WO2008007669A1 - Dispositif et procédé de gestion de système informatique - Google Patents
Dispositif et procédé de gestion de système informatique Download PDFInfo
- Publication number
- WO2008007669A1 WO2008007669A1 PCT/JP2007/063735 JP2007063735W WO2008007669A1 WO 2008007669 A1 WO2008007669 A1 WO 2008007669A1 JP 2007063735 W JP2007063735 W JP 2007063735W WO 2008007669 A1 WO2008007669 A1 WO 2008007669A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- request
- web
- computer system
- log
- usage
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3438—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment monitoring of user actions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0283—Price estimation or determination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3495—Performance evaluation by tracing or monitoring for systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/875—Monitoring of systems including the internet
Definitions
- the present invention relates to a computer system management device, and more particularly to a technique for extracting information useful for computer system management.
- Japanese Patent Application Laid-Open No. 2 005-25372 discloses a technique for accurately estimating the CPU consumption time of each process.
- JP 2005-216078 A simply specifies the process that is the starting point of transaction processing, and all processes involved in transaction processing are specified. It discloses the technology to obtain the profile of CPU consumption, transmission data size and reception data size.
- Japanese Patent Application Laid-Open No. 2005-64971 discloses a technique for estimating the load distribution of the average load amount and the average resource usage amount at each node !.
- One of the purposes of managing resource usage and usage rate is to calculate billing data for billing users of computer system usage fees.
- the charge is generally the amount of resources used, specifically consumption Calculated from CPU time, usage of main memory, number of input and output, and number of print sheets used.
- Japanese Patent Application Laid-Open Publication No. 6-95924 discloses a technique for a process to grasp the charge CPU time using CPU.
- Japanese Patent Laid-Open Publication No. 2005-31870 discloses a plurality of process groups when executing a job. Even when a job is generated, the operation system is made aware of the execution of those multiple process groups as a single job, thereby disclosing the technology for charging appropriately.
- One of the other purposes of managing resource usage and utilization is to measure or predict the load and capacity of current and future computer systems, and use the results to measure the size and use of computer systems. It is to optimize the configuration. For example, if it is anticipated that the load on the computer system will increase in the future, resources may be added to the computer system, thereby avoiding the lack of capacity of the computer system.
- a computer system may be configured of a plurality of servers in a hierarchical structure (for example, in Japanese Patent Application Laid-Open No. 2001-290641, a three-layer configuration of a web server, an application server, and a database server is adopted Management of resource usage and utilization, it is possible to properly determine which level of server to add.
- Japanese Patent Application Laid-Open No. 2006-24017 discloses a system that analyzes the transition of the use state of resources of a computer system and mechanically predicts future trends.
- CPU utilization log and transaction log are obtained from the Web application server, and CPU utilization for each type of transaction is estimated based on CPU utilization and information on the number of transactions processed per transaction.
- Ru CPU utilization for each type of transaction is calculated by multiple regression analysis. Using the calculated CPU usage rate, the estimated processing volume for each fixed time in the future can be obtained for each type of transaction.
- the CPU utilization for each type of transaction and the estimated amount of future processing will be used to determine the configuration and settings of the software and hardware necessary to fully satisfy the application needs on the server.
- a transaction in a web application is generally a session composed of a plurality of web resources (also referred to as HTTP requests), and satisfies the ACID characteristics (atomicity, consistency, isolation, durability). (If rowed, refer to paragraphs [0023] and [0024] of JP-A-2000-207365).
- ACID characteristics atomicity, consistency, isolation, durability.
- HTTP hyper text transfer protocol
- the hyper text transfer protocol (HTTP) used on the WWW is generally considered to be unsuitable for transaction processing. This is because HTTP is based on simple processing of one request / one reply.
- HTTP is based on simple processing of one request / one reply.
- Japanese Patent Application Laid-Open No. 2000-207365 discloses a system for performing transaction processing.
- the server generates transaction identification information when a request that requires a transaction scope (a set of individual independent requests that logically compose one transaction) is sent from the client to the server, Maintain for the duration of multiple independent web requests.
- the client uses this transaction identification information to send a request to the server to request transaction completion, thereby sending a request for additional action to the server that requires transaction scope.
- Apache which is a typical software used as a web server, is a typical example (Ben 'porty Lee, Peter Why, “Apache handbook”, Ohm Co., Ltd., Nov. 25, 1997, pp. 41-57). Apache processes multiple requests arriving asynchronously for one process (or a relatively small number of processes capable of handling multiple types of web requests), while the process is persistent. It employs an architecture that executes while scheduling.
- the object of the present invention is to make it possible to appropriately manage resources by managing resources focusing on the type of Web request in a computer system that handles standard HTTP. To provide technology.
- the computer system of the present invention is a computer system that processes a web request received from a client.
- the computer system comprises: request classification log generation means for generating a request classification log in which the number of web requests processed by the computer system in each measurement period is described for each type of web request; System log generation means for generating a system log describing the use state of the computer system resource in a period, and the web request classification log and the system log are used for processing the web request one
- a resource use status calculation unit that calculates request processing resource usage, which is resource usage, for each type of Web request; According to such a computer system, it is possible to properly manage resources by calculating the request processing resource usage which is the usage of the resource used for the processing of one web request.
- the present invention is preferably applied to a computer system comprising a plurality of hierarchical servers.
- the system log generation unit generates the system log for each hierarchy of the plurality of servers
- the resource usage calculation unit calculates the request processing resource.
- the source usage is calculated for each hierarchy of the plurality of servers.
- the computer system power S a web server constituting the web layer, an application server connected to the web server and constituting the application layer, and a data layer connected to the application server.
- the system log generation unit generates the system log for each of the Web layer, the application layer, and the data layer
- the resource use state calculation unit generates the system log. It is preferable to calculate, for each of the Web layer, the application layer, and the data layer, the usage of resources used for processing of one Web request.
- the computer system further includes a load distribution apparatus that receives the Web request from the client and distributes the Web request to the Web server, the load distribution apparatus determines the time at which the Web request was processed and the load distribution apparatus. It is preferable that a request log in which the type of the web request is described for each of the web requests arriving at the computer system is generated, and the request classification log generation unit generates the request classification log from the request log. Ma ,.
