WO2018109817A1 - Management system and management method - Google Patents

Abstract

This resource pool management system executes resource management processing and system construction processing. The resource management processing involves management as a resource group of a combination of two or more resources for achieving functionality. The system construction processing involves selection from a resource pool of one or more resources as constituent elements of a platform system, which is the system for achieving a required functionality, by taking into account the required functionality, the functionality of each resource included in the resource pool, and the functionality of each managed resource group.

Description

Management system and management method

The present invention relates generally to resource management.

Conventionally, as one of the operations of a computer system, computer resources such as servers and storages (hereinafter simply referred to as resources) are managed as a pool, and if a user needs it, it can be selected from resources belonging to the pool. There is an operation to use the resource of this resource and return the resource to the pool when the resource becomes unnecessary after that.

In such operations, the user manages the resource performance requirements (hereinafter referred to as performance requirements) such as the CPU (Central Processing Unit) operating frequency, memory mounting capacity and storage capacity. (Hereinafter referred to as the resource manager).

Then, the resource manager who has received such communication selects a resource satisfying the performance requirement from the pool, and allocates the resource to the user. Thus, the user uses the resource assigned by the resource manager.

In Patent Document 1 below, in a computer system in which resources to be pooled are not homogeneous, the performance information related to the performance of each resource and the function information related to the function of each resource are acquired, and the acquired performance information of each resource and Based on the function information and the performance requirement and function requirement requested in the resource request, a resource candidate to be allocated to the user is searched from a plurality of resources, and the resource to be allocated to the user from the candidates detected by the search A technique for determining a resource and allocating the determined resource to a user is disclosed.

US2011 / 0010634

A computer system including at least one of a server system (for example, one or more servers) and a storage system (for example, one or more storage devices) according to the needs of the company, and software that operates on the computer system It is also known to provide an integrated platform as an appliance composed of a plurality of entities such as operation management software operating in a management system of a computer system. A company that has introduced an integrated platform can start operation that meets the needs of the company if the integrated platform is powered on.

However, such appliances generally have restrictions such as hardware configuration and usage, and it is difficult to flexibly respond to the addition or change of the needs of the company beyond the pre-design of the appliance. For example, it is difficult for a company to freely use a free server just because there is a free server from the viewpoint of stability of the appliance or other aspects.

Therefore, it may be desired to set a virtual appliance on a plurality of resources including a plurality of types of resources.

As the functionality required for a virtual appliance, there may be functionality that cannot be realized only by the functionality of a single resource. However, Patent Document 1 does not disclose a technique for detecting which of the plurality of resources can provide the required functionality.

The resource pool management system executes resource management processing and system construction processing. The resource management process includes managing a combination of two or more resources that realize functionality as a resource group. The system construction process takes into account the functionality for each managed resource group in addition to the required functionality and functionality for each resource included in the resource pool. Including selecting one or more resources that are components of a base system that is a system that implements the required functionality.

∙ Manage a combination of two or more resources that realize functionality as a resource group. Accordingly, it is possible to select a resource that is a constituent element of the base system in consideration of functionality for each managed resource group.

It is a block diagram which shows the structure of the information system which concerns on embodiment. It is a block diagram which shows the structure of a server. Shows the structure of the resource table. The structure of a resource group table is shown. The structure of a resource group definition table is shown. The structure of a topology table is shown. The structure of a topology related table is shown. The structure of a base system template table is shown. The structure of a base system deployment table is shown. It is a block diagram which shows the structure of the information system to which the rack 3 was added. It is a part of flowchart which shows the flow of a detection process. It is the remainder of the flowchart which shows the flow of a detection process. An example of a resource table before detection processing is shown. The example of the resource table after a detection process is shown. The example of the topology table after a detection process is shown. The example of a resource group table after a detection process is shown. It is a flowchart which shows the flow of a deployment process.

In the following description, a table is used for information, but the data structure of the information is not limited, and other data structures may be used. Since the information does not depend on the data structure, for example, the “kkk table” can be called “kkk information”. Information such as a table may be divided into two or more information elements, or all or part of two or more types of information may be aggregated into the same information.

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

In the following description, the “storage unit” includes one or more memories. The 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 and may typically be a non-volatile storage device (eg, an auxiliary storage device). The PDEV may be, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

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

In the following description, the “management system” may be composed of 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. 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 among the management computer and the display computer may be a management system. The management system may include an interface unit, a storage unit, and a processor unit connected to them. 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 I / O device of one or more I / O devices (for example, an input device (for example, a keyboard and a pointing device), 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. That the computer in the management system “displays the display information” may be that the display information is displayed on the 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 processor unit executes a program and performs processing using at least one of a storage unit (for example, a memory) and an interface unit (for example, a communication port). In the following description, the subject of processing may be mainly a program, but a processor unit that executes the program may be the subject. In addition, it can be interpreted that the processing mainly performed by the processor unit is performed by executing one or more programs. The program may be installed in the computer from a program source. The program source may be, for example, a storage medium that can be read by a program distribution server or a computer. 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, when a description is made without distinguishing the same type of element, a reference code or a common code among the reference codes is used, and when a description is given by distinguishing the same type of element, the element is assigned to that element. May use an identified identifier (eg, at least one of a number and a sign) or an entire reference sign.

In the following description, the “computer system” may include at least one of a server system and a storage system. The “server system” may be one or more physical servers, and may include at least one virtual server (for example, VM (Virtual Machine)). The “storage system” may be one or more physical storage devices, and may include at least one virtual storage device (for example, SDS (Software Defined Storage)).

Further, in the following description, “resource” is an abbreviation for computer system resource (computer resource), and is an element that realizes functionality by itself or in combination with one or more other resources, typically, It corresponds to devices such as servers, storage devices (storage), switches, and networks. In other words, the “resource” does not have to correspond to a component (device component) such as a CPU and a memory.

