JPWO2018173481A1 - Service configuration design apparatus and service configuration design method - Google Patents

Abstract

In order to reduce the time required for the configuration design of the system that provides the service, the service configuration design device 10 analyzes the traffic of each node for providing the ICT service and defines a business pattern definition unit 11 that defines the business. A task / resource relational expression generation unit 12 for formulating a relation between a task and resource consumption at each node to generate a task / resource relational expression, and a design derivation tree representing a combination of resources that can be included in the node. A service configuration deriving unit 13 that generates a service configuration based on a business-resource relational expression, and then generates a service configuration after narrowing down resources in the design derivation tree.

Description

  The present invention relates to a service configuration design apparatus and a service configuration design method for designing a configuration of a system that provides an ICT service.

  A system that provides an ICT (Information and Communication Technology) service (for example, a website, a moving image distribution, and a business system) provided via a communication network such as the Internet includes a plurality of components. For example, a Web system includes a plurality of components such as a Web server, an application server, and a database server (for example, refer to Patent Document 1).

  Patent Literature 1 describes a system configuration plan generation device that generates a configuration plan of a system that provides an ICT service in consideration of component requirements and the like desired by a user.

  Patent Literature 2 introduces a service chain including a plurality of functions for realizing one service.

JP-A-2005-219795 WO 2015/194182

  A system for providing an ICT service provided to a terminal such as a mobile phone or a personal computer via a communication network will be described as an example. FIG. 14 is a system configuration diagram showing a general service providing system 500 that provides an ICT service via the Internet 600.

  As shown in FIG. 14, the service providing system 500 includes servers such as Web servers 510a and 510b, an application server 520, and a database (DB) server 530. Further, the service providing system 500 includes a firewall (FW) 540 for eliminating unnecessary traffic and a load balancer (LB) 550 for converting an IP (Internet Protocol) address. Further, the service providing system 500 may incorporate a function of a proxy (Proxy) for storing a cache and a function of a NAT (Network Address Translation) for performing address translation. Hereinafter, modules that provide various server functions and network functions are collectively referred to as nodes.

  Resources such as a CPU (Central Processing Unit), a memory, and a network I / O (Input / Output) may be added to improve processing at the node. Further, an accelerator supporting AES-NI (Advanced Encryption Standard New Instructions) may be used, or parallel processing by scale-out may be executed.

For example, in a node of a certain FW, there are ten types of resource allocation patterns, three types of accelerators (including non-use of accelerators), and the maximum number of scales is 20. If the maximum number of scales is 20, a candidate for an instance configuration (pattern) of the node is There are 10 × 3 × 20 = 600 patterns. Further, when a providing system composed of various types of nodes is configured by linking NAT nodes, proxy nodes, and the like, there are 600 ⁵ = 777,600,000,000,000,000 types of instance configuration candidates. Testing each pattern is required to verify which pattern is optimal. Therefore, the verification time becomes enormous.

  Further, even if the types are the same, the resources that are consumed in the internal processing are different between the nodes that execute different processes, so that the optimal resource allocation patterns are also different. For example, taking FW as an example, a high-performance CPU resource is required when performing a virus check on mail. When filtering video distribution, network I / O resources with high performance are required. When both processes are performed at the same time, the optimal resource allocation pattern is different, such as a need for more resources than the sum of the CPU resources and the network I / O resources required by the respective functions.

  Therefore, it is necessary to verify a huge number of instance configuration candidates and resource allocation patterns before an optimal instance configuration and an optimal resource allocation pattern can be derived. That is, there is a problem that it takes a long time to design an optimal system for providing an ICT service.

  In this specification, the function of each node constituting the service chain corresponds to a service. Therefore, the “service configuration” is a concept representing how a plurality of nodes are combined. Further, the “service configuration” includes the concept of resource allocation in each node. The “service configuration design” is to determine a combination of a plurality of nodes to satisfy a required service level, and to determine the amount of resources allocated to each node.

