JP6668467B2 - Management system and resource scheduling method - Google Patents

Description

  In general, the present invention, when constructing and operating a system using cloud infrastructure resources, automatically deploys the system by selecting an appropriate service from the cloud service menu according to the requirements of the system execution infrastructure, and Execute automatically. The present invention relates to a resource schedule management device and method for constructing and operating a system using a cloud, in which even if there is no cloud service that satisfies all the requirements for a system execution platform, a candidate plan for a combination of services close to the requirement is generated.

  Generally, the configuration of an information system (eg, computing resources as components and their arrangement) is determined based on non-functional requirements.

  According to the technique disclosed in Patent Document 1, when the performance becomes insufficient, a possible countermeasure is proposed by the designated deadline period in order to improve the performance.

WO2014 / 073045

  According to Patent Literature 1, as described above, a possible countermeasure for improving performance within a period up to a deadline is presented. Therefore, if there is no computing resource required for performance improvement before the deadline, no countermeasure can be presented. In this case, it is necessary to wait until the performance is satisfied, such as extending the deadline.

  However, in recent years, it has been desired to quickly build an information system by using technologies such as cloud services (cloud computing), so that new services can be started quickly or existing services can be quickly changed according to needs. It is important to do that.

  A management system for managing a computer system having a plurality of resources including a plurality of types of resources is constructed. The management system accepts the user's input information and displays, to the user, output information including information representing one or more service types that match the input information among the service types belonging to the search range among the plurality of service types. I do. The input information includes an infrastructure resource type, which is a type of a specified component among one or more components of the system basic configuration, and a plurality of non-functional requirements respectively specified for the plurality of non-functional requirements associated with the infrastructure resource type. First level. Two or more non-functional requirements and two or more second levels respectively associated with the two or more non-functional requirements are associated with each of the plurality of service types. Each of the one or more service types matching the input information is the same non-functional as the plurality of non-functional requirements respectively corresponding to the plurality of first levels included in the input information among the service types belonging to the search range. The service type is associated with requirements, and is associated with two or more second levels that are relatively close to the plurality of first levels included in the input information.

  Even if there is no service type associated with two or more second levels that perfectly match the plurality of first levels specified by the user for each of the plurality of non-functional requirements, two or more service levels that are close to the plurality of first levels The user can receive the presentation of the service type associated with the second level.

Schematic diagram (Example 1). 1 is an overall configuration diagram of a system (first embodiment). FIG. 2 is a configuration diagram of a cloud orchestrator server (first embodiment). FIG. 1 is a configuration diagram of a cloud operation management server (first embodiment). FIG. 2 is a configuration diagram of a monitoring server (first embodiment). 5 is a flowchart of processing between servers and data (Example 1). FIG. 3 is a configuration diagram of system basic configuration information and user input information (first embodiment). FIG. 3 is a configuration diagram of target resource information (first embodiment). FIG. 3 is a configuration diagram of design pattern information (first embodiment). FIG. 2 is a configuration diagram of infrastructure resource information (first embodiment). FIG. 3 is a processing flowchart of a cloud orchestrator server (first embodiment). FIG. 4 is a processing flowchart of the cloud management server (first embodiment). System planner input screen (Example 1). FIG. 4 is a schematic diagram of a service menu (server resource) input screen (first embodiment). FIG. 4 is a schematic diagram of a service plan formulation screen (first embodiment). FIG. 3 is a schematic diagram of a design pattern display screen (Example 1). FIG. 4 is a schematic diagram of a processing flow editing screen (first embodiment). Schematic diagram (Example 2). Overall system configuration diagram (second embodiment) FIG. 3 is a configuration diagram of a backup server (second embodiment). FIG. 3 is a configuration diagram of a cloud orchestrator server (second embodiment). FIG. 9 is a flowchart of processing and data between servers (second embodiment). FIG. 9 is a configuration diagram of a backup service menu (second embodiment). FIG. 9 is a configuration diagram of design pattern information (second embodiment). FIG. 7 is a configuration diagram of infrastructure resource information (second embodiment). FIG. 10 is a processing flowchart of a cloud orchestrator server (second embodiment). FIG. 9 is a schematic diagram of a system planner input screen (second embodiment). FIG. 9 is a schematic diagram of a service menu (storage resource) input screen (second embodiment). FIG. 9 is a schematic diagram of a service menu (backup service) input screen (second embodiment). The schematic diagram of a service plan formulation screen (Example 2). FIG. 9 is a schematic diagram of a design pattern display screen (Example 2). FIG. 9 is a schematic diagram of a processing flow editing screen (second embodiment).

  Hereinafter, several embodiments of the present invention will be described.

  In the following description, the “interface unit” includes one or more interfaces. The one or more interfaces may be one or more same type interface devices (eg, one or more NICs (Network Interface Card)) or two or more different types of interface devices (eg, NIC and HBA (Host Bus Adapter)). There may be.

  In the following description, the “storage unit” includes one or more memories. At least one memory may be a volatile memory or a non-volatile memory. The storage unit may include one or more PDEVs in addition to one or more memories. “PDEV” means a physical storage device, which may typically be a non-volatile storage device (eg, an auxiliary storage device). The PDEV may be, for example, a hard disk drive (HDD) or a solid state drive (SSD).

  In the following description, the “processor unit” includes one or more processors. At least one processor is typically a CPU (Central Processing Unit). The processor may include a hardware circuit that performs some or all of the processing.

  Further, in the following description, processing may be described with “program” as the subject, but the program is executed by the processor unit, and the determined processing is appropriately performed by the storage unit and the interface unit. Since the processing is performed while using at least one, the subject of the processing may be a processor unit (or a computer or a computer system having the processor unit). The program may be installed on the computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. Further, in the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

  In the following description, “xxx information” is mainly in a table format, but the information may be represented by any data structure. In the following description, the configuration of various types of information is merely an example, and various types of information may be divided into two or more information elements, or all or some of two or more types of information may be aggregated into the same information. Is also good.

  In the following description, time such as date and time or time may be expressed as year, month, day, hour, minute, second, or a part thereof.

  In the following description, the “management system” may be configured by one or more computers. Specifically, for example, when the management computer has a display device and the management computer displays information on its own display device, the management computer may be a management system. Further, for example, when a management computer (for example, a server) transmits display information to a remote display computer (for example, a client) and the display computer displays the information (when the management computer displays information on the display computer) ), A system including at least the management computer of the management computer and the display computer may be the management system. The management system may include an interface unit, a storage unit, and a processor unit connected thereto. The interface unit may include at least one of a user interface unit and a communication interface unit. The user interface unit includes at least one of one or more I / O devices (for example, an input device (for example, a keyboard and a pointing device) and an output device (for example, a display device)) and a display computer. Good. The communication interface unit may include one or more communication interface devices. The fact that the computer in the management system “displays the display information” may be that the display information is displayed on a display device of the computer, or that the computer transmits the display information to the display computer. (In the latter case, display information is displayed by a display computer). In the following description, the management system is at least the cloud orchestrator server of the cloud orchestrator server and the cloud operation management server.

  In the following description, a “user” is a person who receives a cloud computing service, and an “infrastructure administrator” is a person who provides a cloud computing service.

  FIG. 1 is a schematic diagram of the first embodiment. According to FIG. 1, an example in which a DB (database) server of a Web system is constructed is employed for the purpose of explaining the outline of the present embodiment.