In the following description, “functionality” is a function, service, or system realized by one or more resources.

In the following description, an “administrator” is a person who manages resources such as servers, storage, and networks. The administrator can perform at least one of maintenance work such as addition, deletion, and replacement of resources and management of a table group 30 on the management server 3 described later.

Also, in the following description, “user” is a person who requests deployment of a base system.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows a configuration of an information system according to the present embodiment.

1, the information system includes a resource pool 4 and a management system 1 that manages the resource pool 4. The management system 1 includes one or more management clients 2 and one or more management servers 3. These management client 2 and management server 3 are connected to each other via a management network 5. The resource pool 4 is also connected to the management network 5. The resource pool 4 is supplied with power from one or a plurality of power supply panels 6.

The management client 2 is a computer device used by a user, and includes, for example, a personal computer, a workstation, or a main frame. The management client 2 accesses the management server 3 via the management network 5 and executes various processes provided by the management server 3. The management client 2 transmits a infrastructure system deployment request to the management server 3. The management client 2 operated by the administrator transmits one or a plurality of infrastructure system deployment requests for each user. The “base system deployment request” is a request to deploy the base system. In the infrastructure system deployment request, for example, at least a deployment condition of the infrastructure system software ID and the deployment condition is specified. The “base system” is a system composed of a plurality of resources extracted from the resource pool 4, and the “base system software” is software (for example, virtual server software (for example, hypervisor)) installed in the base system. High-reliability database management system or cluster-type analysis software). The combination of the base system and the base system software may be a system (for example, a virtual appliance) such as a computer system that provides a user-desired service. The “deployment condition” is a condition related to a resource that is deployed (allocated) as a base system, and may include at least one of the number of servers, functionality, and resource sharing condition, for example. The designation of at least a part of the deployment condition may be a template ID and a version that will be described later. The designated template ID and version, and information described later (for example, information 306C to 306E) associated therewith may be at least a part of the deployment condition.

The management server 3 detects and manages each resource constituting the resource pool 4. In response to the infrastructure system deployment request from the management client 2, the management server 3 extracts a resource that matches the deployment request, and deploys (assigns) the infrastructure system that conforms to the deployment request to the extracted resource. The management server 3 is composed of, for example, a personal computer, a workstation or a main frame. The management server 3 includes a memory 11, a NIC (Network Interface Card) 12, an I / O interface 13, an input device 14, an output device 15, and a CPU 10 connected thereto.

The CPU 10 is an example of a processor unit. The CPU 10 is a device that controls operation of the entire management server 3. The CPU 10 executes various programs stored in the memory 11, thereby executing various control processes as the entire management server 3.

The memory 11 is an example of a storage unit. The memory 11 is used for holding various programs and control information, and is also used as a work memory for the CPU 10.

At least one of the NIC 12 and the I / O interface 13 is an example of an interface unit. The NIC 12 is a network interface for connecting the management server 3 to the management network 5. The I / O interface 13 is an interface for connecting the input device 14 and the output device 15 to the management server 3 and includes a USB (UniversalUniversSerial Bus) interface or the like.

The input device 14 includes a keyboard, a mouse, a microphone, and the like, and is used when the administrator inputs various operations to the management server 3. The output device 15 includes a monitor display and a speaker, and presents information such as images and sounds to the administrator 3.

The resource pool 4 has a plurality of racks 40. A plurality of resources including a plurality of types of resources can be mounted on the plurality of racks 40. As a result, the resource pool 4 includes a plurality of resources including a plurality of types of resources. One or more power supply panels 6 are provided for each rack 40. According to the example of FIG. 1, the rack 3 is not yet mounted, and there is a physical space in which the rack 3 can be mounted. When the rack 3 is added to the space, the power mounted on the rack 3 is supplied with power from the power supply panels 3 and 4.

The plurality of resources are, for example, standard Ethernet switch 41, high-performance Ethernet switch 42, FC (Fibre Channel) switch 43, standard server 44, high-performance server 45, FC storage 46, PDU (Power Distribution Unit) 47, and high performance. A core Ethernet switch 48 is included (Ethernet is a registered trademark). Each of the plurality of resources included in the resource pool 4 is connected to the management network 5 via an I / O interface (not shown) of the resource and communicates with the management server 3. Each rack 40 can be handled as a physical position of a resource mounted on the rack 40. For example, the physical location of the FC storage 1 can be “rack 1”. Further, for the resources connected to all the racks 40 like the high performance core Ethernet switch 1, the physical positions of the resources can be “common”. Each of the standard Ethernet switch 41, the high-performance Ethernet switch 42, the FC switch 43, and the high-performance core Ethernet switch 48 is an example of a network or a switch (for example, a network switch). Each of the standard server 44 and the high-performance server 45 is an example of a server. The FC storage 46 is an example of a storage (storage device).

As shown in FIG. 2, each of the standard server 44 and the high-performance server 45 includes a memory 61, an auxiliary storage device 62 (for example, HDD (Hard Disk Drive), SSD (Solid State Drive)), one or more I / Os. An interface 63, an expansion card 64 (for example, GPGPU (General-purpose computing-on graphics-processing-units)), and a CPU 60 connected thereto are provided. For example, at least one of the auxiliary storage device 62 and the expansion card 64 may not be provided. Further, the I / O interface 63 may be constituted by a NIC, an HBA (Host Bus Adaptor), or the like. The CPU 60 executes necessary control processing according to the program 65 stored in the memory 61.

As shown in FIG. 1, a resource management program 20 and a table group 30 are stored in the memory 11 of the management server 3.

The resource management program 20 includes a plurality of program modules such as a resource management unit 21, a topology management unit 22, and a infrastructure system management unit 23. Each of these program modules may be an independent program, or two or more of these program modules may be one program.