- the computer system further comprises charging data generation means for generating charging data indicating a usage charge charged to the user using the request processing resource usage.
- the resource usage calculation means calculates the default resource usage, which is the usage of resources used independently of the Web request, using the request classification log and the system log.
- the billing data generation means may virtually distribute the default resource usage according to a ratio of the type of the web request to the total usage of the resource for each of the types of the web request.
- the resource usage amount is calculated, and the usage charge is calculated from the request processing resource usage amount and the virtual resource usage amount.
- the default resource usage is the number of web requests for each measurement period and for each type of web request described in the request classification log, and for each measurement period in the system log. In performing regression analysis with respect to the resource use status of the computer system described Therefore it is calculated that force preferred class.
- the computer system further determines, from the request log, the user ID of the user who has issued the web request for each of the web requests arrived at the computer system, and the type of the web request. And a user data generation unit that generates user data that describes the number of the web requests issued by the user for each type of the web request, and the charge data generation unit is imposed on one web request. It is preferable to calculate a unit price, which is a charge, from the virtual resource usage, and to calculate the usage charge from the unit price and the user data.
- the computer system further calculates, from the request processing resource usage, a usage status of the resource for each type of Web request at each predetermined time interval, and calculates the usage status of the resource calculated. It is preferable to provide a system management data generation means for displaying as a graph. In this case, it is preferable that the system management data generation unit has a function of displaying the utilization status of the resource for each type of the Web request in each of the measurement periods as the graph. Preferably, the system management data generation means has a function of displaying the utilization rate of each of the resources in one or more measurement periods as the graph for each type of the Web request.
- the system management data generation unit displays each utilization rate of each of the resources in a plurality of measurement periods as each of the types of the Web request, when displaying the utilization rate as the type of the respective Web request as the measurement. It has a function that can be realized by finding the average value using the usage rate in the period!
- the usage state of the CPU of the computer system is described in the system log for each of the predetermined time intervals, and the resource usage state calculation unit is configured to execute the processing of one Web request.
- the computer system adds the CPU time of each hierarchy of the plurality of servers to calculate the Web time. It is preferable to have a system management data generation unit that calculates request response time for each type of the Web request.
- the management apparatus is a computer that receives and processes Web requests from clients. It is a management device that manages a computer system.
- the management apparatus is a request classification log generation means (21) for generating a request classification log described for each type of the Web request, the number of Web requests processed by the computer system in each measurement period.
- a system log generation unit for generating a system log in which a use state of resources of the computer system in each measurement period is described; and the web request classification log and the system log, using the web request.
- a resource use status calculation means is provided for calculating, for each type of the Web request, a request processing resource use amount, which is a use amount of a resource used for one process. According to such a management apparatus, resource management can be appropriately performed by calculating the request processing resource usage amount, which is the usage amount of the resource used for processing of one Web request.
- a management method for managing a computer system that receives and processes a web request from a client comprises: calculating the number of web requests processed by the computer system in each measurement period; Using the request classification log and the system log, generating a request classification log described for each type, generating a system log describing resource usage of the computer system during each measurement period, and And calculating a request processing resource usage amount, which is a usage amount of resources used for the processing of one Web request, for each type of the Web request.
- a management method it is possible to appropriately manage resources by calculating the request processing resource usage which is the usage of the resource used for the processing of one Web request.
- a computer program according to the present invention is a computer program for managing a computer system that receives and processes Web requests from clients.
- the computer program generates a request classification log in which the number of web requests processed by the computer system in each measurement period is described for each type of the web request, and the computer system in each measurement period Using a request processing resource, which is a usage amount of resources used for processing of one Web request, using a step of generating a system log describing the usage state of the resource, and the request classification port and the system log. Calculate the amount for each type of web request
- the computer to execute the following steps. According to such a computer program, it is possible to properly manage resources by calculating the request processing resource usage, which is the usage of resources used for processing of one Web request.
- resource usage of web requests can be calculated for each type of web request, and the computer system can be properly managed based on the calculated usage of resources.
- FIG. 1 is a block diagram showing the configuration of a computer system according to an embodiment of the present invention.
- FIG. 2 shows the contents of the request log.
- Fig. 3 is a table showing the contents of the system log.
- FIG. 4 is a block diagram showing the software configuration of the operation management apparatus.
- FIG. 5 is a diagram showing the contents of a request classification log.
- FIG. 6A is a table showing an example of the contents of a unit price table.
- FIG. 6B is a table showing another example of the content of the unit price table.
- FIG. 7 is a diagram showing the contents of user data.
- FIG. 8A is a table showing the contents of a CPU output table.
- FIG. 8B is a diagram showing the contents of a memory output table.
- FIG. 9A is an example of a graph showing CPU utilization for each type of Web request.
- FIG. 9B is an example of a graph which displays the utilization of each resource in a specific measurement period for each type of Web request.
- FIG. 1 is a block diagram showing the configuration of a computer system 1 according to an embodiment of the present invention.
- the computer system 1 of the present embodiment receives a Web request from the client 2 and processes the received Web request.
- the Web request referred to in this specification is a message defined by HTTP, and is usually used to refer to a file stored in the computer system 1 or to start a CGI (common gateway interface). Ru.
- the web request is executed by the method described therein.
- Web requests include, for example, GET request including a GET method, PUT request including a PUT method, POST request including a POST method, and DELETE request capability including a DELETE method.
- management is performed based on the type of web request.
- the computer system 1 of the present embodiment employs a three-tier system.
- the computer system 1 includes a data layer including a load distribution device 3, a web server 4 that configures a web layer (presentation layer), an application server (AP server) 5 that configures a functional layer, and the like. And a database server (DB server) 6.
- the Web server 4 and the AP server 5 are plural, and the DB server 6 is single. The number of servers in each layer can be changed as appropriate.
- the load balancer 3 receives the web request sent from the client 2 and distributes the web request to the web server 4 using an appropriate load balancing algorithm.