  The cloud orchestrator server (not shown) receives the information 101 from the user (S1-1). The user input information 101 includes a system basic configuration 210 and an importance level 101b. The system basic configuration 210 is information indicating the basic configuration of the information system to be constructed. Here, the “basic configuration” includes the infrastructure resources of the information system and the relationship between the infrastructure resources. The importance level 101b is a level for each of one or more non-functional requirements for infrastructure resources. The “infrastructure resource” may be the highest-level resource in the information system. An infrastructure resource is a combination of one or more computing resources. A plurality of computing resources including a plurality of types of computing resources include a plurality of physical resources (for example, a CPU and a memory) and a plurality of logical resources (for example, a VM (Virtual Machine), an application program, and a logical volume). Is fine. The importance level 101b may be associated with each of all the infrastructure resources, or the importance level 101b may be associated with only some of the infrastructure resources. In the example of FIG. 1, the infrastructure resource associated with the importance level 101b is the “DB tier” (DB server). There are five levels of importance of each non-functional requirement, and the higher the degree of importance, the higher the numerical value of the level. The number of importance levels of all the non-functional requirements may not be the same number “5”. For example, the number of importance levels for the first non-functional requirement may be “5” or the number of importance levels for the second non-functional requirement may be more or less than “5” (however, The number of degree levels is an integer of 2 or more). The “non-functional requirements” may be any requirements other than the functional requirements among the defined requirements, and may be, for example, availability, performance, scalability, and the like provided by the system execution platform. More specifically, the non-functional requirements for availability can be acceptable downtime, the non-functional requirements for performance can be latency and throughput, and the non-functional requirements for scalability can be scaled up and scaled out. Is fine. The definition of “non-functional requirement” may be a definition according to a guideline issued by a predetermined organization to which a plurality of private companies belong, or may be a definition determined by a private company such as a cloud service provider. In this embodiment, as the downtime of the availability, the importance level is 5 within 30 seconds, the importance level is 4 within 1 minute, the importance level is 3 within 5 minutes, and the importance level is 2 within 1 hour. In this embodiment, the performance is a disk I / O (Input / Output) performance.

  Next, the cloud orchestrator server receives a selection of a resource type (target resource type) from the user in the infrastructure resource menu (target resource) 102 which is a service menu of the infrastructure resource type (S1-2). The target resource type selected by the user is called “selected target resource type”. In this embodiment, in the infrastructure resource menu (target resource) 102 (menu), at least one of “network resource”, “server resource”, and “storage resource” may be included as the infrastructure resource type. In this embodiment, it is assumed that the infrastructure resource menu (target resource) 102 is information including information on server resources. It is assumed that there are two or more options (target resource types) for at least one infrastructure resource type (a server resource in this embodiment). In the example of FIG. 1, it is assumed that the server resources include bare metal, VM resources, and HA (High Availability) configuration resources as options. In the example of FIG. 1, it is assumed that the user has selected the HA configuration resource as the target resource type from the server resources in order to reduce downtime (non-functional requirement having a high importance level). The HA configuration resources include an FT (Fault Tolerant), an HA cluster, and cluster software. The FT is a resource type configured by a plurality of different physical resources (for example, a plurality of different server machines). The HA cluster is a resource type composed of a plurality of different hypervisors. The cluster software is a resource type configured by a plurality of different VMs executed on a plurality of different hypervisors.

  With respect to various resources represented by the infrastructure resource menu (target resource) 102, a provision level is associated with each non-functional requirement. Therefore, also for the selected target resource type, the provision level is associated with each non-functional requirement. Note that typically, the criterion for the provision level and the criterion for the importance level are the same. That is, for example, in this embodiment, as the downtime of availability, the provision level is 5 within 30 seconds, the provision level is 4 within 1 minute, the provision level is 3 within 5 minutes, and the provision level is 2 within 1 hour. It is said.

  The cloud orchestrator server performs a presentation process (S1-3). The presentation process includes, for example, a comparison between the importance level and the achievement level (S1-3-1), and a display of output information including information indicating a comparison result (S1-3-2).

  The comparison between the importance level and the achievement level is performed by comparing the first type of comparison with the achievement level of the selected target resource type and the second type of comparison with the expanded achievement level. At least one type of comparison is included. The output information 105 includes an achievement level 105a. In the example of FIG. 1, the achievement level 105a includes a first type of comparison result (left side) and a second type of comparison result (right side). The achievement level 105a includes a radar chart indicating the achievement level of each non-functional requirement and the importance level of each non-functional requirement for the selected target resource type “HA configuration resource”. The higher the achievement for the non-functional requirements, the higher the numerical value of the achievement level. That is, for each non-functional requirement, if the achievement level (predefined numerical value) is equal to or more than the importance level (numerical value input by the user), it means that the non-functional requirement is satisfied. Is less than the importance level, the larger the difference between the achievement level and the importance level, the more the non-functional requirement is not satisfied. Specifically, for each non-functional requirement, when the importance level is “x” and the achievement level is “y”, the non-functional requirement is satisfied if x ≦ y (y−x ≧ 0). If x> y (y-x <0), the non-functional requirement is not satisfied.

  According to the first type of comparison result, for the FT, the downtime is satisfied because the achievement level is the same as the importance level, but the cost is not satisfied because the achievement level is less than the importance level. . On the other hand, for the cluster software, the downtime is not satisfied because the achievement level is lower than the importance level, but the cost is satisfied because the achievement level is equal to or higher than the importance level.

  The second type of comparison is performed when the selected target resource type is a highly scalable resource type, or when the selected target resource type includes a highly scalable resource type. A highly scalable resource type is a resource type whose achievement level of the non-functional requirement “extensibility” is equal to or higher than the importance level (or, even if the achievement level is lower than the importance level, the difference between them is a predetermined value. Resource type). According to the first comparison result, since the cluster software corresponds to the resource type having high expandability, the second type of comparison is performed on the cluster software. The second type of comparison result indicates an achievement level for each of scale-out and scale-up as an extension of cluster software (an example of a resource type with high expandability). Each of scale out and scale up is an example of an expansion mode. For each of the resource types, an amount of change in the achievement level (a value to be added or subtracted) is associated with each extension mode and each non-functional requirement, and a second comparison is performed based on the amount of change. Will be According to an example of the second comparison result, since the cluster software is scaled out, the level of achievement of the latency is increased, the degree of the latency not being satisfied is reduced, and the cluster software is scaled up. As a result, the level of achievement of the latency is further increased, and the degree to which the latency is not satisfied is further reduced.

  The achievement level 105a (comparison result) included in the output information 105 is a polygonal radar chart having a plurality of vertices respectively corresponding to a plurality of non-functional requirements, but the display format of the calculation result is limited to the radar chart. I can't. For example, the calculation result may be a matrix of non-functional requirements and resource types. Each cell of the matrix represents a result of comparison between the achievement level and the importance level (for example, whether the achievement level (y) is equal to or greater than the importance level (x), or a difference between them if the achievement level (y) is less than the importance level (x)). Is also good.

  According to the example of FIG. 1, the achievement level is displayed for each of the various resource types included in the selected target resource type. However, if one resource type in the selected target resource type includes all the non-functional requirements, If the importance level is satisfied, the cloud orchestrator server may skip any comparison and display of the achievement level for other resource types (there may be no execution). If there is an achievement level that does not match the importance level of at least one non-functional requirement for any of the target resource types, the cloud orchestrator server fulfills at least one target resource type for each non-functional requirement. The degree level and the importance level may be compared and the result of the comparison may be displayed. Note that "when there is an achievement level that does not match the importance level of at least one non-functional requirement" may be "when at least one non-functional requirement is not satisfied". That is, if the importance level is satisfied for all the non-functional requirements specified by the user though the resources are excessive, it does not correspond to “when there is an achievement level that does not match the importance level of at least one non-functional requirement”. May be.

  Further, according to the example of FIG. 1, the user selects the target resource type (S1-2), but this selection may be skipped. In other words, information input (selection of the basic system configuration, selection of the resource type (for example, “DB layer”) in the basic system configuration, designation of the non-functional requirement for the selected resource type, and When the importance level is specified, the cloud orchestrator server uses the input information of the user (the type ID of the infrastructure resource type (for example, the type name), the non-functional requirements associated therewith, and the importance level thereof). By searching for the target resource type definition (type ID (eg, type name), non-functional requirement, and achievement level thereof associated with each of the plurality of target resource types), the target resource type closest to the input information is retrieved. Output information including information representing (system configuration method) may be displayed. In this embodiment, the “system configuration method” is defined by a system configuration pattern such as bare metal, FT, HA cluster, and cluster software, and the number of components having a configuration according to the system configuration pattern. The “closest target resource type” is a target resource type that has the highest matching between the specified importance level and the defined achievement level for the specified non-functional requirement. As for the consistency, the degree of importance of the non-functional requirement may be considered in addition to the consistency of each non-functional requirement (the consistency between the importance level and the achievement level). For example, for a non-functional requirement with a higher importance level and a non-functional requirement with a lower importance level, the weight that the importance level and the achievement level match is the weight of the non-functional requirement with the higher importance level Is high. Further, “match between the importance level and the achievement level” may mean that the achievement level exceeds the importance level in addition to the perfect match. As described above, the search range of the search using the input information of the user may be the entire target resource type, or may be the target resource type selected by the user among the plurality of target resource types.