  An object of the present invention is to reduce the time required for the configuration design of a service providing system.

  A service configuration designing apparatus according to the present invention analyzes a traffic of each node for providing an ICT service and defines a business pattern, and formulates a relationship between the business and resource consumption in each node. And a service configuration deriving means for generating a service configuration based on a design derivation tree representing a combination of resources that can be included in a node. The means generates a service configuration after narrowing down resources in the design derivation tree based on the business / resource relational expression.

  The service configuration design method according to the present invention analyzes a traffic of each node for providing an ICT service, defines a service, formulates a relationship between the service and resource consumption in each node, and defines a service / resource relationship. An expression is generated, and based on the business / resource relational expression, resources in a design derivation tree representing a combination of resources that can be included in a node are narrowed down, and then a service configuration is generated based on the design derivation tree.

  A recording medium according to the present invention is a computer-readable recording medium storing a service configuration design program, and the service configuration design program, when executed by a processor, executes traffic of each node for providing an ICT service. Process to define tasks by analyzing tasks, processing to formulate the relationship between tasks and resource consumption at each node to generate task and resource relational expressions, and processing to include in nodes based on task and resource relational expressions. After narrowing down resources in the design derivation tree representing combinations of possible resources, a process of generating a service configuration based on the design derivation tree is executed.

  ADVANTAGE OF THE INVENTION According to this invention, the time required for the configuration design of the system which provides a service is shortened.

FIG. 2 is a block diagram showing a service configuration designing apparatus for designing an optimal configuration of a service execution apparatus together with related apparatuses. FIG. 4 is an explanatory diagram illustrating an example of a list of traffic information (traffic information table). FIG. 4 is an explanatory diagram illustrating an example of a list of resource consumption rates (resource consumption rate table). It is an explanatory view showing an example of a requirement of a service. 9 is a flowchart illustrating a process executed by a business pattern definition unit. FIG. 9 is an explanatory diagram illustrating an example of business pattern information. It is a flowchart which shows the process which a business / resource relational expression generation part performs. It is explanatory drawing which shows an example of a business / resource relational expression. FIG. 4 is an explanatory diagram illustrating an example of a design derivation tree. FIG. 4 is an explanatory diagram illustrating an example of a resource amount of a resource in each resource type. It is a flowchart which shows the process which a service structure derivation part performs. FIG. 3 is an explanatory diagram illustrating an example of a service configuration stored in a service configuration storage device. FIG. 2 is a block diagram illustrating a main part of the service configuration design device. 1 is a system configuration diagram showing a general service providing system that provides an ICT service.

Embodiment 1 FIG.
Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a service configuration designing apparatus (a system configuration designing apparatus for selecting a configuration of a service chain for realizing an ICT service) for designing an optimal configuration of a service execution apparatus together with related apparatuses. Specifically, in addition to the service configuration design device 100, a service execution device 200, a measurement device 201, a configuration change device 202, a service configuration storage device 203, a measurement information storage device 204, and an input device 300 are shown. In the example shown in FIG. 1, the service execution device 200 provides an environment in which a plurality of services (Service 1 and Service 2 in the example shown in FIG. 1) each implemented by a plurality of nodes operate.

  The input device 300 is a device that supplies the requirements of a newly generated service (ICT service) to the service configuration design device 100.

  The measuring apparatus 201 targets each node for providing the ICT service with traffic information (for example, a traffic amount which is an amount of data passing through a network), a service time (a time required for data processing), and a resource consumption rate. Are measured, and the measured values are stored in the measurement information storage device 204.

  The configuration changing device 202 reads out the configuration information newly stored in the service configuration storage device 203, and builds a service chain in the service execution device 200 based on the configuration information. The service configuration storage device 203 stores the configuration information selected by the service configuration design device 100.

