that acquires operation performance data related to the execution environment of the business application based on the collection command. The person or the like can automatically acquire the performance data without being aware of the computer that actually executes the business application. As a result, it becomes possible to tune business applications and prevent failures in advance.

As mentioned above, although preferred embodiment was illustrated about this invention, this invention is not limited to above-described embodiment, In the range which does not deviate from the meaning of this invention, it can change suitably. For example, in the present embodiment, the case of a grid computing system configured by each grid agent host (computer) 400 has been described. However, under the environment of a blade server in which a plurality of CPUs execute processing under a single OS. May be applied.
In the present embodiment, the collection item 2002 (see FIG. 2) of the operation performance data stored in the performance monitoring server 100 is determined in advance, but this collection item is stored in the performance monitoring server 100. Means for setting may be provided.

It is a figure which shows the structure of the computing system which concerns on embodiment of this invention. It is a figure which shows an example of the collection item list | wrist shown in FIG. FIG. 2 is a diagram illustrating an example of a collection host list illustrated in FIG. 1. It is a figure which shows an example of the operation performance data shown in FIG. It is a figure which shows an example of the business program DB shown in FIG. It is a figure which shows an example of host information list DB shown in FIG. It is a figure which shows an example of the collection item list DB shown in FIG. FIG. 10 is a sequence diagram showing a flow of setting operation performance data collection items linked to execution of a business application in a business manager. It is a sequence diagram which shows the flow from which a performance monitoring server acquires operation performance data. It is a flowchart which shows the process of the collection item distribution part of a performance monitoring server. It is a flowchart which shows the process of the operation performance data acquisition part of a performance monitoring server.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Computing system 100 Performance monitoring server 110 Control means 111 Business execution notification management part (collection command acquisition means)
112 Collection Item Distribution Department (Collection Item Distribution Unit)
113 Operation performance data acquisition unit (operation performance data acquisition means)
120 Storage Means 121 Collection Item List 122 Collection Host List 123 Operation Performance Data 200 Business Manager (Business Server)
210 Business execution control unit (collection command generation means)
220 Business Program DB
300 Grid manager 310 Agent management unit 320 Host information list DB
400 Grid agent host (computer)
410 Business program execution unit 420 Collection item management unit (collection item management means)
430 Operation performance data collection unit (operation performance collection means)
440 Collection Item List DB
500 networks

Claims (7)

A plurality of computers that execute various programs;
A business server that assigns the program to the computer and requests the assigned computer to execute the program;
The operational performance data acquisition is performed using a computing system including a performance monitoring server that collects operational performance data related to the execution environment of the program from the computer and monitors the performance of the computer based on the operational performance data. An operation performance data acquisition method for
The business server includes business information indicating the assigned program and computer identification information for identifying a computer that executes the program, and generating a collection instruction for the program,
The performance monitoring server has a step of reading a collection item corresponding to business information included in the collection command from a memory and distributing it to a computer having the computer identification information;
The computer has a step of collecting operational performance data relating to a distributed collection item;
An operation performance data acquisition method comprising the step of acquiring operation performance data related to the collection item from a computer that has collected the operation performance data related to the collection item by distributing the collection item. .   2. The operation performance data acquisition method according to claim 1, wherein the operation performance data acquisition step acquires the operation performance data at regular intervals. A performance monitoring server that collects operation performance data related to the execution environment of the program from a plurality of computers that execute various programs, and monitors the performance of the computer based on the operation performance data,
Collection instruction acquisition means for acquiring the collection instruction of the program, including business information indicating the program, and computer identification information for identifying a computer executing the program;
A collection item distribution means for reading out a collection item corresponding to the business information included in the collection command from the memory and distributing it to the computer having the computer identification information;
Operating performance data acquisition means for acquiring operating performance data related to the collected item from a computer that has collected operating performance data related to the collected item by distributing the collected item;
A performance monitoring server comprising: A business server that assigns various programs to a plurality of computers and requests the assigned computers to execute the programs,
4. The performance monitoring according to claim 3, comprising business information indicating the assigned program and computer identification information for identifying a computer that executes the program, generating a collection command for the program, A business server comprising collection command generation means for notifying a server. A computer that executes various assigned programs,
Operational performance collection means for collecting operational performance data related to the execution environment of the program;
Collection item management means for acquiring collection items distributed from the performance monitoring server according to claim 3 and notifying the performance monitoring server of operational performance data related to the acquired collection items;
A computer comprising: A plurality of computers that execute various programs;
The program is assigned to the computer, the assigned computer is requested to execute the program, and includes business information indicating the assigned program, and computer identification information for identifying a computer that executes the program, A business server that generates program collection instructions;
A performance monitoring server that collects operation performance data related to the execution environment of the program from the computer based on the collection command and monitors the performance of the computer based on the operation performance data. And
The performance monitoring server
A collection item distribution means for reading a collection item corresponding to the business information included in the collection command from a memory and distributing the collection item to a computer having the computer identification information;
Operating performance data acquisition means for acquiring operating performance data related to the collected item from a computer that has collected operating performance data related to the collected item by distributing the collected item;
A computing system comprising: A plurality of computers that execute various assigned programs;
A performance monitoring server that collects operational performance data related to the execution environment of the program from the computer and monitors the performance of the computer based on the operational performance data;
The performance monitoring server
Collection instruction acquisition means for acquiring the collection instruction of the program, including business information indicating the program, and computer identification information for identifying a computer executing the program;
A collection item distribution means for reading out a collection item corresponding to the business information included in the collection command from the memory and distributing it to the computer having the computer identification information;
Operating performance data acquisition means for acquiring operating performance data related to the collected item from a computer that has collected operating performance data related to the collected item by distributing the collected item;
A computing system comprising:

Images