The table group 30 is one or more tables. The table group 30 includes a resource table 31, a resource group table 32, a resource group definition table 33, a topology table 34, a topology related table 33, a base system template table 36, and a base system deployment table 37.

The resource management unit 21 is a program for managing information related to resources existing in the resource pool 4. More specifically, for example, the resource management unit 21 manages information about resources existing in the resource pool 4 using the resource table 31. Further, the resource management unit 21 detects one or a plurality of resources having specific functionality that meets the conditions defined in the resource group definition table 33 as a resource group, and displays information about the detected resource group as a resource group. Managed by table 42. In addition, when a resource is added, updated, or deleted, the resource management unit 21 detects a change in the resource and updates the resource table 31.

The topology management unit 22 detects a physical connection relationship or logical connection relationship between resources by a known topology detection method, and manages the detected relationship in the topology table 34. Here, as the known topology detection method, for example, in Ethernet, there is an adjacency information exchange protocol called LLDP (Link Layer Discovery Protocol) which is an IEEE standard. By using information of adjacent resources detected by the LLDP from an Ethernet switch that supports this LLDP, a server, or an OS (Operating System) on the server, the adjacency information shown in the topology table 34 described later is created. Is possible. In SAN (Storage Area Network), authentication results called Fabric Login and Port Login are collected from, for example, FC switches, making it easy to connect between FC switches, between FC switches and servers, and between FC switches and storage. It is possible to grasp the connection relationship between them, and it is possible to create adjacency information shown in a topology table 34 described later.

When receiving the infrastructure system deployment request from the management client 2, the infrastructure system management unit 23 uses the infrastructure system template managed by the infrastructure system template table 36, and deploys from the resources managed by the resource management unit 21. A resource that matches a deployment condition that complies with the request is selected, and a base system that complies with the deployment request is deployed to the selected resource. The infrastructure system management unit 23 manages the information regarding the deployed infrastructure system and resources used in the infrastructure system in the infrastructure system deployment table 37.

Hereinafter, each table included in the table group 30 will be described.

FIG. 3 shows the configuration of the resource table 31.

The resource table 31 is a table for managing the functionality of each resource that constitutes the resource pool 4. The resource table 31 holds information such as resource ID 301A, resource type 301B, resource configuration 301C, functionality 301D, and physical location 301E for each resource. In the description of FIG. 3, one resource is taken as an example, and the resource is called “target resource”.

Resource ID 301A is an identifier for uniquely identifying the target resource. In the example illustrated in FIG. 3, for convenience, an identifier that easily identifies the type of the target resource is assigned as the resource ID.

The resource type 301B is information representing the type of the target resource. The resource configuration 301C is information regarding the component of the target resource. For example, when the target resource is a standard server or a high-performance server, the resource configuration 301C includes information indicating the type and performance of the CPU 60 of the target resource, information indicating the performance of the memory 61 of the target resource, and I / O of the target resource. Information representing the details of the interface may be included.

Functionality 301D is information representing one or more functionalities included in the target resource. The physical position 31E is information indicating the physical position of the target resource (for example, the ID of the rack 40 in which the resource is mounted). The physical position 301E may include other types of information instead of or in addition to the ID of the rack 40.

FIG. 4 shows the configuration of the resource group table 32.

The resource group table 32 is a table for managing information related to functionality composed of a plurality of resources. The resource group table 32 holds information such as resource group ID 302A, member resource 302B, and functionality 302C for each resource group. In the present embodiment, a “resource group” is not a mere collection of a plurality of resources but a collection of a plurality of resources that realize one or more functionalities. In the description of FIG. 4, one resource group is taken as an example, and the resource group is referred to as a “target resource group”.

The resource group ID 302A is an identifier for uniquely identifying the target resource group. The member resource 302B is a list of IDs of resources determined to belong to the target resource group. The functionality 302C represents one or more functionality realized by the target resource group (typically functionality shared by a plurality of member resources represented by the member resource 302B of the target resource group).

FIG. 5 shows the configuration of the resource group definition table 33.

The resource group definition table 33 is a table in which information representing a method for detecting a resource group from the resource pool 4 is stored. The resource group definition table 33 holds information such as the resource group 303A, the target functionality 303B, and the member resource discovery method 303C for each resource group type. In the description of FIG. 5, one resource group type is taken as an example, and the resource group type is referred to as “target resource group type”.

The resource group type 303A is an identifier (for example, a name) for uniquely identifying the target resource group type.

The target functionality 303B represents one or more functionalities realized by the resource group belonging to the target resource group type.

The member resource discovery method 303C represents a discovery method including at least one of procedures and conditions for discovering member resources of resource groups belonging to the target resource group type. For example, for at least one resource group type, the discovery method represented by the member resource discovery method 303C includes member resource conditions including the functionality (realization) of the resource that is the member resource and the connection configuration between the resources. .

FIG. 6 shows the configuration of the topology table 34.

The topology table 34 is a table that stores information representing the connection relationship between resources. The topology table 34 holds information such as a link ID 304A, a first link end ID 304B, a second link end ID 304C, and a network type 304D for each link. In the description of FIG. 6, one link is taken as an example, and the link is referred to as a “target link”.

The link ID 304A is an identifier that uniquely identifies the target link. “Link” means a connection between resources and no other resource intervening.

The first link end ID 304B is an identifier of the first link end that is one end of the target link. The first link end ID 304B includes at least a resource identifier of an identifier of a resource belonging to the first link end and an identifier of a resource component (for example, port) belonging to the first link end.

The second link end ID 304C is an identifier of the second link end that is the other end of the target link. The second link end ID 304C includes at least a resource identifier of an identifier of a resource belonging to the second link end and an identifier of a resource component (for example, port) belonging to the second link end.

The network type 304D represents the type of network including the target link (for example, “IP network” or “FC network”).

FIG. 7 shows the configuration of the topology related table 35.