  The service configuration design apparatus 100 includes a service configuration derivation unit (system configuration derivation unit) 101, a business / resource relational expression generation unit 102, a business pattern definition unit 103, a service requirement storage unit 110, a design derivation tree storage unit 111, a business / resource A relational expression storage unit 112 and a business pattern storage unit 113 are included.

  The service configuration deriving unit 101 selects an appropriate service configuration (connection of a plurality of nodes for providing an ICT service) and the like. The business / resource relational expression generation unit 102 generates a business / resource relational expression (a relational expression between the business of the node and the resource consumption rate), and stores the business / resource relational expression in the business / resource relational expression storage unit 112. The business pattern definition unit 103 defines a business pattern (for example, an ICT service), and stores business pattern information related to the business pattern in the business pattern storage unit 113.

  The service requirement storage unit 110 stores the requirement of the service supplied from the input device 300. The design derivation tree storage unit 111 stores design derivation tree data.

  Note that the service configuration derivation unit 101, the business / resource relational expression generation unit 102, and the business pattern definition unit 103 operate according to a program stored in a non-temporary nonvolatile memory such as a ROM (Read Only Memory), a flash memory, and a hard disk. One or more processors (for example, CPU) execute the processing.

  The service requirement storage unit 110, the design derivation tree storage unit 111, the business / resource relational expression storage unit 112, and the business pattern storage unit 113 can be realized by a storage device such as a hard disk.

  Next, the operation of the service configuration design apparatus 100 will be described. First, the business pattern definition will be described.

  The measuring apparatus 201 includes information (traffic information) on traffic (data input to the node via the network and data output from the node), CPU, RAM (Random access) for each node in the service execution apparatus 200. memory), network I / O, storage IOPS (Input / Output operations per second), and the like, information on resource utilization (resource consumption rate) is measured, and the measured values are stored in the measurement information storage device 204.

  FIG. 2 is an explanatory diagram illustrating an example of a list of traffic information (traffic information table) stored in the measurement information storage device 204.

  In the example shown in FIG. 2, a communication log is set in the traffic information table together with a corresponding time stamp, service ID, node ID, and log type. In the example shown in FIG. 2, a plurality of ICT services (Service1, Service2, etc.) are provided by the service execution device 200.

  FIG. 3 is an explanatory diagram illustrating an example of a list of resource consumption rates (resource consumption rate table) stored in the measurement information storage device 204.

  In the example shown in FIG. 3, the CPU consumption rate, the RAM consumption rate, the IOPS consumption rate, and the NW (Network) I / O consumption rate are represented in the resource consumption rate table by the corresponding time stamp, service ID, and node. It is set together with the ID.

  The service configuration design apparatus 100 receives a requirement for a newly generated service from the input device 300 and stores the requirement in the service requirement storage unit 110. The service requirement includes at least information indicating connection of nodes as service configuration requirements, information indicating service level (for example, data transfer speed), and information indicating node type (type of node).

  FIG. 4 is an explanatory diagram illustrating an example of service requirements stored in the service requirement storage unit 110. In the example illustrated in FIG. 4, as illustrated in FIG. 4A, the service requirement storage unit 110 stores service configuration requirements, service level requirements, and business pattern IDs (pattern IDs) as service requirements. Further, as shown in (B), the service requirement storage unit 110 also stores information indicating a node type included in a service configuration requirement. In the example illustrated in FIG. 4, the service specified by the service ID “service1” includes a plurality of ICT services (a video distribution service specified by patternA and a predetermined application service specified by patternB).

  For example, using Juniper (registered trademark) FW, LVS (Linux (registered trademark) Virtual Server) load balancer, Apache (registered trademark) Web server, Tomcat server, and MySQL (registered trademark) DB, video traffic and application traffic are combined. When a service capable of processing 100 Mbps is required, the service requirement is specified as shown in the first line in FIG.