- Web server 4 receives incoming web requests and uses the functions of AP server 5 to handle web requests. When using the function of the AP server 5, the web server 4 generates an AP request and causes the AP server 5 to perform desired processing.
- the AP server 5 executes business logic in response to an AP request sent from the Web server 4 and processes the processing result in the format (eg, HTML format and XML format) used on the WWW. Return to 4.
- the AP server 5 accesses the database server 6 if necessary.
- the database server 6 includes a database, and searches and updates the database in response to a request sent from the AP server 5.
- an operation management apparatus 7 is provided in the computer system 1.
- the operation management device 7 a computer device provided with an operation management communication device 7a, a storage device 7b, an arithmetic device 7c, a display device 7d, and an input device 7e is used.
- Various software for managing the computer system 1 is installed in the operation management device 7.
- various logs specifically, a request log 11 sent from the load balancer 3, and a system port sent from each of the Web server 4, the application server 5, and the database server 6 Is used.
- the system log sent from Web server 4 is 12
- the system log sent from the AP server 5 is referenced 13 and the system log sent from the DB server 6 is referenced! /.
- the request log 11 is a record regarding the web request that has arrived at the computer system 1, and typically describes the time when the web request was processed and the type of the web request.
- FIG. 2 is a diagram showing the contents of the request log 11 in the present embodiment. In FIG. 2, only the part corresponding to one web request in the request log 11 is shown.
- the request log 11 contains the IP address 11a of the client 2 who issued the Web request, the identity l ib of the client 2 (as defined in RFC 1413), and the user ID l of the client 2 who issued the Web request. lc is described.
- the identity l ib is described as "" (hyphen). This indicates that the information on the client l's identity l ib was not available.
- the user ID l lc is given to a user who has succeeded in HTTP authentication, and if the authentication fails, the user ID l lc is described as "" (hyphen).
- a time l id at which the web request was processed, a type l we of the web request (ie, a method described in the web request), and the web request The URL of the document referenced by (uniform resource locator) 1 If and the HTTP protocol version number l lg are described.
- the time at which the Web request was processed l id may be defined as the time at which the processing of the Web request has been completed, which may be defined as the time at which the Web request arrived.
- the type of Web request is GET request
- the version number of the HTTP protocol used is 1.1.
- the request log 11 includes a status code l lh indicating that the processing of the Web request was successful or unsuccessful, and the header of the data to be returned to the client 2 as a result of the processing of the Web request. The number of bytes 1 li is written.
- the request log 11 sent from the load balancing device 3 is sent from each of the Web servers 4 instead of the power S used to manage the computer system 1.
- the incoming request log may be used.
- system logs 12, 13 and 14 are records which describe the use status of resources in Web server 4, application server 5 and database server 6 for each measurement period.
- each measurement period is defined as a defined time interval (eg, 5 minutes).
- the length of the measurement period may be variable.
- FIG. 3 is a table showing the contents of the system log 12.
- the CPU usage rate and the memory usage of the corresponding Web server 4 are described in each system log 12 at intervals of 5 minutes. That is, in the system log 12, the CPU utilization and the memory usage of the corresponding Web server 4 are described for each of the measurement periods each having a length of 5 minutes.
- the system logs 13, 14 have similar contents.
- the system log 13 describes the CPU utilization and memory usage of the corresponding AP server 5 at 5-minute intervals
- the system log 14 describes the CPU utilization and memory usage of the DB server 6 at 5-minute intervals Described in
- the function of creating a system log with such contents is generally performed on a general operating system such as Windows (registered trademark of Microsoft Corporation) or UNIX (registered trademark of The Open Group). It is equipped. Using operating system functions eliminates the need to provide a special mechanism for generating system logs in the operating system, web server 4, AP server 5, and DB server 6, and reduces the implementation cost. It is valid.
- Windows registered trademark of Microsoft Corporation
- UNIX registered trademark of The Open Group
- the operation management device 7 manages the computer system 1 using the request log 11 and the system logs 12, 13 and 14 described above.
- the operation management apparatus 7 of the present embodiment has two features. The first feature of the operation management apparatus 7 of this embodiment, the operation management apparatus 7, the amount of resources consumed for processing a Web re Taesuto calculated for each type of We b request, the calculated The use of resources is used to manage the computer system 1.
- a second feature of the operation management apparatus 7 of the present embodiment is that it is possible to calculate, for each of the Web layer, the functional layer, and the data layer, the usage of resources consumed to process Web requests. The calculation of resource usage for each of the Web layer, the functional layer, and the data layer makes it possible to more appropriately grasp the state of the computer system 1.
- FIG. 4 is a block diagram showing an example of the software configuration installed in the operation management device 7.
- the storage device of the operation management device 7 includes a request classification tool 21, a system port integration tool 22, a resource usage analysis tool 23, a billing data generation tool 24, a system management data generation tool 25, and user data.
- the creation tool 26 is installed. Operation and management of the computer system 1 are performed by these software programs.
- the request classification tool 21 of the operation management unit 7 analyzes the request log 11 sent from the load distribution unit 3 and calculates the number of web requests processed by the computer system 1 for each measurement period and for the number of web requests. Calculated for each type. The calculation result is stored as the request classification log 31 in the storage device 7b.
- FIG. 5 is a table showing the contents of the request classification log 31.
- the request classification log 31 describes the number of processed web requests at intervals of 5 minutes for each type of web request.
- a web request for example, GET request
- a web request for example, P OST request
- type “R2” is 10 times processed! /, Shown by the Request Classification Log 31.
- the system log integration tool 22 includes the Web server 4, the AP server 5, and the DB server
- Web layer system log 32 is a record that describes resource usage in the entire Web layer during each measurement period.
- the system log integration tool 22 generates a web layer system log 32 by integrating the system log 12 received from each of the web servers 4.
- the system log integration tool 22 calculates the CPU utilization of the entire Web layer by averaging the CPU utilization of each Web server 4, and adds the memory utilization of each Web server 4. By doing this, the memory usage of the entire Web layer is calculated.
- the calculated CPU utilization and memory usage are described in the Web layer system log 32.
- functional layer system log 33 and data layer system log 34 are functional layer and data layer, respectively. It is a record which describes the use condition of the resource in the whole for every measurement period.