The topology related table 35 is a table that stores information representing the relationship between the physical position and the functionality. The topology related table 35 holds information such as topology related ID 305A, topology position 305B, adjacent details 305C, physical position 305D, and functionality 305E for each topology related. In the description of FIG. 7, one topology relation is taken as an example, and the topology relation is referred to as “target topology relation”.

The topology related ID 305A is an identifier for uniquely identifying the target topology related. The topology position 305B is a position estimated from the adjacent details 305C corresponding to the target topology, and represents a position on the network topology. The adjacent detail 305C is information used for estimating a position on the network topology, and is information including an identifier of one or more resources (or resource components) adjacent to the position. The physical position 305D represents a physical position associated with a position (position on the network topology) estimated from the adjacent details 305C corresponding to the target topology relation, that is, an estimated physical position (for example, the ID of the rack 40). The functionality 305E represents the functionality (for example, the power supply system assigned to the rack represented by the physical location 305D) associated with the estimated physical location corresponding to the target topology association.

The significance of the topology related table 35 is, for example, as follows. That is, for the resource pool 4 (for example, the data center), the power supply panel 6 and the switch port corresponding to the physical position are designed in advance for each physical position. Therefore, when a resource (for example, the rack 40 on which the resource is mounted) is added to the physical location, the resource management program 20 determines the network (topology) to which the resource is connected based on the ID of the port to which the resource is connected. The position (305B) can be specified, the physical position (physical position 305D) corresponding to the network is estimated from the table 34, and the functionality (functionality 305E) associated with the estimated physical position is specified from the table 34. The functionality represented by the functionality 305E is different from the functionality represented by the information that can be acquired from the resource (physical hardware) (that is, the functionality possessed by the resource). Functionality associated with a place called rack 40). That is, in the present embodiment, the realizable functionality may differ depending on the physical position where the resource is arranged. Therefore, by referring to the topology-related table 35 described above, it is possible to determine whether or not the functionality can be realized in consideration of the physical position where the resource is arranged. In other words, the physical location can also be handled as a kind of resource, and thus the physical location can also be handled as a member resource of the resource group.

FIG. 8 shows the configuration of the base system template table 36.

The infrastructure system template table 36 is a table in which information related to infrastructure system templates is stored. The infrastructure system template table 36 includes, for each infrastructure system template (hereinafter referred to as infrastructure template), a template ID 306A, a version 306B, a main resource 306C, a related resource 306D, a resource constraint 306E, an unapproved environment verification 306F, a verification record 306G, and an automatic deployment program Information such as 306H is held. In the description of FIG. 8, one base template is taken as an example, and the base template is referred to as “target base template”.

Template ID 306A is an identifier (for example, name) for uniquely identifying the target base template. The version 306B represents the version of the target base template.

For each resource as a main resource belonging to the target base template, the main resource 306C includes information indicating the type of the resource (for example, a standard server or a high-performance server) and information indicating the version of firmware to be applied to the resource. Including. Here, the “main resource” is typically a server on which the base system software is installed, but is not limited thereto.

The related resource 306D includes, for each resource as a related resource belonging to the target base template, information indicating the type of the resource (for example, FC switch), information indicating the version of firmware to be applied to the resource, and the firmware version And an application flag that is a flag indicating whether application is mandatory. Here, the “related resource” is a resource that is inevitably used depending on the location and connection relationship of the main resource, for example, an Ethernet switch or an FC switch connected to the main resource via a network cable. It is not limited to. The application flag includes a value “strict” indicating that application is required and a value “non-strict” indicating that application is not required.

The resource constraint 306E represents a selection condition (restriction) related to the main resource and related resources belonging to the target base template. The selection condition represented by the resource constraint 306E includes, for example, functionality associated with the target base template (that is, functionality required for the base system corresponding to the template).

The non-approved environment verification 306F is information (for example, the program or its ID) regarding the non-approved environment verification program associated with the target base template. The non-approved environment verification program is a program executed to check whether the current firmware version satisfies the requirements of the target base template in the following cases (a) and (b).
(A) The resource represented by the related resource 306D is found, but the firmware version of the found resource does not match the firmware version represented by the related resource 306D.
(B) The value indicated by the suitability flag represented by the related resource 306D is “non-strict”.

The verification result 306G is information indicating the result of executing the non-approved environment verification program represented by the non-approved environment verification 306F corresponding to the target base template in the past (for example, information indicating the version of the firmware that has passed the verification, and Log data (or a pointer to log data) at the time of execution of the verification program.

The automatic deployment program 306H is information (for example, a program or an ID) related to a deployment program that executes installation of infrastructure system software, application of related firmware, and application of various settings.

FIG. 9 shows the configuration of the base system deployment table 37.

The infrastructure system deployment table 37 is a table in which information related to the deployed infrastructure system is stored. The infrastructure system deployment table 37 holds information such as a deployment ID 307A, a template ID 307B, a deployment version 307C, a deployment main resource 307D, and a deployment related resource 307E for each infrastructure system. In the description of FIG. 9, one base system is taken as an example, and the base system is referred to as a “target base system”.

The deployment ID 307A is an identifier for uniquely identifying the deployed target infrastructure system. The template ID 307B is an identifier of the base template applied to the deployment of the target base system. The deployment version 307C represents the version of the applied base template.

The deployed main resource 307D includes, for each main resource included in the deployed base system, an identifier of the main resource and information indicating the firmware version applied to the main resource.

The deployment related resource 307E includes, for each related resource included in the deployed base system, an identifier of the related resource and information indicating the firmware version applied to the related resource.

Next, the resource group management method according to this embodiment will be described.