  Further, according to the example of FIG. 1, if the achievement level of the specific non-functional requirement of the target resource type is equal to or higher than the importance level of the non-functional requirement, the cloud orchestrator server sets For at least one of the plurality of viewpoints (for example, the viewpoint closest to the user requirement (importance level)), a comparison between the importance level and the achievement level for each of the plurality of non-functional requirements (for example, a radar chart) Can be displayed. An example of the “specific non-functional requirement” here is “extensibility”, and the “viewpoint” is a variation of extensibility, for example, scale-up and scale-out.

  Hereinafter, Example 1 will be described in detail.

  FIG. 2 is an overall configuration diagram of the system according to the first embodiment.

  The overall system configuration includes a management system, an execution system, and a remote site.

  The management system includes a cloud orchestrator server 201, a cloud operation management server 202, and a monitoring server 203.

  The execution system includes a physical resource (server) 204a, a physical resource (storage) 204b, and a physical resource (network) 204c. The execution system may be a pool of a plurality of resources including a plurality of types of resources (typically, computing resources).

  The remote site includes a terminal 206. The terminal 206 may be a computer (for example, a personal computer) of a user (for example, a client of a cloud service provider). The terminal 206 accesses the Web browser 207 via the network 205. The user input information 101 is input from the terminal 206.

  The cloud orchestrator server 201 stores a program planner 208, a construction automation 209, a data basic system configuration 210, a construction processing method (script) 211, and a design pattern (system configuration / construction processing flow) 212. The cloud orchestrator server 201 includes an interface unit including an interface that communicates with the network 205, a storage unit including at least a memory of a memory and an auxiliary storage device, and a processor unit including at least one processor. A processor unit is connected to the interface unit and the memory unit. Note that the configuration including the interface unit, the storage unit, and the processor unit may be the same as at least one of the servers 202, 203, 204a, and 204b instead of or in addition to the cloud orchestrator server 201. The same applies to 206. At least one of the servers 201, 202, and 203 may be a virtual server executed in a server system including one or more physical servers. Elements 208, 209, 210, 211 and 212 may be stored in the storage unit. Elements 208 and 209 may be computer programs executed by the processor unit.

  The cloud operation management server 202 includes server resource management 213a, storage resource management 213b, network resource management 213c, and infrastructure service management 214 of program virtual resource management 213, data infrastructure service menu (target resource) 102, infrastructure resource information (Configuration / assignment status) 215 is stored. The elements 214, 102, 215, 213a, 213b, and 213b may be stored in the storage unit of the cloud operation management server 202. The elements 214, 213a, 213b, and 213c may be computer programs executed by the processor unit of the cloud operation management server 202.

  The monitoring server 203 stores program configuration management 216, operating performance management 218, data configuration information 217, and operating performance information 219. The elements 216, 218, 217, and 219 may be stored in the storage unit of the monitoring server 203. Elements 216 and 218 may be computer programs executed on the processor unit of monitoring server 203.

  FIG. 3 is a configuration diagram of the cloud orchestrator server 201.

  The screen display processing 305, the processing of the planner 208, and the processing of the construction automation 209 of the cloud orchestrator server 201 shown in FIG. 2 are performed by hardware including a memory 301a, a processor 302a, a storage device 303a, and a communication interface 304a. Specifically, it can be realized (for example, it can be realized by executing one or more programs by the processor 302a). Note that the illustrated input device may be a keyboard and a pointing device, and the output device may be a display device. However, these may be the terminal 206.

  The system basic configuration / non-functional requirement registration processing 306, the target resource candidate plan generation processing 307, the design pattern selection processing 308, and the construction processing flow generation processing 309 included in the planner 208 are programs read into the memory 301a (for example, the program 208). Is executed by the processor 302a.

  The construction processing flow automation processing 310 included in the construction automation 210 can be realized by the processor 302a executing a program (for example, the program 210) read into the memory 301a.

  The storage device 303a stores a system basic configuration 210, a construction processing method (script) 211, and a design pattern (system configuration & construction processing flow) 212.

  The input / output information exchanged from the terminal 206 via the web browser 207, the infrastructure service menu 102 of the cloud operation management server 202, and the infrastructure resource information 215 may be input / output via the communication interface 304a.

  FIG. 4 is a configuration diagram of the cloud operation management server 202.

  The screen display processing 401, the processing of the infrastructure service management 214, and the processing of the virtual resource management 213 of the cloud operation management server 202 shown in FIG. 2 are performed by hardware including a memory 301b, a processor 302b, a storage device 303b, and a communication interface 304b. This can be specifically realized by hardware (for example, it can be realized by executing one or more programs by the processor 302b). Note that the illustrated input device may be a keyboard and a pointing device, and the output device may be a display device. However, these may be the terminal 206.

  The target resource non-functional requirement definition process 402 and the target resource allocation method definition process 403 included in the infrastructure service management 214 are realized by the processor 302b executing a program (for example, the program 214) read into the memory 301b. It is possible.

  The virtual resource allocation process 404 included in the virtual resource management 213 can be realized by the processor 302b executing a program (for example, the program 213) read into the memory 301b.

  The storage device 303b stores an infrastructure resource menu (target resource) 102 and infrastructure resource information 215.

  The input / output information exchanged from the terminal 206 via the web browser 207, the construction processing script 211 of the cloud orchestrator server 201, and the information of the design pattern 212 may be input / output via the communication interface 304b.

  FIG. 5 is a configuration diagram of the monitoring server 203.

  The screen display processing 501, the processing of the configuration management 216, and the processing of the operation performance management 218 of the monitoring server 203 illustrated in FIG. 2 are specifically performed by hardware including the memory 301c, the processor 302c, the storage device 303c, and the communication interface 304c. (For example, by executing one or more programs by the processor 302c). Note that the illustrated input device may be a keyboard and a pointing device, and the output device may be a display device. However, these may be the terminal 206.

  The configuration management setting / cancellation process 502 and the configuration information collection process 503 included in the configuration management 216 can be realized by the processor 302c executing a program (for example, the program 216) read into the memory 301c.

  The monitoring setting / cancellation processing 504 and the monitoring information collection processing 505 included in the operation performance management 218 can be realized by the processor 302c executing a program (for example, the program 218) read into the memory 301c.

  The storage device 303c stores configuration information 217 and operation performance information 219.

  The input / output information exchanged from the terminal 206 via the web browser 207, the construction processing script 211 and the design pattern 212 of the cloud orchestrator server 201, and the infrastructure resource information 215 of the cloud operation management server 202 are input / output via the communication interface 304c. May be.

  FIG. 6 shows processing and data flow between servers. Specifically, for example, FIG. 6 shows a data flow of each processing among the terminal 206, the cloud orchestrator server 201, the cloud operation management server 202, and the monitoring server 203.

  The target resource non-functional requirement definition processing 402 inputs at least a part of the infrastructure resource menu (target resource) 102 (for example, the information 801 in FIG. 8) to the cloud operation management server 202 and stores it. As a result, non-functional requirements for each resource type belonging to the target resource are defined.

  The target resource allocation method definition process 403 inputs at least a part of the infrastructure resource menu (target resource) 102 (for example, a selection of a resource type in the information 102 is received from a user), and the input information (for example, Based on the selected target resource type (information indicating the selected target resource type), a design pattern 212 (here, a system configuration) that calls the processes of the cloud operation management server 202 and the monitoring server 203 is generated and stored.

  The target resource candidate plan generation process 307 receives the user input information 101 and uses the information 101 (for example, the importance level of each of the non-functional requirements for the selected infrastructure resource type) from the cloud operation management server 202 to the target. Select a resource candidate plan.

  The design pattern selection processing 308 inputs a target resource candidate plan among the selected target resource candidate plans (for example, accepts selection of a target resource candidate plan from the selected target resource candidate plans from a user), and receives the input. A system configuration according to the target resource candidate plan is selected from the stored design patterns 212.

  The construction processing flow generation processing 309 inputs the selected system configuration (for example, accepts the approval of the selected system configuration from the user) and executes the construction processing flow (part of the information stored in the design pattern 212). To edit.

  The construction processing flow automation processing 310, the edited construction processing flow is input, the deployment is executed in the virtual resource allocation processing 404 of the cloud operation management server 202, and the monitoring setting is executed in the monitoring setting / cancellation processing 504 of the monitoring server 203. I do.