  In FIG. 4B, “fw” indicates a firewall. "nat" indicates a NAT device. "lb" indicates a load balancer. "px" indicates a proxy. “dpi” indicates a DPI (Deep Packet Inspection) device. "app" indicates an application server. "ids" indicates an intrusion detection system. "web" indicates a Web server. "db" indicates a database.

  FIG. 5 is a flowchart showing the processing executed by the business pattern definition unit 103.

  The business pattern definition unit 103 reads the traffic information corresponding to the node in the new service from the measurement information storage device 204 (Step S11). Then, the business pattern definition unit 103 executes the process of step S13 for each node type stored in the service requirement storage unit 110 (steps S12 and S14).

  In step S13, the business pattern definition unit 103 defines a business pattern based on the protocol information and the content information in the traffic information.

  For example, when using traffic information as shown in the traffic information table illustrated in FIG. 2, the business pattern definition unit 103 sets the protocol information to “HTTP” and the content information to “Content-Type: video / flv”. Is detected, and a business pattern called “video traffic” is defined.

  Then, the business pattern definition unit 103 stores the business pattern information related to the defined business pattern in the business pattern storage unit 113 (Step S15).

  FIG. 6 is an explanatory diagram showing an example of business pattern information stored in the business pattern storage unit 113. In the example shown in FIG. 6, data of a name indicating a business pattern and data indicating an identification pattern (protocol information, content information, etc.) are registered in correspondence with the pattern ID.

  Next, a process of generating a business / resource relational expression will be described.

  FIG. 7 is a flowchart showing a process executed by the business / resource relational expression generation unit 102.

  The business / resource relational expression generation unit 102 reads the log and the resource consumption rate included in the traffic information from the measurement information storage device 204 (Step S21). The traffic volume can be known from the log. Further, the business / resource relational expression generation unit 102 reads business pattern information from the business pattern storage unit 113 (Step S22).

  Then, the business / resource relational expression generation unit 102 classifies the traffic information for each business pattern (step S23). Further, the business / resource relational expression generation unit 102 executes the processing of step S25 for each node type (steps S24 and S26).

  In step S25, the business / resource relational expression generation unit 102 performs a mathematical analysis such as regression analysis using the traffic amount of the business pattern as a target variable and the consumption rate of each resource as an explanatory variable to generate a business / resource relational expression. I do. At this time, the business / resource relational expression generation unit 102 sets a threshold value of the accuracy of the relational expression, and deletes explanatory variables having low correlation with the objective variable. That is, the business / resource relational expression generation unit 102 generates a business / resource relational expression narrowed down to resource types effective for the performance of the node type.

  The business / resource relational expression generation unit 102 stores the generated business / resource relational expression in the business / resource relational expression storage unit 112 (step S27).

  FIG. 8 is an explanatory diagram showing an example of the business / resource relational expression stored in the business / resource relational expression storage unit 112. In the example shown in FIG. 8, the resources are a CPU, a RAM, an NWIO, and an IOPS.

  The business / resource relational expression generation unit 102 calculates the significance probability (p value) of the coefficient of the explanatory variable, and stores it in the business / resource relational expression storage unit 112 together with “significant F” indicating the significance of the relation. .

  As an example, when the business / resource relational expression generation unit 102 sets “significant F ≦ 0.05” as a threshold of the accuracy of the relational expression, the explanation is such that the correlation with the objective variable is low so as to satisfy this threshold. Remove variables.

  For example, in the relational expression of “node ID = fw1” and “business pattern = patternA (video traffic)”, the explanatory variables of the RAM are deleted, and the explanatory variables are effective for the performance of the node type fw1, CPU, network I / O. O and disk IOPS.

  Next, the service configuration optimization processing will be described. In the present embodiment, in the service configuration optimizing process, a design guidance tree and a business / resource relational expression are used.

  FIG. 9 is an explanatory diagram illustrating an example of a design derived tree (hereinafter, also referred to as a derived tree) stored in the design derived tree storage unit 111.