Information regarding resources such as servers, storage, and networks is held in the resource table 31. When a resource is added to the resource pool 4, the resource management program 20 detects the added resource (for example, receives information related to the added resource from the resource pool 4), and stores information 301A to 301E related to the detected resource in the resource table 31. Add to. Further, the resource management program 20 uses a combination of one or more resources including additional resources based on the information 301A to 301E regarding the additional resources, the target functionality 303B for each resource group type, and the member resource discovery method 303C. Detects the presence or absence of resource groups that have the functionality that is realized. When the resource management program 20 detects such a resource group, the resource management program 20 stores information 302 A to 302 C regarding the detected resource group in the resource group table 32. Further, when the resource management program 20 receives a platform system deployment request for a certain user from the management client 2, the resource management program 20 selects and deploys a resource that matches the deployment conditions according to the deployment request.

Hereinafter, the detection process according to the present embodiment will be described. The state and configuration before execution of the detection process are the same as those shown in FIGS. 1 and 3 to 9. Here, it is assumed that the resource group definition table 33 (FIG. 5), the topology related table 35 (FIG. 7), and the infrastructure system template table 36 (FIG. 8) are created in advance by, for example, an administrator. The resource table 31 (FIG. 3), the resource group table 32 (FIG. 4), the topology table 34 (FIG. 6), and the infrastructure system deployment table 37 (FIG. 9) may be generated by the resource group management method described above. For example, it may be stored in advance by an administrator.

In the state and configuration shown in FIG. 1 and FIGS. 3 to 9, it is assumed that the following is performed as shown in FIG.
・ Add rack 3.
-Add various resources in rack 3.
-Connection of power supply system (specifically, A-system PDU3 is connected to power supply panel 3 and B-system PDU3 is connected to power supply panel 4)
Connect any port of the high-performance Ethernet switch 2 in the rack 3 to the port 3 (not shown) of the high-performance core Ethernet switch 1 and the port 3 (not shown) of the high-performance core Ethernet switch 2 using a network cable. To do.

Note that the resources in the rack 3 are connected to the management network 5 and can communicate with the management server 3.

Furthermore, it is assumed that a rack management program 49 that provides a management function for the rack 3 is executed on the memory (not shown) of the high-performance Ethernet switch 2 in the rack 3. In this embodiment, the rack management program 49 operates on the high-performance Ethernet switch 2, but the rack management program 49 is not limited to this, and may operate on the memory of the high-performance server 3, for example. The rack management program 49 may be incorporated in the rack 3 at the time of factory shipment, for example, and may be activated when the rack 3 is powered on. The rack management program 49 may exist for each rack 40.

In addition, for all resources in the management server 3 and the rack 3, settings necessary for management communication such as an IP (Internet Protocol) address are set in the I / O interface used for management communication.

11 and 12 show the flow of detection processing. In the following description, it is assumed that the rack 3 is added as illustrated in FIG. The detection process is an example of a resource management process that is a process related to resource management. A resource may be a member resource of a resource group or a single resource that does not belong to a resource group.

When the rack management program 49 detects that the network cable is connected to the rack 3 (detects that the rack 3 is connected to the network), it notifies the resource management program 20 of the addition of the rack 3 (S1). .

Next, when the resource management program 20 receives a notification of expansion of the rack 3 (S2), it requests the rack management program 49 for information on the resources in the rack 3 (S3).

In response to the request, the rack management program 49 transmits information on the resources in the rack 3 to the resource management program 20 (S4). It is assumed that the information transmitted in S4 has the same content as the information described in FIG.

The resource management program 20 receives information related to the resource 3 in the rack 3 from the rack management program 49, and at least one of information according to the information (for example, received information and information given based on the information) ) 301A to 301E are added to the resource table 31, and the rack management program 49 is requested for information related to adjacent details (S5). Here, FIG. 13A shows a state in which information 301A to 301D regarding the resources in the rack 3 is added to the resource table 31 in S5. According to FIG. 13A, the physical location 301E is not stored, and the functionality 301D regarding the resources “A-system PDU3” and “B-system BDU3” is also not stored. This is because the information transmitted from the rack management program 49a in S4 (information on the resources in the rack 3) is information registered in the rack management program 49 at the time of shipment of the rack 3 (or resources). This is because at the time of shipment, the installation location of the rack 3 and the power supply system to which the PDU is connected are not known, and therefore, the rack management program 49 at the time of shipment does not hold information on the physical position and the power supply system.

Next, the rack management program 49 detects a link between the inside and outside of the rack 3 using a known method such as the above-described LLDP, and sends link information, which is information about the detected link, to the resource management program 20. Transmit (S6).

The resource management program 20 receives the link information, and stores information 304A to 304D related to the link represented by the received information in the topology table 34. An example of information stored in the topology table 34 is shown in FIG. Further, the resource management program 20 compares the adjacency details 305C of the topology related table 35 (FIG. 7) with the information 304A to 304D (information regarding the link related to the rack 3) stored in the topology table 34, and searches for a matching record. (S7).

If a matching record is found (S8: Y), the process proceeds to S10. If no matching record is found (S8: N), the administrator is notified that there is an unexpected connection using means such as email, and the process ends (S9).

In S10, the resource management program 20 uses the information stored in the record in the resource table 31 (added in S5) using the information stored in the record found in S7 (the record of the topology related table 35). , Information on records in the rack 3) is updated. More specifically, for example, it is assumed that a record including the topology-related ID 305A “RL3” is found in the topology-related table 35 in S7. This is because the adjacent details 305C in the record are adapted to the information 301A to 301D regarding the resources in the rack 3. The resource management program 20 copies the physical position 305D (here, “rack 3”) in the found record to the record in the resource table 31 (record corresponding to the resource in the rack 3) as the physical position 301E. Further, the resource management program 20 assigns the A system “power supply system 3” and the B system “power supply system 4” to the resource ID 301A “A system PDU3” from the functionality 305E in the record including the topology-related ID 305A “RL3”. "And a record including the resource ID 301A" B-system PDU3 "are stored as functionality 301D (FIG. 13B).