- the system log 14 of the DB server 6 may be used as the data layer system log 34 as it is.
- the resource usage analysis tool 23 is a request classification log 31, a web layer system log 32, a functional layer system log 33, and a data layer system log 34, which are used to process one web request.
- the usage amount is calculated for each type of web request for each of the web layer, functional layer, and data layer.
- the resource usage analysis tool 23 calculates the CPU time and the memory usage necessary for processing the web request for each of the web layer, the functional layer, and the data layer.
- the length of all the measurement periods is a constant value At will be described.
- the CPU usage rate for the entire Web layer is p W (i)
- the memory usage for the entire web layer is M w (i)
- the number of processed type ⁇ Rf web requests is N (i)
- ⁇ /- ⁇ () ⁇ N Rj (i) -U Rj w + At-p z w , ... (1
- ⁇ is the length of each measurement period, and p W does not process Web requests.
- the request classification log 31 is described, and the CPU utilization p w (i) and the memory usage M w (i) of the entire Web layer are described in the Web layer system log 32.
- Computer systems generally have Web requests to maintain their operation. It should be noted that some resources are used regardless of nothing. For example, on a web server running Apache, a predetermined number of processes will run forever regardless of the presence or absence of web requests. Therefore, CPU and memory are used just to keep the process executable. In addition, computer systems often run processes to manage the system, which use CPU and memory. In the following, the CPU usage rate in the situation where Web request processing is not performed (non-request processing situation) is called default CPU usage rate, and the memory usage in the situation where Web request processing is not performed is called default memory usage. .
- X is a transposed matrix of matrix X
- matrix X is expressed by the following equation:
- Rl R2 Rn Z usage M w , M W , ⁇ ⁇ ⁇ ⁇ ⁇ M w and default memory usage M w is for the Web tier
- Rl R2 Rn Z Analysis result data 36 is stored in the storage unit 7b.
- the same procedure is performed for the functional layer and the data layer.
- the CPU time and memory usage consumed by the functional layer for processing one Web request, and the default CPU usage rate and default memory usage of the functional layer are the request classification log 31 and the functional layer system log 33 Are calculated by the same procedure as described above.
- the calculated CPU time, memory usage, default CPU usage and default memory usage are stored in the storage device 7b as the analysis result data 36 of the functional layer.
- the CPU time and memory usage consumed to process one Web request of type “Rj ⁇ in the functional layer are described as U F and M F respectively, and the default CPU of the functional layer is described. Usage rate and memory usage
- CPU time and memory usage in the data layer for processing of one Web request and default CPU usage and default memory usage capacity of the data layer S, request classification log 31 and data Calculated from the tier system log 34 according to the same procedure as above.
- the calculated CPU time, memory usage, default CPU time and default memory usage are stored in the storage device 7b as data layer analysis result data 37.
- the CPU time and memory usage consumed to process one type ⁇ Rj ⁇ web request in the data layer are described as U D and M D respectively.
- Fault CPU time and memory usage are noted as p U and M u respectively.
- the length of each measurement period can also be variable.
- the following equation is established instead of the equation (la):
- Rj can be calculated by multiple regression analysis. More specifically, CPU time u w , U
- P W are a plurality of measured values p W ( ⁇ ⁇ p W (m), N (1
- memory usage M w , M w , ⁇ ⁇ ⁇ ⁇ M w and default memory usage M can be calculated by the following equation:
- Similar analysis can be performed on resources other than CPU time and memory usage. For example, similar analysis can be performed for network interface usage (usually byte / sec) and disk interface (SCSI interface) usage (usually byte / sec). You may analyze "read” and "write” separately about network interface and disk interface usage. Disk interface usage may be analyzed rather than disk interface usage. In this case, the disk interface utilization rate ⁇ uses the disk queue length W and the following formula:
- n Q / (Q + l)
- Analysis result data of the web layer, functional layer, and data layer obtained by the resource usage analysis tool 23 35, 36, 37 (ie, the web layer, functional layer for processing one web request) , CPU time and memory usage consumed by each data layer, and default CPU usage and default memory usage) are used for two purposes: first, Web layer, functional layer, and Data Layer Analysis Results Data 35, 36, 37 are used for billing the user.
- the unit price to be charged for processing of one Web request is calculated for each type of Web request, and the calculated unit cost and the number of Web requests determine the charge to be charged to each user. Be done.
- the analysis results data 35, 36, 37 of the web layer, functional layer, and data layer are used to analyze the load and capacity of the computer system 1 in terms of the type of web request. used. For example, the utilization rate of specific resources in each measurement period is analyzed for each type of web request using the analysis result data 35, 36, 37. In addition, the utilization rate of each resource in a specific measurement period is analyzed for each type of Web request using the analysis result data 35, 36, 37.
- the usage fee charged to each user should basically be determined according to the amount of resources used to process the web request issued by each user.
- One problem is that resources are consumed to maintain the operation of the computer system 1 (with or without a web request).
- resources are consumed to maintain the operation of the computer system 1 (with or without a web request).
- memory is consumed only when the process becomes executable without processing Web requests, the amount of memory used is large regardless of the presence or absence of Web requests.
- common resource usage it is desirable to charge each user for usage of resources to maintain the operation of computer system 1 (hereinafter referred to as common resource usage). There is a case.
- a unit price table 38 is prepared in the storage unit 7 b of the operation management apparatus 7 and this unit price table 38 is used for charging processing. If the length of all measurement periods is a constant value
- the unit price table 38 includes each of the Web layer, the functional layer, and the data layer. CPU unit cost and memory unit cost per measurement period (five minutes in this embodiment) are described. In the following, the CPU unit price per measurement period (5 minutes in this embodiment) in the Web layer is described as C w
- the memory cost per measurement period is described as c w .
- the unit price table 38 includes The CPU unit price per unit time and the memory unit price may be described for each functional layer and data layer.
- CPU unit price in the measurement period "i" in the Web layer when the CPU unit cost per unit time in the Web layer and c W, a A t (i) 'c W .
- the memory unit price in the measurement period "i" in the Web layer when the memory unit price per during unit time in the Web layer and c W, a A t (i) 'c W .