  The design derivation tree storage unit 111 stores a derivation tree in which all combinations of resources are set in a tree for each of a plurality of types of nodes for providing one or a plurality of ICT services. I have. Note that a derived tree shown as ID: derived tree 1 in FIG. 9 is stored in the design derived tree storage unit 111 in advance. The derived tree shown as ID: derived tree 2 is an example of a derived tree in a state after the pruning process has been performed by the service configuration deriving unit 101.

  FIG. 10 illustrates the resource amounts of the resources in each of the combinations (resource types) of the resources that can be included in each node. The resource amounts illustrated in FIG. 10 are stored in the design derivation tree storage unit 111. For example, CPU [Large] indicates that the CPU resource amount of the combination of resources (CPU, RAM, NWIO, and IOPS) is large.

  FIG. 11 is a flowchart illustrating a process executed by the service configuration deriving unit 101. Here, the requirement of the service specified by service1 shown in the first line in FIG.

  The service configuration deriving unit 101 reads out service requirements from the service requirement storage unit 110. That is, the service requirement shown in the first row in FIG. 4A stored in the service requirement storage unit 110 is read.

  Next, the service configuration derivation unit 101 sets a business pattern and a traffic amount as an ICT service (step S31). In this example, “video traffic” (patternA) and “application service” (patternB) are set as business patterns. Then, the service configuration deriving unit 101 reads out the data representing the derived tree from the design derived tree storage unit 111, and reads out the business type / resource relational expression of the node type included in the service from the business / resource relational expression storage unit 112 (step S32). ).

  The service configuration derivation unit 101 collects information on business / resource relational expressions, business patterns, and traffic amounts for each node type. Then, the service configuration deriving unit 101 executes the processing of steps S34 to S37 for each node type (steps S33 and S38).

  In step S34, the service configuration deriving unit 101 determines whether or not the number of business patterns is one (step S34). When the number of business patterns is one, the service configuration deriving unit 101 determines the amount of resources based on the magnitude (significance probability (p value)) of the explanatory variable in the business-resource relational expression of the business pattern. Is changed (step S36). That is, the service configuration deriving unit 101 performs a pruning process of the derived tree. Note that the service configuration deriving unit 101 assigns a minimum amount to a resource having no corresponding variable. The resource amount is defined as illustrated in FIG.

  For example, the service configuration deriving unit 101 changes the allocation width of the resource amount based on the significant probability (p value) of the explanatory variable of the business / resource relational expression of the business pattern. Specifically, as an example, the derived tree is determined in accordance with a determination rule such as “Resource amount Large when p value ≦ 0.01, Medium when 0.01 <p value ≦ 0.05, Small when 0.05 ≦ p value”. Pruning. As a result, a derived tree shown as ID: derived tree 2 is generated from the derived tree shown as ID: derived tree 1 in FIG.

  In the example shown in FIG. 9, in the ID: derived tree 1, the node “fw1” has three branches of CPU [Small], CPU [Medium], and CPU [Large]. It is narrowed down to CPU [Large].

  When the number of business patterns is two or more, the service configuration deriving unit 101 uses a plurality of business / resource relational expressions, and generates a pruning pattern from the plurality of business / resource relational expressions. As an example, as the pruning pattern, an average value of significance probabilities (p values) of explanatory variables of a plurality of business / resource relational expressions is used.

  For example, the significant probability (p-value) of the explanatory variable of video distribution (video traffic: patternA) is (cpu, RAM, NWIO, IOPS) = (0.001,-, 0.04, 0.03) and the application service (application traffic: patternB) If the significant probability (p-value) of the explanatory variable is (cpu, RAM, NWIO, IOPS) = (0.042,0.008,-, 0.04) (see FIG. 8), the service configuration deriving unit 101 calculates the average value. , (Cpu, RAM, NWIO, IOPS) = (0.021,0.008,0.04,0.03). In this case, the service configuration deriving unit 101 generates a service configuration for a plurality of tasks (ICT services) collectively.