Next, the resource management program 20 sequentially extracts records from the resource group definition table 33 (S11), and among the resources in the rack 3, the same functionality 301D as the target functionality 303B in the extracted records. Are searched from the resource table 31 (FIGS. 3 and 13B) (S12).

When a resource corresponding to the same functionality 301D is found (S13: Y), the process proceeds to S14. When such a resource is not found (S13: N), the process proceeds to S18.

In S14, the resource management program 20 searches for member resources other than the found resource by the discovery method represented by the member resource discovery method 303C in the record extracted in S11 (S14).

Here, a specific search example will be described by taking a record (see FIG. 5) including the resource group ID 303A “FC SAN fabric” as an example. The record includes the target functionality 303B “FC SAN” and the member resource discovery method 303C “a resource group having the target functionality and mutual FC network adjacency”. Therefore, in S12, the resource management program 20 selects the resource “high performance server 3”, “high performance” corresponding to the functionality 301D “FC SAN” from the table 31 (resources in the rack 3) shown in FIG. 13B. Server 4 "," FC storage 3 "and" FC switch 3 "are specified. Next, the resource management program 20 searches for adjacent resources in the FC network using the topology table 34 for each of these resources. For example, regarding “high performance server 3”, “FC switch 3” is detected as an adjacent resource. Furthermore, “high-performance server 4” and “FC storage 3” are detected as adjacent resources for “FC switch 3”. As described above, “high performance server 3”, “high performance server 4”, “FC storage 3”, and “FC switch 3” are the resource groups having the target functionality described in the member resource discovery method 303C. 7 is a list of member resources of a resource group discovered by “there is an FC network adjacent relationship”.

Next, the resource management program 20 determines whether or not there is a record in the resource group table 32 having the member resource 302B including all the member resources found in S14. If the determination result is true (S15: Y), the process proceeds to S17. If the determination result is false (S15: N), the process proceeds to S16.

In S16, the resource management program 20 adds a new record to the resource group table 32. The resource management program 20 stores a uniquely identifiable identifier as the resource group ID 302A in the new record, stores the IDs of all the member resources found in S14 as the member resource 302B, and stores them as the functionality 302C. Then, the value of the resource group type 303A in the record used for the search (the record extracted in S11 from the resource group definition table 33) is stored.

In S17, the resource management program 20 adds the ID of the member resource found in S14 to the member resource 302B in the record found in S15 (S15: Y).

In S18, the resource management program 20 determines whether or not execution has been performed for all resources searched in S12. If the determination result is false (S18: N), the process returns to S14, and the same processing is executed for the remaining resources. On the other hand, if the determination result is true (S18: Y), the process proceeds to S19.

In S19, the resource management program 20 determines whether the processes of S11 to S18 have been executed for all resource group definitions. If the determination result is true (S19: Y), the processing flow ends. If the determination result is false (S19: N), S11 is executed for the next resource group definition.

FIG. 15 shows an example of the updated resource group table 32 as a result of executing S1 to S19 for expansion of the rack 3. In the data of the resource group table 32 shown in FIG. 15, the underline indicates the resource group table 32 before the detection process is executed as a result of the detection process by the resource group management method according to the present embodiment (see FIG. 15). Data updated from 4). As shown in FIG. 15, a high performance server and a high performance server 4 are added to the member resource 302B in the “GPGPU cluster 1” resource group, and a high performance Ethernet is added to the “IP QoS transport network 1” resource group. Switch 2, high-performance server 3, and high-performance server 4 are added to member resource 302B, and “FC SAN fabric 2”, “power system 3”, and “power system 4” are newly added to resource group table 32. ing.

FIG. 16 shows a flow of deployment processing according to the present embodiment. The deployment process is an example of a system construction process that is a process related to the construction of the base system.

The resource management program 20 receives the infrastructure system deployment request from the management client 2 (S31). It is assumed that at least the template ID and version and the number of servers are specified in this infrastructure system deployment request.

In the description of the deployment process, it is assumed that the state and configuration after the detection process is performed. That is, it is assumed that the configurations and states shown in FIGS. 5, 7, 8, 9, 10, 13B, 14 and 15 are employed. Further, it is assumed that the template ID “high-performance analysis server system”, version “001”, and the number of servers “2” are specified as deployment conditions in the infrastructure system deployment request received in S31.

Next, the resource management program 20 searches the base system template table 36 for a record that matches the template ID and version specified in the received deployment request (S32). Here, it is assumed that a record including the template ID 306A “high performance analysis server system” and the version 306B “001” (in other words, a base template corresponding to the record) is found.

Next, in S33, the resource management program 20 uses the selectable main resource type described in the main resource 306C in the found record as a search key, and the resource type 301B matches the type from the resource table 31. Search for resources. In S33, the resource management program 20 extracts a resource that is not included in the deployment main resource 307D of the infrastructure system deployment table 37, that is, a resource that is not used, among the resources found by the search.

In this embodiment, since the main resource 306C of the found base template is “high performance server (firmware: 00-02)”, the high performance server 1, the high performance server 2, and the high performance server 3 are determined from the resource table 31. And the high performance server 4 is extracted. Furthermore, since the high-performance server 1 has already been used from the infrastructure system deployment table 37 shown in FIG. 9, the resource management program 20 obtains a high-performance server other than the high-performance server 1, that is, a high The performance server 2, the high performance server 3, and the high performance server 4 are extracted as candidates.

Next, when a candidate unused resource is found as a result of S33 (S34: Y), the process proceeds to S35. If not found (S34: N), the process proceeds to S37.

In S37, the resource management program 20 notifies the user corresponding to the deployment request received in S31 that the deployment has failed.