- the unit price table 38 includes one per measurement period.
- the web system computer system operation cost (per unit time), CPU contribution rate, and memory contribution rate are also possible to describe the web system computer system operation cost (per unit time), CPU contribution rate, and memory contribution rate. If the length of all measurement periods is fixed, the CPU unit cost per measurement period is calculated by multiplying the computer system operation cost per measurement period by the CPU contribution rate. Furthermore, the memory unit price per measurement period can be calculated by multiplying the computer system operation cost of the Web layer by the contribution ratio of the memory. On the other hand, when the length of the measurement period is variable, the CPU unit cost per unit time is calculated by multiplying the computer system operation cost per unit period by the contribution rate of the CPU. Furthermore, the memory unit price per unit time can be calculated by multiplying the computer system operation cost of the Web layer by the contribution ratio of the memory.
- Billing data generation tool 24 an analysis result data 35 of the Web layer, by using the CPU unit price C W and the memory unit price C W described in the bid table 38, a single Web request
- the unit price W (i) of the web layer for one process is determined as follows.
- the CPU time consumed to process the web request of type ⁇ Rj ⁇ ⁇ ⁇ is N (i)-U W. Also, during each measurement period, for processing web requests
- the default CPU time consumed independently is A fp w if the length of all measurement periods is a constant value A t, and if the length of measurement periods is variable, A t (i t ) -w zz.
- “Virtual CPU consumption time” in layer b is determined. Specifically, if the length of all measurement periods is constant, the Web layer of the type of measurement period “i” ⁇ Rj Web Web request Virtual CPU consumption time in CPU-Rj w (i) is the following formula:
- the virtual CPU consumption time CPU ⁇ Rj W (i) has the following formula:
- M_Rf (i) N Rj (i)-M Rj "+ M Z '
- the unit price W (i) in the web layer of one request is the following formula:
- ⁇ (l) ⁇ (l) + ⁇ (0, ⁇ ⁇ ⁇ (9)
- FIG. 7 is a table showing an example of the contents of the user data 39.
- the aggregation period is 5 minutes.
- ⁇ N (i) is
- RjUk Measurement period "i" means the number of Web requests of type ⁇ Rj ⁇ ⁇ issued by user Uk.
- the billing data generation tool 24 determines the unit price w (i) in the web layer of one web request calculated as described above and the user data 39 for each user in the measurement period ' ⁇ '.
- ⁇ means the sum of all Web request types!
- the unit price of one Web request and the charge to be charged to the user are similarly determined.
- the accounting data generation tool Honoré 24 the fee F F (i) of the functional layer, rates of the data layer F D a (i) be calculated by the following formula
- F (i) and ⁇ D (i) are Web requests in the functional layer and data layer, respectively.
- the charge data generation tool 24 is a charge to be charged to each user Uk.
- F (i) F W (i) + F F (i) + F D (i).
- the charge data generation tool 24 is a charge F (i), F w (i), F F to be charged to each user Uk.
- F D (i) is generated to describe charging data 40 and stored in the storage device 7b.
- the system management data generation tool 25 has a function for calculating resource utilization and usage in each of the Web layer, function layer, and data layer for each type of Web request and for each measurement period.
- the system management data generation tool 25 uses the analysis results data 35, 36, 37 of the web layer, the functional layer, and the data layer to determine the CPU usage rate in each layer for each type of web request and for each measurement period.
- the CPU output table 41 describing the calculated and calculated CPU utilization is stored in the storage device 7b. During the measurement period ' ⁇ ', CPU utilization in the web layer of type ⁇ Rj Web web request ⁇ W (i), in the functional layer
- FIG. 8A shows the contents of the part corresponding to the Web layer in the CPU output table 41.
- “CPU usage rate ⁇ W (i) in the Web layer of the type Rj Web request” is not calculated for one type Rf Web request.
- the system management data generation tool 25 calculates the memory usage of each of the web layer, the functional layer, and the data layer for each type of web request and for each measurement period, and calculates the memory usage.
- the described memory output table 42 is stored in the storage device 7b.
- Figure 8B Is a table showing the contents of the portion of the memory output table 42 corresponding to the web layer.
- the memory usage ⁇ W (i) in the Web layer of the type ⁇ Rj ⁇ web request during the measurement period ' ⁇ ' the memory usage ⁇ F (i) in the functional layer, the memory in the data layer
- FIG. 8B shows the contents of the portion of the memory output table 42 corresponding to the Web layer.
- the system management data generation tool 25 has a function of displaying the contents of the CPU output table 41 and the memory output table 42 by a graph. Specifically, the system management data generation tool 25 has a function of displaying resource utilization and usage for each type of Web request in each measurement period as a graph. For example, as shown in FIG. 9A, the system management data generation tool 25 displays the CPU utilization and default CPU utilization of the web layer for each type of web request in each measurement period as a graph. The system management data generation tool 25 is configured to display a similar graph of memory usage. Such a graph is useful for grasping the current state of the computer system 1.
- the system management data generation tool 25 has a function of displaying the usage status of each resource in a specific measurement period for each type of Web request.
- the system management data generation tool 25 is based on CPU utilization, memory utilization, and network utilization of the Web layer in a specific measurement period (10:00 to 10:05 in FIG. 9B). Display usage rates and disk usage rates as graphs for each type of web request.
- the total amount of memory managed by the operating system is the value managed by the normal operating system.
- the memory usage status is the memory usage rate, as shown in FIG. 9B, the vertical axis
- the utilization status of multiple resources can be displayed together in a single graph defined as “source utilization rate (%)”. Similar to the graph of FIG. 9A, the graph of FIG. 9B is useful for grasping the current state of the computer system 1.
- the system management data generation tool 25 further uses the utilization status of each resource in a plurality of measurement periods (for example, two measurement periods of 10:00 to 10:05 and 10:05 to 10:10). It can also have the ability to display each web request. Specifically, after calculating the resource utilization rate in each measurement period, calculating the average value of them can be displayed as a graph in the same manner as above.