  FIG. 7 shows the configuration of the system basic configuration 210 and the configuration of the user input information 101.

  The system basic configuration 210 includes a system type 210a, an infrastructure resource type 210b, and a hierarchy (use) 210c. The system type 210a indicates the name of the type of the information system to be constructed. The infrastructure resource type 210b indicates the name of each type of one or more infrastructure resources constituting the information system to be constructed. The hierarchy (use) 210c represents the name of a resource type (hereinafter, resource type hierarchy) included in the infrastructure resource type for each infrastructure resource type. The infrastructure resource type may be a large category, and the resource type hierarchy may be a small category within the large category.

  As described above, the user input information 101 includes the system basic configuration 210 and the importance level 101b. The importance level 101b includes a size 701, a number 702, and a non-functional requirement group 703.

  The size 701 represents the size specified by the user for each resource type hierarchy. The size has three levels of “S” (Small), “M” (Medium), and “L” (Large), but may have two levels or four or more levels. At least one of the servers 201, 202, and 203 may hold information indicating a relationship between a size level and size details (for example, a resource and its capacity or number) for each resource type hierarchy.

  The number 702 indicates the number specified by the user for each resource type hierarchy. For example, for the resource type hierarchy “DB layer”, the number 702 is “2”, which is a resource (component) belonging to the resource type (selected target resource type) selected for the infrastructure resource type “server resource”. Is "2". According to the example of FIG. 1, when the selected target resource type is “HA configuration resource”, the number 702 “2” means that FT has two hardware (for example, physical servers), and HA For clusters, this means that there are two hypervisors, and for cluster software, it means that there are two VMs.

  The non-functional requirement group 703 indicates a plurality of non-functional requirements that can be specified (selected) by the user and the importance level specified by the user for each non-functional requirement. Specifically, for example, the non-functional requirement group 703 includes downtime 703a, latency 703b, throughput 703c, expandability 703d, capacity 703e, RTO (Recovery Time Objective) 703f, RPO (Recovery Point Objective) 703g, and cost. 703h. Each of the various types of information 703a to 703h includes an importance level specified by the user for each resource type hierarchy.

  FIG. 8 shows the configuration of the infrastructure resource menu (target resource) 102.

  The infrastructure resource menu (target resource) 102 includes an infrastructure resource type 103, a target resource group 104, and a non-functional requirement group 801.

  The infrastructure resource type 103 indicates an infrastructure resource type.

  The target resource group 104 includes a configuration 104a and a type 104b. The configuration 104a indicates the type of the configuration of the infrastructure resource type. The type 104b represents, for each configuration type of the infrastructure resource type, a resource type (a target resource type) belonging to the configuration type. For the configuration type “single”, there are “bare metal” and “VM” as resource types. As for the configuration type “HA”, since the resource type corresponding to the type is “HA configuration resource”, “FT”, “HA cluster”, and “cluster software” exist as described above.

  The non-functional requirement group 801 indicates a plurality of non-functional requirements that can be designated (selected) by the user and the achievement level of each non-functional requirement. Specifically, for example, the non-functional requirement group 801 includes downtime 801a, latency 801b, throughput 801c, expandability 801d, capacity 801e, RTO 801f, RPO 801g, and cost 801h. Each of the various information 801a to 801h includes an achievement level for each resource type.

  FIG. 9 shows a configuration of the design pattern 212. The design pattern 212 may be generated for each resource type belonging to the infrastructure resource type. The design pattern 212 in FIG. 9 is information corresponding to the design type “cluster software”.

The design pattern 212 is as follows:
(*) An infrastructure resource type 903 indicating an infrastructure resource type to which a corresponding resource type “cluster software” belongs;
(*) A target resource group 904 including a type 904b representing the corresponding resource type "cluster software" and a configuration 904a representing the configuration type to which it belongs;
(*) A non-functional requirement group 905 (detailed illustration is omitted) composed of entries (records) corresponding to the resource type “cluster software” in the non-functional requirement group 801 (information including the achievement level). ,
(*) Design pattern group 901
including. The design pattern group 901 includes a system configuration 901a and a construction processing flow 901b.

  The system configuration 901a includes a system configuration diagram of the cluster software. Here, between VMs on different hypervisors, a cluster management program (Manager) executed by the VM monitors the life and death by heartbeat, and there is no response from one OS or application even after a certain time has elapsed. Starts the other OS or application and continues service operation.

  The construction processing flow 901b is a system construction processing flow of a cluster software configuration. To automatically construct the system configuration 901a, a flow 901b represents a flow as a construction processing procedure.

  FIG. 10 shows the configuration of the infrastructure resource information 215.

  The infrastructure resource information 215 includes an infrastructure resource type 1004 representing the selected infrastructure resource type “server resource”, a type 1005b representing each resource type belonging to the corresponding infrastructure resource type “server resource”, and a configuration representing the configuration type to which it belongs. A target resource group 1005 including a resource 1005a, an acquisition source 1001, a resource tag 1002, and a resource allocation status (free resource) 1003. The acquisition source 1001 indicates, for each resource type, a location where a computing resource that can be allocated exists, in other words, a location where a resource type (information system) to be constructed is constructed. The resource tag 1002 represents a resource type tag. In the example of FIG. 10, the tag is the same as the resource type name. The resource allocation status 1003 indicates, for each resource type, the ratio of the amount of the free resource set in the allocatable resource set to the amount of the allocatable resource set. The “assignable resource set” may be at least a part of a cloud (a shared pool of a plurality of computing resources including a plurality of types of computing resources), and can be assigned as a component of an information system corresponding to the resource type. All resources are sufficient. The “free resource set” may be one or more free resources (unallocated resources).

  FIG. 11 shows a processing flow of the cloud orchestrator server 201.

  The planner 208 determines (selects) the system basic configuration 210 in the system basic configuration / non-functional requirement registration processing 306 and, for each resource type hierarchy in the system basic configuration, the size, the number, the non-functional requirements, and the non-functional requirements. The processing 1101 for accepting the importance level of is executed, and the result (input information) is stored in the user input information 101.

  In the target resource candidate plan generation processing 307, the planner 208 executes processing 1102 of receiving a selection of a resource type from one or more infrastructure resource types (one or more resource type layers) represented by the user input information 101 from the user.

In the process 307, next, the planner 208 assigns the resource type satisfying the following (condition 01) to (condition 04) to the target resource 102 (or a plurality of target resource types belonging to the target resource type selected by the user). A process 1103 for selecting from is executed.
(Condition 01) Resource type corresponding to the selected resource type (resource type hierarchy or the type of infrastructure resource to which it belongs).
(Condition 02) Among the resource types corresponding to (Condition 01), a resource type in which a plurality of non-functional requirements identical to a plurality of non-functional requirements associated with the selected resource type are associated.
(Condition 03) Among the resource types corresponding to (Condition 02), for each non-functional requirement, a resource type associated with the same achievement level as the importance level.
(Condition 04) If there is no resource type corresponding to (Condition 03), among the resource types corresponding to (Condition 02), a plurality of achievement levels close to a plurality of importance levels associated with the selected resource type are set. Associated resource type.

  The resource type corresponding to (Condition 04) is, for example, that the higher the non-functional requirement, the higher the achievement level matches the importance level (or is higher than the importance level), or the achievement level Is less than the importance level, the resource type that most satisfies that the difference is as small as possible may be used.

  In the process 307, finally, the planner 208 executes a process 1104 for confirming a resource allocation status (availability) corresponding to the selected target resource type, and stores the result in the target resource candidate plan 1151. The “vacancy status” here is information registered in the infrastructure resource information 215 (FIG. 10). The virtual resource is regarded as one resource for each target resource type, and in step 307, it is checked whether the amount of unused (usable) resources for all the resources is larger than the required amount of resources. If it is determined that the required resource amount is less than the required amount (insufficient), the planner 208 can propose a plan to reduce the allocated amount or change to another target resource.

  The candidate plan calculation method of the process 1103 may be, for example, the candidate plan calculation method 1108. According to this method 1108, for each non-functional requirement, a difference is made between the importance level of the resource type selected by the user and the achievement level of the target resource type, and the value of the importance level is used as a coefficient to calculate a linear The optimal solution (the target resource type closest to (including the same as) the importance level specified by the user for the plurality of non-functional requirements) is calculated using the integer programming.