In S35, the resource management program 20 includes whether the found resource satisfies the resource constraint 306E corresponding to the found base template, the functionality 301D corresponding to the found resource, and the found resource. This is determined using the functionality 302C corresponding to the resource group. Specifically, for example, the resource management program 20 determines whether or not functionality 301D corresponding to the found resource has functionality that satisfies the resource constraint 306E. If the determination result is false, the resource management program 20 determines whether or not the functionality 302C corresponding to the resource group including the found resource has functionality that satisfies the resource constraint 306E.

In the present embodiment, since the resource constraint 306E corresponding to the found base template is “GPGPU cluster”, only “GPGPU cluster 1” in the resource group table 32 shown in FIG. At least one of them is included and the resource constraint 306E is satisfied.

In S38, the resource management program 20 creates all combinations of resources for the number of servers specified in the deployment request received in S31 among the resource candidates narrowed down using the resource constraint 306E in S35. Furthermore, when all the resources belonging to the resource combination are included in the resource group determined to satisfy the resource constraint 306E in S35, the resource management program 20 selects the resource combination as a final candidate. Specifically, for example, when there are two or more resource groups belonging to the same resource group type “FC SAN fabric”, the resource management program 20 does not select a resource from two or more resource groups, Select a resource from a resource group (same resource group).

If no final candidate is found in S38 (S39: N), the process proceeds to S37, and if found (S39: Y), the process proceeds to S40.

In S40, the resource management program 20 extracts the resource type and firmware version included in the related resource 306D corresponding to the found base template. The firmware version acquired in S40 is hereinafter referred to as “requested firmware version”.

In S41, the resource management program 20 searches the topology table 34 for a resource adjacent to the final candidate resource (a resource connected to the final candidate resource via a link), and Check the firmware version of the existing resource. The resource management program 20 may acquire the firmware version of the adjacent resource by accessing the resource. The resource information collected for creating the topology table 34 may include the firmware version of the resource, and the firmware version may be registered in the topology table 34 (or another table). It may be acquired from the table 34. When there is a resource that matches the resource type extracted in S40 as an adjacent resource, the resource management program 20 further determines whether the firmware version of the adjacent resource matches the requested firmware version. If the determination result is true (S41: Y), the process proceeds to S47. If the determination result is false (S41: N), the process proceeds to S42.

In S42, the resource management program 20 checks whether the requested firmware version is “strict” or “non-strict”. If the requested firmware version is “strict”, the process proceeds to S37. When the requested firmware version is “non-strict”, the resource management program 20 matches the firmware version of the adjacent resource with the verification result 306G of the base template (for example, information indicating the firmware version already verified in the past). It is determined whether or not to perform (S43). If the determination result is true (S43: Y), the process proceeds to S47. If the determination result is false (S43: N), the process proceeds to S44.

In S44, the resource management program 20 applies the verification program represented by the unapproved environment verification 306F corresponding to the base template to the adjacent resource (adjacent resource (related resource) having firmware whose version does not match the requested firmware version). And execute. When the execution result of the verification program is NG (S45: N), the process proceeds to S37. If the execution result of the verification program is OK (S45: Y), the process proceeds to S46.

In S46, the resource management program 20 corresponds to the base template information including information indicating the current firmware version of the adjacent resource (related resource) and information indicating the adjacent resource (for example, resource ID). It adds to verification performance 306G. As a result, the current firmware version of the adjacent resource (related resource) is registered as the verified firmware version. The verification result 306G may also include information regarding a log representing the execution result.

In S47, the resource management program 20 executes the deployment program represented by the automatic deployment program 306H corresponding to the infrastructure template, thereby deploying the infrastructure system software to the final candidate resource.

According to the deployment process shown in FIG.

The resource management program 20 searches for a base template that matches the deployment request. The resource management program 20 executes the following processes (A) to (D) for the found base template (hereinafter referred to as the corresponding template).
(A) Narrow down resources that satisfy resource constraints 306E (for example, functionality) corresponding to the template and candidate resources that are resource groups;
(B) Search for a free server as an installation destination (deployment destination) of the base system software from the narrowed candidate resources (the free server is a server corresponding to the resource type represented by the main resource 306C corresponding to the template) Unused server),
(C) Search for resources that are adjacent to the selected candidate resource and that belong to the resource type represented by the related resource 306D corresponding to the template.
(D) It is determined whether or not the firmware version of the resource found in (C) matches the requested firmware version_ (the firmware version represented by the related resource 306D corresponding to the template).
(E) If the determination result of (D) is false, determine whether the requested firmware version is “non-strict”;
(F) When the determination result of (E) is true, the verification program represented by the non-approved environment verification 306F corresponding to the template is assigned to the adjacent resource (related resource (related resource) having a firmware version that does not match the requested firmware version). )).
(G) When the execution result of (F) is OK, information including information indicating the current firmware version of the adjacent resource is added to the verification result 306G corresponding to the template.

Processes (D) to (G) may be performed on the main resource instead of or in addition to the adjacent resource (related resource). The main resource 306C may also include a firmware version (requested firmware version including “strict” or “non-strict”) for each server as the main resource.

The significance of the processes (E) to (G) among the above processes is, for example, as follows. That is, for resource types corresponding to “strict”, use of firmware versions other than the firmware version guaranteed by verification on the template providing side is prohibited. However, it is conceivable that flexible resource use on the user side is desired (for example, it is desired to use a specific resource without requesting permission from the provider of the corresponding template). Alternatively, a common resource (typically a switch such as a network switch) that can be any element of a plurality of infrastructure systems including the first and second infrastructure systems is conceivable and common. The firmware version of the resource may conform to the firmware version corresponding to the template of the first base system, but may not match the firmware version corresponding to the template of the second base system. Therefore, “non-strict” can be adopted for a certain firmware version of a certain resource. “Non-strict” means that adoption of a firmware version different from the requested firmware version may be permitted. Specifically, when the execution result of the verification program corresponding to the template is OK (verification is successful) (in other words, when the verification result is OK on the user side instead of the template providing side) The adoption of the firmware is permitted, and the verification result 306G such as the adopted firmware version is registered in the infrastructure system deployment table 37.