- the utilization rate and amount of use of resources of the computer system 1 as a whole may be a web request. It is also useful to calculate for each type of and each measurement period. In one embodiment, CPU utilization and memory usage as the entire computer system 1 are calculated for each type of Web request and for each measurement period. Specifically, in the measurement period “i”, the overall CPU usage rate ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ (i) of the computer system 1 of the web request of type “Rj” is
- ⁇ (i) N (i)-U / At (i), ⁇ ⁇ ⁇ ⁇ ⁇ (14 & ')
- Type ⁇ Rj The sum of CPU time consumed by one Web request, and the following formula:
- Type ⁇ Rj ⁇ Web request is the sum of memory usage of one, the following formula:
- the system management data generation tool 25 compares the load of the computer system 1 It has a function to estimate the response time (response time) of Web request when the target is light for each type of Web request. Response time is an effective parameter to analyze the computer system 1's ability. Specifically, the system management data generation server 25 calculates the response time T of the type “Rf web request according to the following equation.
- U W , U F and U D are in the Web layer, functional layer and data layer respectively
- the calculated response time is stored as response time data 43 in the storage device 7b.
- response time can be measured even in a production environment where some request processing is always performed. Therefore, it is not necessary to prepare a dedicated experiment environment.
- the type of the web request is defined by the method included in it, but may be a different definition.
- the type of web request may be defined by the URL referenced by the web request.
- the computer system 1 is hierarchized into three layers of the web layer, the functional layer, and the data layer, the computer system 1 may be a computer system hierarchized into two layers or four or more layers. It is also possible that the invention applies. The present invention can also be applied to non-hierarchized computer systems.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Debugging And Monitoring (AREA)
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008524801A JP4900881B2 (ja) | 2006-07-10 | 2007-07-10 | コンピュータシステム、管理装置、及びコンピュータプログラム |
US12/373,254 US8171133B2 (en) | 2006-07-10 | 2007-07-10 | Management apparatus and management method for computer system |
EP07790553A EP2040171A4 (en) | 2006-07-10 | 2007-07-10 | DEVICE AND METHOD FOR COMPUTER SYSTEM MANAGEMENT |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2006-188999 | 2006-07-10 | ||
JP2006188999 | 2006-07-10 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008007669A1 true WO2008007669A1 (fr) | 2008-01-17 |
Family
ID=38923227
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2007/063735 WO2008007669A1 (fr) | 2006-07-10 | 2007-07-10 | Dispositif et procédé de gestion de système informatique |
Country Status (4)
Country | Link |
---|---|
US (1) | US8171133B2 (ja) |
EP (1) | EP2040171A4 (ja) |
JP (1) | JP4900881B2 (ja) |
WO (1) | WO2008007669A1 (ja) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008225651A (ja) * | 2007-03-09 | 2008-09-25 | Nec Corp | 負荷分析システム、方法、及び、プログラム |
JP2012529234A (ja) * | 2009-06-03 | 2012-11-15 | マイクロソフト コーポレーション | サーバの利用状況の判定 |
JP6048555B1 (ja) * | 2015-09-10 | 2016-12-21 | 日本電気株式会社 | 分類情報作成装置、分類情報作成方法、分類情報作成プログラム、検索装置、検索方法、及び、検索プログラム |
JP2018124878A (ja) * | 2017-02-02 | 2018-08-09 | 富士通株式会社 | 性能要件推定プログラム、性能要件推定装置、および性能要件推定方法 |
JP2018206199A (ja) * | 2017-06-07 | 2018-12-27 | Kddi株式会社 | 管理装置、管理方法、及びプログラム |
US10735440B2 (en) | 2015-09-10 | 2020-08-04 | Nec Corporation | Communication destination determination device, communication destination determination method, and recording medium |
Families Citing this family (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090327495A1 (en) * | 2008-06-27 | 2009-12-31 | Oqo, Inc. | Computing with local and remote resources using automated optimization |
US20090328036A1 (en) * | 2008-06-27 | 2009-12-31 | Oqo, Inc. | Selection of virtual computing resources using hardware model presentations |
WO2010118255A2 (en) * | 2009-04-08 | 2010-10-14 | The University Of North Carolina At Chapel Hill | Methods, systems, and computer program products for network server performance anomaly detection |
JP2011134018A (ja) * | 2009-12-22 | 2011-07-07 | Canon Inc | 情報処理装置、情報処理システム、制御方法、及びプログラム |
JP2011170464A (ja) * | 2010-02-16 | 2011-09-01 | Ricoh Co Ltd | ライセンス管理システム、ライセンス管理方法およびプログラム |
EP2601563A4 (en) * | 2010-08-06 | 2016-04-20 | Hewlett Packard Development Co | SYSTEMS AND METHOD FOR QUOTING ENERGY CONSUMPTION |
US10797953B2 (en) * | 2010-10-22 | 2020-10-06 | International Business Machines Corporation | Server consolidation system |
TWI414161B (zh) * | 2011-01-28 | 2013-11-01 | Univ Nat Chiao Tung | 負載分配方法 |
US9178767B2 (en) * | 2011-12-09 | 2015-11-03 | Telefonaktiebolaget L M Ericsson (Publ) | Intelligent traffic quota management in split-architecture networks |