  In the design pattern selection process 308, the planner 208 determines a design pattern (an infrastructure resource type (resource type hierarchy) selected by the user and a design pattern (system configuration) corresponding to the selected target resource type corresponding to the target resource candidate plan 1151. The processing 1105 for selecting)) is executed. In processing 1105, the planner 208 performs predetermined recording on the pattern 212 so that the design pattern 212 corresponding to the selected result (design pattern having the selected system configuration) can be specified in the processing 309 described below. Good.

  In the construction process flow generation process 309, the planner 208 inputs the design pattern selected in the process 1105, and executes a process 1106 of selecting a process flow corresponding to the design pattern. In the processing 1106, the planner 208 stores the result in the construction processing flow 212.

  The construction automation 209 inputs the construction processing flow 212 in the construction processing flow automation processing 310, and executes processing 1107 for automatically executing each processing according to the processing flow. In step 1107, the system is automatically deployed on the physical resource. That is, an information system is constructed.

  FIG. 12 shows a processing flow of the cloud operation management server 202.

  In the target resource non-functional requirement definition processing 402, the infrastructure service management 214 inputs the infrastructure resource information 215, defines the non-functional requirements (and the achievement level) of the target resource type corresponding to the infrastructure service type, and Is executed in the infrastructure resource menu (target resource) 102.

  In the target resource allocation method definition process 403, the infrastructure service management 214 inputs the infrastructure resource menu (target resource) 102 and generates a design pattern 212 that is a method for allocation to physical resources corresponding to the infrastructure resource type. A process 1202 of adding the design pattern 212 (adding it to the cloud orchestrator server 201) is executed.

  In the virtual resource allocation process 404, the virtual resource management 213 inputs the infrastructure resource information 215 and the infrastructure resource menu (target resource) 102, actually allocates physical resources, and stores the result in the infrastructure resource information 215. In the present process 404, the virtual resource management 213 performs a process 1203 for selecting a physical resource corresponding to the virtual resource, a process 1204 for searching for a free status of the physical resource, and a process 1205 for acquiring and allocating a virtual resource from the free resource. Execute.

  13 to 17 show the cloud orchestrator management screen according to the first embodiment. The cloud orchestrator management screen is a screen (typically, a GUI (Graphical User Interface)) that is executed in the screen display processing 305 of the cloud orchestrator server 201 and displayed on the Web browser 207 of the terminal 206. The “user operation” below is an operation performed by the user, such as a click using a pointing device.

  FIG. 13 illustrates a system planner input screen 1300 that is a first example of the cloud orchestrator management screen according to the first embodiment.

  The system planner input screen 1300 receives an input of a non-functional requirement to be used from a user. For example, assume that “Web system” is selected from the system basic configuration menu 1301. Next, the procurement deadline of the information system that satisfies the importance level of the non-functional requirement desired by the user is input by a user operation of the designated period menu 1302, the designated date and time menu 1303, and the start and end radio buttons 1304. Note that the input method of the procurement time limit is not limited to the method using at least one of these UIs (user interfaces) 1302 to 1304. Similarly, for the input of other types of information, the input method described with reference to FIGS. 13 to 17 is merely an example, and is not limited to the input method described with reference to FIGS.

  From the system basic configuration menu 1301, for the resource types “WebAP layer” and “DB layer” belonging to the infrastructure resource type “server resource” (“server” icon 1305), the size, the number, the non-functional requirements and the importance thereof The level is entered by the user. For example, when the check boxes of “input value”, “display graph” and “save value” of the non-functional requirement icon 1306 are selected, a non-functional requirement area 1307 is displayed on the screen 1300. When the user inputs the importance level of each non-functional requirement in an input value area 1308 in the area 1307, a radar chart showing the non-functional requirement definition (the relationship between the non-functional requirements and the importance level) is displayed in a graph area 1309. Is displayed, and the definition of the non-functional requirement is saved (for example, saved as part of the user input information 101).

  FIG. 14 illustrates a service menu screen 1400 that is a second example of the cloud orchestrator management screen according to the first embodiment.

  The service menu screen 1400 receives an input of a cloud service menu to be provided from an infrastructure administrator. It is assumed that the infrastructure resource type “server resource” is selected from the target resource list menu 1401. Next, when one resource type “HA configuration” (“HA configuration” icon 1402) belonging to the resource type is selected, and the “value input”, “graph display”, and “value storage” check boxes are selected. A non-functional requirement area 1403 is displayed on the screen 1400. When the infrastructure administrator inputs the achievement level for each non-functional requirement in the input value area 1404 of the area 1403, the definition of the non-functional requirement (the relationship between the non-functional requirement and the achievement level) is displayed in the graph area 1405. A radar chart is displayed and its non-functional requirement definition is saved (eg, saved as part of target resource 102).

  FIG. 15 illustrates a service plan formulation screen 1500 that is a third example of the cloud orchestrator management screen according to the first embodiment.

  The service plan formulation screen 1500 is a screen for the user to determine a cloud service plan that the user wants to use in self-service. It is assumed that the user has selected “Web system” from the basic system configuration menu 1501. The user uses the UIs 1502, 1503, and 1504 to input the procurement date of the information system that satisfies the importance level of the non-functional requirement desired by the user.

  A “non-functional requirement” icon 1506 belonging to the “DB layer” icon 1505 is selected by a user operation. The service plan area 1507 includes a service menu area 1508 and a service plan candidate area 1509. According to the “user requirement” of the radar chart displayed in the service plan candidate area 1509, it can be seen that the non-functional requirement “downtime” has the highest importance level.

  When the “server resource” icon 1511 is selected from the target resource list menu 1510 in the service menu area 1508, and the “non-functional requirement display” check box of the “HA configuration” icon 1512 expanded thereby is selected, the service plan candidate is displayed. In the area 1509, the achievement level of the non-functional requirement of each resource type belonging to the target resource type “HA configuration” is displayed on the radar chart. Further, when the “non-functional requirement display” check box of the “cluster software” icon 1513 is selected, the achievement level of the non-functional requirement of the extension plan of the resource type “cluster software” is displayed on the radar chart.

  In the example of FIG. 15, from the radar chart (chart indicating the importance level according to the user requirement and the achievement level of the selected resource type) on the service plan candidate area 1509, the user selects the resource type “cluster software” It is possible to select the adoption of the plan “without extension” and select the “Save plan” check box of the “DB layer” icon 1505.

  FIG. 16 illustrates an example of a design pattern display screen 1600 that is a fourth example of the cloud orchestrator management screen according to the first embodiment.

  The design pattern display screen 1600 displays the design pattern selected by the user. It is assumed that “Web system” is selected from the basic system configuration menu 1601. The user inputs a procurement deadline using the UIs 1602, 1603, and 1604.

  When the user selects the “plan display” and “configuration display” check boxes of the “DB layer” icon 1605 and the “display” check box of the “non-functional requirement” icon 1706, the service plan area 1607 is displayed in the design pattern area 1607. 1608 and a system configuration area 1609 are displayed. In the service plan area 1608, a radar chart showing the importance level according to the user requirement and the achievement level according to the resource type (service plan) selected by the user is displayed. That is, from the information displayed in the area 1608, the user can compare the importance level according to the user requirement with the achievement level according to the resource type selected by the user. In the system configuration area 1609, a diagram representing the actual system configuration of the resource type (service plan) selected by the user is displayed.

  FIG. 17 illustrates an example of a process flow editing screen 1700 that is a fifth example of the cloud orchestrator management screen according to the first embodiment.

  The processing flow editing screen 1700 is a screen for editing a design pattern construction flow. It is assumed that “Web system” in the basic system configuration menu 1701 has been selected. The user inputs the procurement date using the UIs 1702, 1703, and 1704.

  When the user selects the “configuration display” and “flow display” check boxes of the “DB layer” icon 1705, a system configuration area 1707 and a construction flow area 1708 are displayed in the design pattern area 1706. By editing the contents of the construction flow displayed in the construction flow area 1708 by a user operation, it is possible to customize the arrangement and setting information, which are the system construction contents. After customization, if the “Save Flow” check box is selected, the flow after customization is saved.

  Example 2 will be described. At this time, differences from the first embodiment will be mainly described, and description of common points with the first embodiment will be omitted or simplified.