When the determination result of the process (E) is true, the resource management program 20 determines whether or not the verification result 306G corresponding to the template is verification OK (indicating a result that the execution result of the verification program was OK). You may judge. When the determination result is false, the resource management program 20 may execute the process (F).

Also, flags such as “non-strict” and “strict” are not necessarily required (for example, firmware versions may be “non-strict” for all resources).

The above is the description of this embodiment.

According to the present embodiment, in addition to individual resources, a concept of “resource group” that is a combination of a plurality of resources providing functionality is introduced. For each resource group type, a discovery method 303C is defined as a condition for providing functionality associated with the resource group. By combining the resources found according to the discovery method 303C, the functionality represented by the corresponding target functionality 303B can be provided. For example, as a communication technology for realizing high-quality network communication, there is PFC (Priority-based Flow Control) defined as 802.1Qbb specification of IEEE (The Institute of Electrical and Electronics Engineers, Inc.). According to the embodiment described above, it is possible to construct a resource group that is configured by a plurality of resources including inter-server resources that support PFC as functionality and that can realize PFC as functionality.

Further, according to the present embodiment, the topology is referred to find the resource along the discovery method 303C. Note that referring to the topology is only one method for discovering member resources, and depending on the definition of the discovery method 303C, reference to the topology may be unnecessary.

Further, according to the present embodiment, the functionality 305E is associated with the physical position estimated from the network position (position on the topology) detected based on the topology. When the topology is referenced according to the discovery method 303C, the functionality 305E associated with the topology can be associated as the resource group functionality 302C.

Further, according to the present embodiment, the firmware version corresponding to the template of the first base system for the common resource that can be any element of the plurality of base systems including the first and second base systems. However, it is possible that the firmware version corresponding to the template of the second base system is not compatible. Even in such a case, if the requested firmware version corresponding to the common resource is “non-strict” for the template of the second infrastructure system (and the execution result of the verification program corresponding to the template is OK). The common resources can be deployed as elements of the second infrastructure system as well as elements of the first infrastructure system.

As mentioned above, although one Embodiment of this invention was described, it cannot be overemphasized that this invention is not limited to this Embodiment, and can be variously changed in the range which does not deviate from the summary.

1: Management system

Claims (14)

1. An interface unit including one or more interfaces connected to the resource pool;
   A processor unit including one or more processors connected to the interface unit,
   The processor unit executes a resource management process and a system construction process,
   The resource management process includes managing a combination of two or more resources realizing functionality as a resource group,
   The system construction process takes into account the functionality for each managed resource group in addition to the required functionality and the functionality for each resource included in the resource pool. Selecting one or more resources that are components of a base system, which is a system that implements the required functionality,
   Management system.
2. The resource management process includes:
   Detecting that one or more resources have been added to the resource pool;
   In response to the detection, for each of two or more resource group types, according to the resource discovery method defined for the resource group type, the resource as a member of the resource group belonging to the resource group type is added to the added 1 Including discovering from these resources,
   The management system according to claim 1.
3. The resource discovery method for at least one resource group type includes member resource conditions including functionality realized by a resource that is a member of a resource group belonging to the resource group type, and a connection configuration between resources.
   The management system according to claim 2.
4. In the resource management process, the functionality of at least one of the added one or more resources is functionality associated with the physical location of the resource.
   The management system according to claim 2.
5. The physical location of the at least one resource is a physical location estimated from a network location to which the at least one resource is connected.
   The management system according to claim 1.
6. In the system construction process, at least one of the selected one or more resources is the functionality of the resource among the free resources that are not included in any base system. , At least one of the functionality of the resource group containing the resource is a resource that matches the required functionality,
   The management system according to claim 1.
7. In the system construction process, the one or more selected resources are all included in any one resource group of two or more resource groups belonging to the same resource group type.
   The management system according to claim 1.
8. In the system construction process, the firmware version of a specific resource among the selected one or two or more resources does not match the required firmware version, but a verification success was obtained as an execution result of the firmware verification program Firmware version,
   The management system according to claim 1.
9. The system construction process is as follows:
   Selecting a base system template from a plurality of base system templates each associated with the required functionality, the required resource type and the required firmware version;
   Determining whether the firmware version of the specific resource belonging to the resource type associated with the infrastructure system template matches the requested firmware version associated with the selected infrastructure system template; When,
   If the determination result is negative, executing the firmware verification program for the specific resource;
   including,
   The management system according to claim 8.
10. The system construction process further includes
    When verification success is obtained as an execution result of the firmware verification program, information including information indicating the firmware version of the specific resource is associated with the selected base system template as a verification record.
    The management system according to claim 8.
11. The firmware verification program is executed when associated with a value that means that it is not strictly required to match the requested firmware version;
    The management system according to claim 8.
12. The resource pool includes a plurality of resources including a plurality of types of resources,
    The plurality of resources includes a plurality of servers and one or more switches,
    One or more servers of the plurality of servers are connected to each of the one or more switches,
    The specific resource is at least one of the one or more switches.
    The management system according to claim 8.
13. Resource pools,
    A management system connected to the resource pool,
    The management system executes resource management processing and system construction processing,
    The resource management process includes managing a combination of two or more resources realizing functionality as a resource group,
    The system construction process takes into account the functionality for each managed resource group in addition to the required functionality and the functionality for each resource included in the resource pool. Selecting one or more resources that are components of a base system, which is a system that implements the required functionality,
    Information system.
14. Execute resource management processing,
    Execute system construction process
    The resource management process includes managing a combination of two or more resources realizing functionality as a resource group,
    The system construction process takes into account the functionality for each managed resource group in addition to the functionality required for each resource included in the resource pool, and from the resource pool, Selecting one or more resources that are components of a base system, which is a system that implements the required functionality,
    Management method.

Images