US8948191B2 (en) * | 2011-12-09 | 2015-02-03 | Telefonaktiebolaget L M Ericsson (Publ) | Intelligent traffic quota management |
JP5880962B2 (ja) * | 2012-06-12 | 2016-03-09 | ソニー株式会社 | 電子装置、計算方法、プログラム、および情報処理装置 |
US9189355B1 (en) * | 2013-05-21 | 2015-11-17 | Intuit Inc. | Method and system for processing a service request |
CN103455549A (zh) * | 2013-07-26 | 2013-12-18 | 苏州亿倍信息技术有限公司 | 一种数据处理的方法及系统 |
JP6264849B2 (ja) * | 2013-11-12 | 2018-01-24 | 富士通株式会社 | 分析方法、分析装置、及び分析プログラム |
US20150332295A1 (en) * | 2014-02-18 | 2015-11-19 | Mastercard International Incorporated | Method of Forecasting Resource Demand |
US20150309908A1 (en) * | 2014-04-29 | 2015-10-29 | Hewlett-Packard Development Company, L.P. | Generating an interactive visualization of metrics collected for functional entities |
US10832176B2 (en) | 2014-12-08 | 2020-11-10 | Mastercard International Incorporated | Cardholder travel detection with internet service |
US10255561B2 (en) | 2015-05-14 | 2019-04-09 | Mastercard International Incorporated | System, method and apparatus for detecting absent airline itineraries |
US10904127B2 (en) | 2016-06-13 | 2021-01-26 | Intel Corporation | Systems and methods for detecting a zombie server |
US10644963B2 (en) * | 2016-06-13 | 2020-05-05 | Intel Corporation | Systems and methods for detecting a zombie server |
CN108229564B (zh) * | 2018-01-05 | 2022-08-02 | 创新先进技术有限公司 | 一种数据的处理方法、装置及设备 |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0695924A (ja) | 1992-09-10 | 1994-04-08 | Nec Corp | 性能データ採取方式 |
JP2000207365A (ja) | 1999-01-13 | 2000-07-28 | Internatl Business Mach Corp <Ibm> | トランザクション処理方法及び記憶媒体 |
JP2001290641A (ja) | 2000-04-10 | 2001-10-19 | A & I System Kk | アプリケーションソフトウェア共用システム |
JP2001344224A (ja) | 2000-05-31 | 2001-12-14 | Toshiba Corp | クライアント/サーバ型協調動作方法 |
JP2002342182A (ja) * | 2001-05-21 | 2002-11-29 | Hitachi Ltd | ネットワークシステムにおける運用管理の支援システム |
JP2005025372A (ja) | 2003-06-30 | 2005-01-27 | Toshiba Corp | 計算機システムのcpu負荷分析装置、そのcpu負荷分析方法及びcpu負荷分析用プログラム |
JP2005031870A (ja) | 2003-07-10 | 2005-02-03 | Hitachi Ltd | ジョブ課金方法 |
JP2005064971A (ja) | 2003-08-15 | 2005-03-10 | Nippon Telegr & Teleph Corp <Ntt> | 負荷発生分布推定方法および装置、並びに高負荷判定方法および装置 |
JP2005148911A (ja) | 2003-11-12 | 2005-06-09 | Nec Corp | 負荷分散方法及び装置とシステム並びにプログラム |
JP2005216078A (ja) | 2004-01-30 | 2005-08-11 | Nec Corp | 計算機システム性能測定分析用のトランザクションプロファイルの生成システム、その生成方法及びプログラム |
JP2006024107A (ja) * | 2004-07-09 | 2006-01-26 | Fuji Seal International Inc | Rfidインレットの積層方法、及びrfidインレット積層体 |
JP2006024017A (ja) | 2004-07-08 | 2006-01-26 | Internatl Business Mach Corp <Ibm> | コンピュータ資源のキャパシティを予測するためのシステム、方法およびプログラム |
JP2006188999A (ja) | 2005-01-07 | 2006-07-20 | Honda Motor Co Ltd | 内燃機関の過給圧制御装置 |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3398599B2 (ja) | 1998-06-24 | 2003-04-21 | シャープ株式会社 | アプリケーションデータ処理装置 |
US7320131B1 (en) * | 2001-06-06 | 2008-01-15 | Cisco Technology, Inc. | Methods and apparatus for selecting a server to process a request |
EP1495417A4 (en) * | 2002-03-14 | 2007-10-03 | Unisys Corp | SYSTEM AND METHOD FOR ESTIMATING RESOURCE USE |
AU2002368528A1 (en) | 2002-12-27 | 2004-07-22 | Unisys Corporation | Improvements in a system and method for estimation of computer resource usage by transaction types |
JP4223411B2 (ja) | 2004-01-19 | 2009-02-12 | 株式会社日立製作所 | 実行多重度制御システム、実行多重度制御システムの制御方法、及びプログラム |
JP2006012065A (ja) | 2004-06-29 | 2006-01-12 | Honda Motor Co Ltd | 資源負荷適正化管理システム |
US7559053B2 (en) | 2004-08-24 | 2009-07-07 | Microsoft Corporation | Program and system performance data correlation |
JP4180638B2 (ja) | 2004-10-28 | 2008-11-12 | 富士通株式会社 | 分析方法及び装置 |
JP2006155276A (ja) | 2004-11-30 | 2006-06-15 | Hitachi Ltd | 計算機利用に関する予実算管理システム |
US7908314B2 (en) * | 2005-03-23 | 2011-03-15 | Hitachi, Ltd. | Method for controlling a management computer |
US20070276743A1 (en) * | 2006-05-24 | 2007-11-29 | Zalta Danny P | System and method for calculating fees associated with services |
US8200555B2 (en) * | 2006-06-08 | 2012-06-12 | International Business Machines Corporation | Method to monitor amount of usage of applications in a server and their billing |
-
2007
- 2007-07-10 WO PCT/JP2007/063735 patent/WO2008007669A1/ja active Application Filing
- 2007-07-10 US US12/373,254 patent/US8171133B2/en not_active Expired - Fee Related
- 2007-07-10 EP EP07790553A patent/EP2040171A4/en not_active Withdrawn
- 2007-07-10 JP JP2008524801A patent/JP4900881B2/ja not_active Expired - Fee Related
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0695924A (ja) | 1992-09-10 | 1994-04-08 | Nec Corp | 性能データ採取方式 |
JP2000207365A (ja) | 1999-01-13 | 2000-07-28 | Internatl Business Mach Corp <Ibm> | トランザクション処理方法及び記憶媒体 |
JP2001290641A (ja) | 2000-04-10 | 2001-10-19 | A & I System Kk | アプリケーションソフトウェア共用システム |
JP2001344224A (ja) | 2000-05-31 | 2001-12-14 | Toshiba Corp | クライアント/サーバ型協調動作方法 |
JP2002342182A (ja) * | 2001-05-21 | 2002-11-29 | Hitachi Ltd | ネットワークシステムにおける運用管理の支援システム |
JP2005025372A (ja) | 2003-06-30 | 2005-01-27 | Toshiba Corp | 計算機システムのcpu負荷分析装置、そのcpu負荷分析方法及びcpu負荷分析用プログラム |
JP2005031870A (ja) | 2003-07-10 | 2005-02-03 | Hitachi Ltd | ジョブ課金方法 |