  FIG. 18 is a schematic diagram illustrating the outline of the resource scheduling method according to the second embodiment. According to FIG. 18, for the purpose of explaining the outline of the present embodiment, an example in which a storage of a Web system is constructed is employed.

  In the present embodiment, the user inputs the importance level of the non-functional requirement for the system basic configuration 210 of the input information 101 and the storage of the components (for data and backup). Non-functional requirements for storage are performance, capacity, data protection, and the like provided by the system execution platform. Specifically, for example, performance is latency, capacity is stored data size, and data protection is RTO ( Recovery Time Objective) and RPO (Recovery Point Objective). According to the example of FIG. 18, latency, capacity, cost, RTO and RPO are adopted as the non-functional requirements. In this embodiment, the level of RTO is set to level 5 within 1 hour, level 4 to within 4 hours, level 3 to within 8 hours, and level 2 to within 1 day. Latency is storage I / O performance.

  It is assumed that the user has selected the target resource type “storage resource” from the infrastructure resource menu (target resource) 102. Non-functional requirement achievement levels are defined in five levels for storage resources. For example, regarding latency, SSD (Solid State Drive) is defined as 5, SAS (Serial Attached SCSI) is defined as 3, SATA (Serial AT Attachment) is defined as 2, and SDS (Software Defined Storage) is defined as 1.

  In the first embodiment, an achievement level is defined for a system configuration method (more precisely, a target resource type according to the system configuration method). On the other hand, in the second embodiment, an achievement level is defined for the system operation method in addition to or instead of the system configuration method. There may be a plurality of “system operation methods” even if the system configuration is the same. For example, as a system operation method, a storage-based backup (a backup without a host between storages) as a first example of a backup service and a host-based backup (a backup between the storages via a host) as a second example of a backup service are provided. Backup). As the storage-based backup, for example, a synchronous backup (for example, a backup of a type in which the I / O completion is notified to the host after the backup of data according to the I / O request from the host is completed), an asynchronous backup (for example, from the host) Even if the backup of the data according to the I / O request is not completed, the completion of the I / O is notified to the host, and the data is backed up asynchronously with the processing of the I / O request.) There is. As a host-based backup, there is a disk backup. Regarding the backup service, the RTO defines replication as 5, snapshot as 2, and disk backup as 3.

  As in the first embodiment, the cloud orchestrator server performs a presentation process. The presentation process includes a comparison between the importance level and the achievement level, and a display of output information including information indicating a result of the comparison (S1-3-2).

  The output information includes a non-functional requirement (importance level) required by the user and a non-functional requirement (achievement level) satisfied by the target resource type for each non-functional requirement. The leftmost radar chart is a chart for user requirements (for data), that is, storage for data. The middle radar chart is a chart for user requirements (for backup), that is, storage for backup. According to the output information, an SSD that emphasizes latency for data and a SATA that emphasizes capacity for backup use are close to user requirements.

  The output information is a radar chart showing the user requirement (for backup), the achievement level for each of the plurality of system operation methods associated with the user requirement (for backup), and the user requirement (importance level). (Rightmost radar chart). Here, it can be seen that the disk backup is close to the user requirements in the comprehensive evaluation.

  In the second embodiment, for example, the following is possible. That is, information input (selection of system basic configuration, selection of resource types (for example, “data storage” and “backup storage”) in the system basic configuration, specification of non-functional requirements for the selected resource type, and specification Is performed, the cloud orchestrator server sends the user input information (type ID of the infrastructure resource type (for example, type name), the non-functional requirement associated with the By searching for the target resource type definition (type ID (for example, type name), non-functional requirement, and achievement level associated with each of the plurality of target resource types) using the importance level), the input information is retrieved. Contains information indicating the target resource type (system configuration method) closest to You may display the output information. In this embodiment, the “system configuration method” is the type of storage, for example, SSD, SATA, SDS, or the like.

  Further, in this embodiment, the cloud orchestrator server determines that at least one of a plurality of system operation methods (for example, a system operation method closest to a user requirement (importance level)) for a specific infrastructure resource type. ), A comparison (eg, radar chart) of the importance level and the achievement level for each of the plurality of non-functional requirements can be displayed. An example of the “specific infrastructure resource type” here is “storage (for backup)”, and the “system operation method” is a backup method.

  Also, in the present embodiment, as in the first embodiment, if the achievement level of the specific non-functional requirement of the target resource type is equal to or higher than the importance level of the non-functional requirement, the cloud orchestrator server For at least one of the plurality of types, a comparison (e.g., a radar chart) between the importance level and the achievement level for each of the plurality of non-functional requirements can be displayed.

  Hereinafter, the second embodiment will be described in detail.

  FIG. 19 is an overall configuration diagram of the system according to the second embodiment.

  According to FIG. 19, a backup server 1901 is added to the overall system configuration according to the first embodiment. The backup server 1901 stores a backup management 1902, a backup service management 1903, and a backup service menu 1801. The elements 1902, 1903, and 1801 may be stored in the storage unit of the backup server 1901. The elements 214, 213a, 213b, and 213c may be computer programs executed by the processor unit of the cloud operation management server 202.

  FIG. 20 is a configuration diagram of the backup server 1901.

  The screen display processing 2001, the processing of the backup service management 1903, and the processing of the backup management 1902 of the backup server 1901 illustrated in FIG. 19 are performed by hardware including a memory 301f, a processor 302f, a storage device 303f, and a communication interface 304f. Specifically, it can be realized (for example, it can be realized by executing one or more programs by the processor 302f).

  The backup service non-functional requirement definition process 2002 and the backup service setting method definition process 2003 included in the backup service management 1903 can be realized by the processor 302f executing the program read into the memory 301f.

  The backup setting / cancellation process 2004 included in the backup management 1902 can be realized by the processor 302f executing a program read into the memory 301f.

  In the storage device 303, backup service definition information (backup service menu) 1801 and a backup setting method 2005 are stored.

  Input / output information exchanged from the terminal 206 via the Web browser 207, the construction processing script 211 and design pattern 212 of the cloud orchestrator server 201, and infrastructure resource information 215 of the cloud operation management server 202 are input / output via the communication interface 304. May be.

  FIG. 21 is a configuration diagram of the cloud orchestrator server 201.

  The planner 208 can further execute a backup service candidate plan generation process 2101 as an internal process.

  FIG. 22 shows processing and data flow between servers.

  In the backup service non-functional requirement definition process 2002, a backup service menu 1801 is input. Thereby, the non-functional requirements and the achievement level of the backup service of the backup server 1901 are defined.

  The backup service setting method definition process 2003 inputs a backup service menu 1801 and generates a design pattern for calling the backup server 1901.

  The backup service candidate plan generation processing 2101 receives the user input information 101 and selects a backup service candidate plan from the backup server 1901.

  The design pattern selection processing 308 inputs the selected backup service candidate plan and selects the design pattern 212.

  The construction processing flow generation processing 309 inputs the selected design pattern 212 and edits the construction processing flow 212.

  The construction processing flow automation processing 310 and the construction processing flow 212 are input, the deployment is executed in the virtual resource allocation processing 404 of the cloud operation management server 202, and the monitoring setting is executed in the monitoring setting / cancellation processing 504 of the monitoring server 203. The backup setting is executed in the backup setting / cancellation process 2004 of the backup server 1901.

  FIG. 23 shows the configuration of the backup service menu 1801.

  The backup service menu 1801 includes a backup service type 2304, a service menu 2301, and a non-functional requirement group 2302.

  The backup service type 2304 indicates the name of the type of the backup method (backup service) which is an example of the system operation method (large category). The service menu 2301 indicates the name of a service type (service menu) belonging to the backup service type (large category).

  The non-functional requirement group 2302 represents a plurality of non-functional requirements that can be specified (selected) by the user and the achievement level specified by the user for each non-functional requirement. Specifically, for example, the non-functional requirement group 2302 includes downtime 2302a, latency 2302b, throughput 2302c, expandability 2302d, capacity 2302e, RTO 2302f, RPO 2302g, and cost 2302h. Each of the various information 2302a to 2302h includes an achievement level for each service type.

  FIG. 24 shows a configuration of information of the design pattern 212.