  Then, the service configuration deriving unit 101 executes the process of step S36.

  The service configuration deriving unit 101 generates an instance candidate of the service configuration by combining the component candidates for each node (step S39).

  As described above, in the present embodiment, when selecting a candidate of a component of a system for providing a service based on log information collected when an ICT service or the like is operated, a configuration is obtained by pruning a derived tree. Since element candidates are narrowed down, constituent element candidates are derived in a short time.

  Note that the service configuration deriving unit 101 performs a test on the component candidates and selects the best service configuration. Then, the service configuration deriving unit 101 stores the selected service configuration in the service configuration storage device 203.

  FIG. 12 is an explanatory diagram illustrating an example of a service configuration stored in the service configuration storage device 203. As shown in FIG. 12, the service configuration storage device 203 stores data capable of specifying the way of cooperation of each service (each node) for realizing the ICT service and the amount of resources in each service.

  The example illustrated in FIG. 12 is an example that satisfies the service requirements illustrated in FIG.

  Further, the configuration changing device 202 reads a service configuration newly stored in the service configuration storage device 203, and builds a service chain on the service execution device 200 based on the service configuration.

  In the present embodiment, for example, a service configuration suitable for a new ICT service is designed, or a service configuration (service chain configuration) is automatically improved based on log information measured at the time of operation of the ICT service. It is used for purposes such as improving service quality.

  FIG. 13 is a block diagram showing a main part of the service configuration design device. The service configuration design apparatus 10 analyzes a traffic of each node for providing an ICT service and defines a business and defines a business (a business pattern definition unit) 11 (implemented by the business pattern definition unit 103 in the embodiment). ) And a task / resource relational expression generation unit (task / resource relational expression generating means) 12 that formulates the relation between the task and the resource consumption in each node to generate a task / resource relational expression. Is realized by the business / resource relational expression generating unit 102) and a service configuration deriving unit (service configuration deriving unit) 13 that generates a service configuration based on a design derivation tree representing a combination of resources that can be included in a node. (In the embodiment, the service configuration deriving unit 101 is realized.) On the basis of the source relationship to generate a service configuration after narrowing down the resources in the design derivation tree.

  The service configuration design program and the recording medium storing the service configuration design program may be configured as follows.

(1) The service configuration design program analyzes the traffic of each node for providing an ICT service to a computer and defines a business, and formulates the relationship between the business and the resource consumption of each node. And generating a business / resource relational expression based on the business / resource relational expression. And a process to be generated.

(2) The service configuration design program of (1) prunes a design derivation tree on the computer in accordance with the degree of the relation of the variable corresponding to the resource that can be included in the node in the business / resource relation to the business. To execute a process of narrowing down resources.

(3) The service configuration design program of (1) or (2) causes the computer to generate a business / resource relational expression by excluding resources having low correlation with the performance of the node.

(4) The service configuration design program according to any one of (1) to (3) causes the computer to collectively generate a service configuration for a plurality of tasks.

(5) A non-transitory storage medium storing a service configuration design program, wherein the service configuration design program, when executed by a processor, analyzes traffic of each node for providing an ICT service. To define a business, formulate a relationship between the business and the resource consumption in each of the nodes to generate a business-resource relational expression, and based on the business-resource relational expression, a resource that can be included in a node. After narrowing down resources in the design derivation tree representing the combination of the above, a service configuration is generated based on the design derivation tree.

(6) When the service configuration design program stored in the storage medium of (5) is executed by the processor, the degree of the relation of the variable corresponding to the resource included in the node in the business / resource relational expression to the business when the processor is executed The resources are narrowed down by pruning the design derivation tree according to.

(7) The service configuration design program stored in the storage medium of (5) or (6) excludes resources having low correlation with the performance of the node and generates a business / resource relational expression.