JP2005064971A (ja) | 2003-08-15 | 2005-03-10 | Nippon Telegr & Teleph Corp <Ntt> | 負荷発生分布推定方法および装置、並びに高負荷判定方法および装置 |
JP2005148911A (ja) | 2003-11-12 | 2005-06-09 | Nec Corp | 負荷分散方法及び装置とシステム並びにプログラム |
JP2005216078A (ja) | 2004-01-30 | 2005-08-11 | Nec Corp | 計算機システム性能測定分析用のトランザクションプロファイルの生成システム、その生成方法及びプログラム |
JP2006024017A (ja) | 2004-07-08 | 2006-01-26 | Internatl Business Mach Corp <Ibm> | コンピュータ資源のキャパシティを予測するためのシステム、方法およびプログラム |
JP2006024107A (ja) * | 2004-07-09 | 2006-01-26 | Fuji Seal International Inc | Rfidインレットの積層方法、及びrfidインレット積層体 |
JP2006188999A (ja) | 2005-01-07 | 2006-07-20 | Honda Motor Co Ltd | 内燃機関の過給圧制御装置 |
Non-Patent Citations (5)
Title |
---|
"Nikkei Computer", 8 March 2004, NIKKEI BUSINESS PUBLICATIONS. INC, article "New Common Sense That Cannot Be Asked Any More" |
"Nikkei Open System", July 2001, NIKKEI BUSINESS PUBLICATIONS. INC, article "From Two-Layered C/S To n-Layered Web System, From EUC To Information Community BtoB", pages: 164 - 170 |
BEN ROLY; PETER ROLY: "Apache Handbook", 25 November 1997, OHMSHA LTD., pages: 41 - 57 |
See also references of EP2040171A4 |
SHUSHI KUBO: "OS Review", 1988, KYORITSU SHUPPAN CO., LTD, pages: 174 - 179 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008225651A (ja) * | 2007-03-09 | 2008-09-25 | Nec Corp | 負荷分析システム、方法、及び、プログラム |
JP2012529234A (ja) * | 2009-06-03 | 2012-11-15 | マイクロソフト コーポレーション | サーバの利用状況の判定 |
JP6048555B1 (ja) * | 2015-09-10 | 2016-12-21 | 日本電気株式会社 | 分類情報作成装置、分類情報作成方法、分類情報作成プログラム、検索装置、検索方法、及び、検索プログラム |
JP2017054333A (ja) * | 2015-09-10 | 2017-03-16 | 日本電気株式会社 | 分類情報作成装置、分類情報作成方法、分類情報作成プログラム、検索装置、検索方法、及び、検索プログラム |
US10735440B2 (en) | 2015-09-10 | 2020-08-04 | Nec Corporation | Communication destination determination device, communication destination determination method, and recording medium |
US10902027B2 (en) | 2015-09-10 | 2021-01-26 | Nec Corporation | Generation of category information for measurement value |
JP2018124878A (ja) * | 2017-02-02 | 2018-08-09 | 富士通株式会社 | 性能要件推定プログラム、性能要件推定装置、および性能要件推定方法 |
JP2018206199A (ja) * | 2017-06-07 | 2018-12-27 | Kddi株式会社 | 管理装置、管理方法、及びプログラム |
Also Published As
Publication number | Publication date |
---|---|
JPWO2008007669A1 (ja) | 2009-12-10 |
US8171133B2 (en) | 2012-05-01 |
EP2040171A1 (en) | 2009-03-25 |
EP2040171A4 (en) | 2010-09-22 |
JP4900881B2 (ja) | 2012-03-21 |
US20090287768A1 (en) | 2009-11-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2008007669A1 (fr) | Dispositif et procédé de gestion de système informatique | |
US8903996B2 (en) | Operating cloud computing services and cloud computing information system | |
Cao et al. | Grid load balancing using intelligent agents | |
US7035919B1 (en) | Method for calculating user weights for thin client sizing tool | |
US8612577B2 (en) | Systems and methods for migrating software modules into one or more clouds | |
US7533170B2 (en) | Coordinating the monitoring, management, and prediction of unintended changes within a grid environment | |
KR101977726B1 (ko) | 가상 데스크탑 서비스 방법 및 장치 | |
KR101826969B1 (ko) | 버스트 모드 제어 | |
Holzman et al. | Hepcloud, a new paradigm for hep facilities: Cms amazon web services investigation | |
US20050060360A1 (en) | Method, system and program product for managing system resources | |
US8132177B2 (en) | System and method for load-balancing in server farms | |
US20050108133A1 (en) | Service shopping and provisioning system and method | |
Agarwal et al. | Metering and accounting for composite e-services | |
US20100077393A1 (en) | Method and system for automatically generating software and configuring application installation | |
US7529185B2 (en) | Method, system and program product for determining bandwidth consumption across a multi-tiered computer environment | |
US20060130067A1 (en) | Method, system and program product for approximating resource consumption of a computer system | |
Anithakumari et al. | Negotiation and monitoring of service level agreements in cloud computing services | |
Solainayagi et al. | Resource Allocation Based on Matchmaking Services in Multiple Clouds Using Trustworthy and Scalable Service Providers Algorithm. | |
Cunha et al. | Renegotiation of electronic brokerage contracts | |
Aldawsari et al. | A survey of resource management challenges in multi-cloud environment: Taxonomy and empirical analysis | |
Sivasamy et al. | Modelling of a cloud platform via M/M1+ M2/1 queues of a Jackson network | |
Chard et al. | Using secure auctions to build a distributed metascheduler for the grid | |
Hu et al. | Minimizing geo-distributed interactive service cost with multiple cloud service providers | |
Koneru et al. | Three tier architecture: To select CSP through CloudServiceBroker in multicloud environment | |
US20110288904A1 (en) | System, Method, and Software for Analyzing Maneuvers of an Application in a Distributed Computing Environment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07790553 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008524801 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12373254 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007790553 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: RU |