The present design pattern 212 represents a disk backup. For example, the design pattern 212 includes a corresponding backup service type 2402 (for example, name “host base”), a service menu 2403 (for example, name “disk backup”) belonging thereto, and
A design pattern group 2401 is included. The design pattern group 2401 includes a system configuration 2401a and a construction processing flow 2401b.

  The system configuration 2401a includes a system configuration diagram for disk backup. Here, among the backup data, the system data and the file data are stored in the image storage in the data storage, and are stored in the backup storage by vDP (vStorage APIs for Data Protection) backup. Business data handled by the business application in the DB is stored in a backup storage from a VM internal disk via a backup server via a network backup using backup software.

  The construction processing flow 2401b is a system construction processing flow in the case of disk backup. This is a construction processing procedure for automatically constructing and operating the system configuration 2401a.

  FIG. 25 shows the configuration of the infrastructure resource information 215.

  As in the first embodiment, the infrastructure resource information 215 includes an infrastructure resource type 1004, a target resource group 1005, an acquisition source 1001, a resource tag 1002, and a resource allocation status (free resource) 1003. The value included in the target resource group 1005 is a resource type associated with each infrastructure service type.

  FIG. 26 shows the flow of processing of the cloud orchestrator server 201.

Compared to FIG. 11, a backup service candidate plan generation process 2101 is added. In the backup service candidate generation process 2101, the planner 208 inputs the user input information 101 and executes a process 2601 of selecting a target backup service menu for a component or a combination thereof. In the process 2101, the planner 208 next compares the requirements of the backup service in the descending order of the importance level of the set non-functional requirements of the user input, and executes a process 2602 of selecting the one near the requirement. Specifically, for example, the planner 208 selects a resource type (service menu) satisfying the following (condition 11) to (condition 14) from the backup service menu 1801.
(Condition 11) A service menu corresponding to the selected resource type (for example, storage).
(Condition 12) A service menu in which, among the service menus corresponding to (Condition 11), a plurality of non-functional requirements identical to a plurality of non-functional requirements associated with the selected resource type are associated.
(Condition 13) A service menu in which the same achievement level as the importance level is associated with each non-functional requirement among the service menus corresponding to (Condition 12).
(Condition 14) If there is no service menu corresponding to (Condition 13), among the service menus corresponding to (Condition 12), a plurality of achievement levels close to a plurality of importance levels associated with the selected resource type are set. Associated service menu.

  For the service menu corresponding to (condition 14), for example, the higher the non-functional requirement, the higher the achievement level matches the importance level (or higher than the importance level), or the achievement level. Is less than the importance level, the service menu that most satisfies that the difference is as small as possible may be sufficient.

  Further, as a candidate plan calculation method of the process 2602, a linear integer programming method 2603 can be adopted as in the first embodiment.

  27 to 32 illustrate a cloud orchestrator management screen according to the second embodiment.

  FIG. 27 illustrates a system planner input screen 2700 that is a first example of the cloud orchestrator management screen according to the second embodiment.

  A user inputs system requirements to be used on a system planner input screen 2700. It is assumed that “Web system” is selected from the basic system configuration menu 2701. The user uses the UIs 2702, 2703, and 2704 to enter the procurement date of the information system that satisfies the importance level of the non-functional requirement desired by the user.

  In the basic system configuration menu 2701, it is assumed that the size and the non-functional requirements are input for the “data” icon and the “backup” icon of the storage icon 2705. When the “Value input”, “Graph display”, and “Save value” check boxes of the “Non-functional requirement” icon 2705 are selected, a non-functional requirement area 2707 is displayed, and the user enters a resource type in the input value area 2708. When the importance level of each non-functional requirement is input for each resource, a graph area 2709 is displayed, a radar chart of the importance level of each non-functional requirement for each resource type is displayed, and the non-functional requirement definition (input contents) is saved. Is done.

  FIG. 28 illustrates a service menu screen 2800 which is a second example of the cloud orchestrator management screen according to the second embodiment.

  On the service menu screen 2800, an infrastructure administrator inputs a cloud service menu to be provided. When the storage resource icon 2802 is selected from the target resource list menu 2801 and the check boxes for value input, graph display, and value storage are selected, a non-functional requirement area 2803 is displayed. When the user inputs the achievement level of each non-functional requirement for each target resource type in the input value area 2804, a graph area 2805 is displayed, and a radar chart of the importance level of each non-functional requirement for each target resource type is displayed. The non-functional requirement definition (input contents) is saved.

  FIG. 29 illustrates a service menu screen 2900 that is a third example of the cloud orchestrator management screen according to the second embodiment.

  In the service menu screen 2900, an infrastructure administrator inputs a provided cloud service menu. When the “host base” icon 2902 and “storage base” icon 2903 of the backup service list menu 2901 are selected and the “value input”, “graph display”, and “value storage” check boxes are selected, the non-functional requirement area is displayed. 2904 is displayed. When the user inputs the achievement level of each non-functional requirement in the input value area 2905 for each service menu, a graph area 2906 is displayed, and a radar chart of the importance level of each non-functional requirement for each service menu is displayed, The non-functional requirement definition (input contents) is saved. This input content is stored in, for example, the backup service menu 1801.

  FIG. 30 illustrates a service plan formulation screen 3000 that is the fourth example of the cloud orchestrator management screen according to the second embodiment.

  The user decides a cloud service plan to use using the service plan formulation screen 3000 in self-service. It is assumed that “Web system” is selected from the basic system configuration menu 3001. A procurement deadline is input using the UIs 3002, 3003, and 3004.

  It is assumed that the “display” check box is selected for the “non-functional requirement” icon 3006 belonging to the “storage” icon 3005. The service plan area 3007 includes a service menu area 3008 and a service plan candidate area 3009. From the user requirements displayed in the service plan candidate area 3009, it can be understood that the capacity for backup and the importance of latency and RTO for data are high.

  When the “storage resource” icon 3011 is selected from the target resource list menu 3010 in the service menu area 3008 and the “non-functional requirement display” check box of the “storage resource” icon 3011 is selected, the service plan candidate area 3009 is displayed. Then, a radar chart showing the level of achievement of the non-functional requirements of each service of the storage resource is displayed. Even if the user requirements (importance) are already displayed on the radar chart and the achievement level is displayed on the radar chart in response to a predetermined user operation such as selecting a check box of the non-functional requirement display. Good. Such a display method can be adopted in the first embodiment.

  Similarly, in the service menu area 3008, when the “non-functional requirement display” check box is selected for the “host base” icon 3013 and “storage base” icon 3014 of the backup service list menu 3012, the service plan candidate area 3009 is displayed. A radar chart showing the degree of achievement of the non-functional requirements of each backup service is displayed.

  From the service plan candidate area 3009 (specifically, the achievement level of the selected service menu requirement with respect to the user requirement), a SATA plan that satisfies the capacity requirement is adopted for the backup and the latency plan for the data is adopted. The user can decide to adopt an SSD plan that meets the requirements. In addition, the user can decide to adopt a disk backup plan that is close to the user requirements in the comprehensive evaluation of the RTO and the cost for the backup service for data.

  FIG. 31 illustrates a design pattern display screen 3100 that is a fifth example of the cloud orchestrator management screen according to the second embodiment.

  The design pattern display screen 3100 displays the design pattern selected by the user. “Web system” is selected from the basic system configuration menu 3101, and the procurement period is input using the UIs 3102, 3103, and 3104.

  For each of the "data" icon and the "backup" icon belonging to the "storage" icon 3105, the "plan display" and "configuration display" check boxes are selected, and the "non-functional requirement" icon "display" check box is selected. Is selected, a service plan area 3107 and a system configuration area 3108 are displayed in the design pattern area 3106. In the service plan area 3107, a radar chart showing user requirements (importance level) and requirements of the selected service plan (achievement level) is displayed. In the system configuration area 3108, an actual system configuration diagram of the selected service plan is displayed.

  FIG. 32 illustrates a process flow editing screen 3200 that is a sixth example of the cloud orchestrator management screen according to the second embodiment.

  Using the processing flow editing screen 3200, the user can edit the construction flow of the design pattern. “Web system” is selected from the basic system configuration menu 3201, and the procurement period is input using the UIs 3202, 3203, and 3204.

  When the “configuration display” and “flow display” check boxes of the “storage” icon 3205 are selected, a system configuration area 3207 and a construction flow area 3208 are displayed in the design pattern area 3206. By editing the contents of the construction flow in the construction flow area 3208, the arrangement and setting information, which is the system construction contents, can be customized. After customization, if the “Save Flow” check box is selected, the flow after customization is saved.