(8) The service configuration design program stored in any of the storage media according to (5) to (7) generates a service configuration collectively for a plurality of tasks.

  As described above, the present invention has been described with reference to the exemplary embodiments and examples. However, the present invention is not limited to the exemplary embodiments and examples. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

  This application claims priority based on Japanese Patent Application No. 2017-058724 filed on March 24, 2017, the entire disclosure of which is incorporated herein.

DESCRIPTION OF SYMBOLS 10 Service configuration design apparatus 11 Business pattern definition part 12 Business / resource relational expression generation part 13 Service configuration derivation part 100 Service configuration design equipment 101 Service configuration derivation part 102 Business / resource relational expression generation part 103 Business pattern definition part 110 Service requirement storage Unit 111 design derivation tree storage unit 112 business / resource relational expression storage unit 113 business pattern storage unit 200 service execution device 201 measuring device 202 configuration change device 203 service configuration storage device 204 measurement information storage device 300 input device 500 service providing system 510a, 510b Web server 520 Application server 530 Database server 540 Firewall 550 Load balancer 600 Internet

A service configuration design program according to the present invention analyzes a traffic of each node for providing an ICT service to a computer and defines a business, and formulates a relationship between the business and resource consumption of each node. And generating a business / resource relational expression based on the business / resource relational expression. After narrowing down resources in a design derivation tree representing a combination of resources that can be included in a node based on the business / resource relational expression, It causes and a process to be generated.

Claims (10)

Business pattern defining means for analyzing traffic of each node for providing an ICT service and defining a business;
A task / resource relational expression generating means for formulating a relation between the task and the resource consumption in each of the nodes to generate a task / resource relational expression,
Service configuration deriving means for generating a service configuration based on a design derivation tree representing a combination of resources that can be included in the node,
A service configuration design apparatus, wherein the service configuration deriving unit generates a service configuration after narrowing down resources in the design derivation tree based on the business / resource relational expression. The service according to claim 1, wherein the service configuration deriving unit narrows down the resources by pruning the design derivation tree according to the degree of the relation of the variable corresponding to the resource which can be included in the node in the business / resource relational expression to the business. Configuration design equipment. The service configuration design apparatus according to claim 1, wherein the business-resource relational expression generation unit generates a business-resource relational expression by excluding resources having low correlation with the performance of the node. The service configuration design device according to any one of claims 1 to 3, wherein the service configuration deriving unit generates a service configuration for a plurality of tasks collectively. Analyze the traffic of each node to provide ICT service and define the business,
Formulate a relationship between the task and resource consumption in each of the nodes to generate a task / resource relational expression,
A service configuration design method for generating a service configuration based on a design derivation tree after narrowing down resources in a design derivation tree representing a combination of resources that can be included in a node based on the business / resource relational expression. The service configuration design method according to claim 5, wherein the resource is narrowed down by pruning the design derivation tree according to the degree of the relation of the variable corresponding to the resource that can be included in the node in the business / resource relational expression to the business. The service configuration design method according to claim 5 or 6, wherein a business / resource relational expression is generated by excluding resources having low correlation with the performance of the node. The service configuration design method according to any one of claims 5 to 7, wherein a service configuration is generated collectively for a plurality of tasks. A computer-readable recording medium storing a service configuration design program,
The service configuration design program, when executed by a processor,
A process of analyzing traffic of each node for providing an ICT service and defining a business;
A process of formulating a relationship between the task and the resource consumption in each of the nodes to generate a task / resource relational expression;
A recording medium for executing a process of narrowing down resources in a design derivation tree representing a combination of resources that can be included in a node based on the business / resource relational expression, and then generating a service configuration based on the design derivation tree. The service configuration design program, when executed by a processor,
The recording medium according to claim 9, wherein a process for narrowing down resources is performed by pruning a design derivation tree in accordance with a degree of relation of a variable corresponding to a resource that can be included in a node in the business / resource relational expression to the business.

Images