  According to the above description, for example, the following can be said.

  First, each non-functional requirement is associated with a qualitative level, rather than a quantitative number. The system execution platform (the cloud, which is a plurality of computing resources including multiple types of computing resources) specifies in advance non-functional requirements such as availability, data protection, performance, and security by quantitative indicators, It is necessary to protect it. For example, in the case of availability, the downtime is within 1 minute, within 5 minutes, within 1 hour, and the like. At this time, it is designated by levels 1 to 3 instead of specific numerical values. By specifying various non-functional requirements at a common level, the requirements of the system required by the user (the level specified by the user) and the requirements of the service menu provided by the cloud service (the level associated in advance) And can be compared.

  Secondly, even when there is no infrastructure resource or operation method that satisfies all the non-functional requirements required by the user, it is possible to select from available resources and operation methods by relaxing the requirements. The cloud service menu includes one related to an infrastructure resource (system configuration method) and one related to a system operation method. Regarding infrastructure resources, there are physical hardware and physical configurations, virtualization platforms and virtual layer configurations, virtual resources and middleware that runs on them, and software that provides infrastructure functions on general-purpose hardware. As for the system operation method, there are a backup method, a monitoring method, and the like. The management system selects, from the service menus required to satisfy the requirements, those having the achievement level as close as possible to the requirements, for the importance level of the non-functional requirements required by the user. As a selection method, there is a method of giving priority to a material having a higher importance, comprehensively evaluating a plurality of non-functional requirements, and selecting a material having a higher achievement. A method for calculating a difference between a non-functional requirement of a user and a non-functional requirement that can be provided by a service menu for each non-functional requirement, and calculating an optimal solution using a linear programming method with importance levels as coefficients. Can be adopted.

  As described above, the management system can quickly provide an infrastructure resource that meets the requirements for the system execution base, and can present, for example, a user with a candidate plan combining service menus. In addition, the infrastructure that forms the cloud (system execution platform) can selectively use limited physical hardware and general-purpose physical hardware according to requirements, thereby achieving resource utilization efficiency.

  The management system is a system that manages a computer system having a plurality of computing resources including a plurality of types of computing resources. The computer system to be managed is, for example, the above-described system execution platform. XaaS (Xas a Service) can be adopted as a service according to the computing resources of the system execution base. XaaS is generally such that some computing resources (eg, hardware, lines, software execution environment, application software, development environment, etc.) necessary for building or operating an information system can be used through a network such as the Internet. Meaning service. The character (or word) adopted as “X” of XaaS differs depending on the type (service model) of XaaS. For example, examples of XaaS include SaaS (Software as a Service), PaaS (Platform as a Service), IaaS (Infrastructure as a Service), and HaaS (Hardware as a Service).

  In the above description, the format of the constructed construction process flow is a format that can be interpreted by the execution engine (construction automation 209) of the flow.

  In the above description, “target resource type” in the target resource 102 (an example of a system configuration method) and “service type (small category)” in the backup service menu (an example of a system operation method) are “service type”. This is an example.

  In the above description, the non-functional requirements include downtime, latency, throughput, RTO, RPO, and KPI (Key Performance Indicator) in order to evaluate availability, performance, data protection, ROI (Return On Investment), and the like. Capacity, cost, etc. are represented at an abstracted level. By expressing at the level of abstraction, it is possible to evaluate with an allowable range or relative value.

  Although several embodiments have been described above, these are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to these embodiments. The present invention can be implemented in other various forms. For example, a combination of the first embodiment and the second embodiment is also possible.

  For example, for each service type (for example, target resource type or service type (small category)), the scheduled use date and time of the computing resource assigned to the service type may be associated. The management system may determine whether or not there is a service type that completely satisfies the user input information, based on whether or not the scheduled use date and time comes before the procurement deadline specified by the user. When the determination result is false, the management system displays, for the user, output information including information (for example, the above-described radar chart) indicating a service type approximate to the user input information using a linear programming method or the like. You may.

  201: cloud orchestrator server, 202: cloud operation management server

Claims (13)

A management system for managing a computer system having a plurality of resources including a plurality of types of resources,
An interface unit including one or more interfaces for receiving user input information;
An output information including one or more processors connected to the interface unit and including information representing one or more service types matching the input information among service types belonging to a search range among a plurality of service types. And a processor unit for displaying for
The input information is
An infrastructure resource type which is a type of a component specified by a user among one or more components of the system basic configuration;
A plurality of first levels respectively specified by a user for a plurality of non-functional requirements associated with the infrastructure resource type,
For each of the plurality of service types,
Two or more non-functional requirements;
Two or more second levels, each associated with the two or more non-functional requirements,
Each of the one or more service types that match the input information is the same as a plurality of non-functional requirements respectively corresponding to the plurality of first levels included in the input information among service types belonging to the search range. Has non-functional requirements associated with it ,
The one or more service types that match the input information are service types determined based on a first-level height among a plurality of non-functional requirements represented by the input information, and are further represented by the input information. For each of the plurality of non-functional requirements, based on whether the second level is greater than or equal to the first level, and if the second level is less than the first level, based on the magnitude of the difference between the second level and the first level. , Is a service type determined according to the calculation of the optimal solution using linear programming,
Management system. The plurality of service types include two or more system configuration methods,
The one or more service types that match the input information include one or more system configuration methods,
The management system according to claim 1. Among the one or more system configuration methods that match the input information, if there is a system configuration method in which the second level of the specific non-functional requirement is equal to or higher than the first level of the specific non-functional requirement, the output information includes: For the system configuration method, for each of the plurality of viewpoints of the specific non-functional requirement, information including a plurality of second levels respectively corresponding to the plurality of non-functional requirements,
The management system according to claim 2. The specific non-functional requirement is scalability,
The plurality of aspects are scale-up and scale-out,
The management system according to claim 3. The plurality of service types include two or more system operation methods for at least one infrastructure resource type,
The one or more service types that match the input information include one or more system operation methods,
The management system according to claim 1. The two or more system operation methods are two or more different backup methods associated with storage as an infrastructure resource type.
The management system according to claim 5. The information representing one or more service types that match the input information includes two or more second levels associated with each of the one or more service types, and a plurality of first levels included in the input information. Is information representing
The management system according to claim 1. The plurality of non-functional requirements and the two or more non-functional requirements are at least two of downtime, latency, throughput, RTO (Recovery Time Objective), RPO (Recovery Point Objective), capacity and cost,
The management system according to claim 1. The input information is information input via a first screen for the user,
The two or more second levels for each of the plurality of service types are information input via a second screen for an administrator.
The management system according to claim 1. The processor unit includes:
Receiving a selection of a service type from a user among one or more service types matching the input information,
Displaying a design pattern, which is a pattern composed of two or more resources, for the selected service type to the user;
The management system according to claim 1. The processor unit includes:
Generate a construction process flow in which the setting procedure for deploying the design pattern is defined and display it to the user,
The management system according to claim 10 . The processor unit includes:
Deploy the construction process flow,
The management system according to claim 11 . A resource scheduling method by a management system for presenting a user with a service type using two or more resources of a plurality of resources of the computer system,
(A) receiving input information of a user,
The input information is
An infrastructure resource type which is a type of a component specified by a user among one or more components of the system basic configuration;
A plurality of first levels respectively specified by a user for a plurality of non-functional requirements associated with the infrastructure resource type,
(B) displaying to the user output information including information representing one or more service types that match the input information among the service types belonging to the search range among the plurality of service types;
For each of the plurality of service types,
Two or more non-functional requirements;
Two or more second levels, each associated with the two or more non-functional requirements,
Each of the one or more service types that match the input information is the same as a plurality of non-functional requirements respectively corresponding to the plurality of first levels included in the input information among service types belonging to the search range. Has non-functional requirements associated with it,
The one or more service types that match the input information are service types determined based on a first-level height among a plurality of non-functional requirements represented by the input information, and are further represented by the input information. For each of the plurality of non-functional requirements, based on whether the second level is greater than or equal to the first level, and if the second level is less than the first level, based on the magnitude of the difference between the second level and the first level. , Is a service type determined according to the calculation of the optimal solution using linear programming,
Resource scheduling method.